Environment real-time monitoring method based on intelligent Internet of Things

Abstract

The invention provides an environment real-time monitoring method based on an intelligent internet of things. The system is mainly characterized in that a sensing layer collects environmental data in real time by using a plurality of sensor nodes of the sensing layer, and the environmental data cover a plurality of parameters such as temperature, humidity, air quality and illumination intensity; the network layer performs data transmission through an NB-IoT technology in a low-power wide area network protocol, so that wide coverage, high reliability and low power consumption are realized; the application layer uses a variational auto-encoder to optimize a time sequence analysis algorithm, and carries out deep analysis and processing on environmental data, thereby realizing accurate prediction and timely monitoring on an environmental state; the user interface dynamically displays environment data through front-end development and data visualization technology, and friendly user experience is provided. A low-power-consumption transmission mode and a secure transmission protocol ensure energy conservation and data security of the system. In conclusion, the method is superior to a traditional method in the aspects of real-time monitoring, data transmission efficiency, intelligent analysis, user experience, energy conservation, data security and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and in particular to a real-time environmental monitoring method for an intelligent Internet of Things. Background Art

[0002] With the rapid development of society and people's increasing requirements for environmental quality, the importance of real-time environmental monitoring has become increasingly prominent. Traditional environmental monitoring methods mainly rely on manual sampling and laboratory analysis, which has many limitations. First, the monitoring cycle is long, resulting in data lag and failure to reflect real-time changes in the environment in a timely manner. Secondly, the coverage is limited, making it difficult to fully obtain environmental information in a large area. In addition, the manual sampling and laboratory analysis process is cumbersome, costly, and easily interfered by human factors, making it difficult to guarantee the accuracy and reliability of the data.

[0003] In terms of data transmission and processing, traditional environmental monitoring systems also face a series of problems. The data transmission efficiency is low, and the collected data is often not transmitted to the analysis center in time, which affects the timeliness of monitoring. At the same time, the energy consumption is high, which increases the monitoring cost. The data analysis capability is limited, and it is difficult to conduct in-depth mining and analysis of large amounts of complex data, and it is impossible to accurately predict and monitor the environmental status in time.

[0004] In order to overcome these shortcomings and meet people's needs for real-time environmental monitoring, the present invention proposes a real-time environmental monitoring method based on intelligent Internet of Things, which aims to collect environmental data in real time through multiple sensor nodes in the perception layer, and efficiently transmit the data to the application layer with the help of the wireless communication protocol of the network layer. The application layer uses intelligent algorithms to accurately analyze and process the data, thereby realizing accurate prediction and timely monitoring of the environmental status. Summary of the invention

[0005] In order to achieve the above-mentioned purpose of accurately predicting and timely monitoring the environmental status, the present invention adopts a real-time environmental monitoring method based on the intelligent Internet of Things, including the following parts:

[0006] S1, the perception layer collects environmental data through multiple sensor nodes, each of which includes one or more sensors for detecting different parameters in the environment;

[0007] S2, the network layer transmits the environmental data collected by the perception layer to the application layer through a wireless communication protocol, and the wireless communication protocol uses a low-power wide area network protocol for data transmission;

[0008] S3, the application layer analyzes and processes the received environmental data, and predicts and monitors the environmental status through an intelligent algorithm. The intelligent algorithm uses a variational autoencoder to optimize the time series analysis algorithm to monitor changes in environmental quality over a period of time in the future;

[0009] S4, the user interface module on the application layer realizes environmental data display through front-end development, data visualization, front-end and back-end communication, back-end support, deployment and optimization;

[0010] The step S3 uses a variational autoencoder to optimize the time series analysis algorithm to monitor changes in environmental quality over a period of time in the future. The implementation steps are:

[0011] S31. First, define the encoder to map the time series data to the latent variable space. The input layer inputs the time series data x. The hidden layer is one or more fully connected layers for extracting features. The output layer has two outputs, which are the mean μ and standard deviation σ of the latent variable. Define the sampling layer to sample the latent variable z from the latent variable distribution. The sampling formula is:

[0012] z=μ+σ·ε

[0013] Where ε is the random noise of the standard normal distribution; the decoder is defined to reconstruct the hidden variable z into time series data. The input layer is the hidden variable z, the hidden layer is one or more fully connected layers for extracting features, and the output layer is the reconstructed time series data

[0014] S32, then define the loss function, reconstruction loss to measure the reconstructed time series data The difference between the original data x is:

[0015]

[0016] Use KL divergence loss to measure the difference between the latent variable distribution q(zx) and the prior distribution p(z):

[0017]

[0018] The total loss is the sum of the reconstruction loss and the KL divergence loss:

[0019]

[0020] Among them, Γ is the total loss, β is the hyperparameter that weighs the reconstruction loss and KL divergence loss;

[0021] S33, then compile the model, select Adam optimizer to compile the model, the optimizer formula is:

[0022]

[0023] Among them, θ is the model parameter and η is the learning rate;

[0024] Use training data to train the model and update model parameters. The training formula is:

[0025]

[0026] Among them, θ * are the parameters updated after model training;

[0027] S34. Finally, use the validation data to evaluate the model performance, adjust the model hyperparameters, input new time series data, and use the trained variational autoencoder model for prediction.

[0028] Preferably, the sensors described in step S1 include temperature sensors, humidity sensors, air quality sensors, light sensors, etc., and the collected data include environmental parameters such as temperature, humidity, air pollutant concentration, light intensity, etc.

[0029] Preferably, the network layer of step S2 adopts a low power wide area network protocol for data transmission, and the steps are as follows:

[0030] S21. First, select NB-IoT, which is suitable for wide coverage, high reliability and low power consumption. After selecting the appropriate protocol, install and configure a gateway device compatible with the LPWAN protocol to receive data from sensor nodes and upload it to the cloud.

[0031] S22, then perform network configuration, first initialize the LPWAN module, set parameters such as frequency, bandwidth, data rate, etc., to ensure that the sensor nodes can communicate with the LPWAN gateway, then configure the gateway device to connect to the Internet, and set parameters such as network server address and gateway ID to ensure that the gateway can successfully receive data from the sensor nodes;

[0032] S23, then packing the processed data into a data frame that complies with the LPWAN protocol, the data frame includes a sensor ID, a timestamp, environmental data, etc., and sending the data frame from the sensor node to the gateway device through the LPWAN protocol, using a low-power transmission mode to ensure energy saving;

[0033] S24. Finally, after the gateway device receives the data frame transmitted by the sensor node, it performs preliminary processing on the data and uploads the processed data to the cloud server through the Internet. It adopts a secure transmission protocol to ensure data security, stores the received environmental data on the cloud server, and establishes a database to index the data according to time, sensor ID, etc.

[0034] Preferably, the user interface module in step S4 realizes environmental data display through front-end development, data visualization, front-end and back-end communication, back-end support, deployment and optimization. First, HTML, CSS and JavaScript are used to build the user interface, and the React framework is combined to improve development efficiency and maintainability; Chart.js and other libraries are used to create line charts, bar charts and other charts to dynamically display environmental data; AJAX is used to communicate with the back-end server to obtain and display data; the back-end uses Node.js and other frameworks to provide API interfaces to process and transmit environmental data in the database; finally, the application is deployed to a cloud service or server, and CI / CD tools are configured for automated deployment. This series of programming is used to ensure a good display effect of environmental data.

[0035] Compared with the prior art, the advantages and positive effects of the present invention are that the present invention collects environmental data such as temperature, humidity, air quality and light intensity in real time through multiple sensor nodes in the perception layer, solving the problems of data lag and long monitoring cycle caused by manual sampling and laboratory analysis in traditional methods. Low-power wide area network protocol is used for data transmission, and NB-IoT technology is selected in particular to ensure wide coverage, high reliability and low power consumption of data transmission, thereby improving transmission efficiency. The application layer uses variational autoencoders to optimize the time series analysis algorithm to provide more accurate environmental status prediction. The user interface realizes dynamic display of data and friendly user experience through front-end development and data visualization technology. DETAILED DESCRIPTION

[0036] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described below in conjunction with embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0037] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways than those described herein. Therefore, the present invention is not limited to the specific embodiments of the following disclosure.

[0038] Embodiment, the traditional environmental monitoring method mainly relies on manual sampling and laboratory analysis, which has the defects of long monitoring cycle, data lag and limited coverage. In addition, the manual sampling and laboratory analysis process is cumbersome, costly, and easily interfered by human factors. The accuracy and reliability of the data are difficult to guarantee, and the data transmission efficiency is low, which affects the timeliness of monitoring. At the same time, the energy consumption is high, which increases the monitoring cost.

[0039] In order to overcome these shortcomings, the present invention proposes a real-time environmental monitoring method based on intelligent Internet of Things, which aims to collect environmental data in real time through multiple sensor nodes in the perception layer, and efficiently transmit the data to the application layer with the help of the wireless communication protocol of the network layer. The application layer uses intelligent algorithms to accurately analyze and process the data, thereby achieving accurate prediction and timely monitoring of the environmental state. The first thing to consider is the data collection part. The present invention collects environmental data through multiple sensor nodes in the perception layer. Each sensor node includes one or more sensors for detecting different parameters in the environment. The sensors include temperature sensors, humidity sensors, air quality sensors, light sensors, etc. The collected data includes environmental parameters such as temperature, humidity, air pollutant concentration, and light intensity.

[0040] Considering that the network layer needs to transmit the environmental data collected by the perception layer to the application layer through the wireless communication protocol, the present invention proposes to use the low-power wide area network protocol for data transmission. First, NB-IoT, which is suitable for wide coverage, high reliability and low power consumption, is selected. After selecting a suitable protocol, a gateway device compatible with the LPWAN protocol is installed and configured to receive the data of the sensor node and upload it to the cloud. Next, the network configuration is performed. First, the LPWAN module is initialized, and parameters such as frequency, bandwidth, and data rate are set to ensure that the sensor node can communicate with the LPWAN gateway. Then, the gateway device is configured to connect to the Internet, and parameters such as the network server address and gateway ID are set to ensure that the gateway can successfully receive the data of the sensor node. Subsequently, the processed data is packaged into a data frame that complies with the LPWAN protocol. The data frame includes a sensor ID, a timestamp, environmental data, etc. The data frame is sent from the sensor node to the gateway device through the LPWAN protocol, and a low-power transmission mode is adopted to ensure energy saving. Finally, after the gateway device receives the data frame transmitted by the sensor node, it performs preliminary processing on the data and uploads the processed data to the cloud server through the Internet. A secure transmission protocol is adopted to ensure data security. The received environmental data is stored on the cloud server, and a database is established to index the data according to time, sensor ID, etc. Use low-power wide area network protocol to transmit environmental data, which is energy-saving and safe, with optimized configuration and efficient storage.

[0041] After the data is collected and transmitted, considering that the traditional time series analysis algorithm has low adaptability and weak ability to capture complex patterns, the present invention uses a variational autoencoder to optimize the time series analysis algorithm to monitor changes in environmental quality in the future. First, the encoder is defined to map the time series data to the latent variable space. The input layer inputs the time series data x. The hidden layer is one or more fully connected layers for feature extraction, and the output layer has two outputs, which are the mean μ and standard deviation σ of the latent variable. The sampling layer is defined to sample the latent variable z from the latent variable distribution. The sampling formula is:

[0042] z=μ+σ·ε

[0043] Where ε is the random noise of the standard normal distribution; the decoder is defined to reconstruct the hidden variable z into time series data. The input layer is the hidden variable z, the hidden layer is one or more fully connected layers for extracting features, and the output layer is the reconstructed time series data Then define the loss function, reconstruction loss to measure the reconstructed time series data The difference between the original data x is:

[0044]

[0045] Use KL divergence loss to measure the difference between the latent variable distribution q(zx) and the prior distribution p(z):

[0046]

[0047] The total loss is the sum of the reconstruction loss and the KL divergence loss:

[0048]

[0049] Among them, Γ is the total loss, β is the hyperparameter that weighs the reconstruction loss and KL divergence loss; then compile the model, select the Adam optimizer to compile the model, and the optimizer formula is:

[0050]

[0051] Among them, θ is the model parameter and η is the learning rate;

[0052] Use training data to train the model and update model parameters. The training formula is:

[0053]

[0054] Among them, θ * The parameters are updated after model training; finally, the validation data is used to evaluate the model performance, adjust the model hyperparameters, input new time series data, and use the trained variational autoencoder model for prediction. Compared with the traditional time series analysis algorithm, the present invention proposes to use variational autoencoders to optimize the time series analysis algorithm, which can better adapt to data changes and thus obtain more accurate prediction results.

[0055] Finally, the user interface module on the application layer realizes the display of environmental data through front-end development, data visualization, front-end and back-end communication, back-end support, deployment and optimization. First, HTML, CSS, and JavaScript are used to build the user interface, and the React framework is combined to improve development efficiency and maintainability; Chart.js and other libraries are used to create line charts, bar charts and other charts to dynamically display environmental data; AJAX is used to communicate with the back-end server to obtain and display data; the back-end uses Node.js and other frameworks to provide API interfaces to process and transmit environmental data in the database; finally, the application is deployed to the cloud service or server, and CI / CD tools are configured for automated deployment. This series of programming integrates a variety of technologies. Compared with traditional methods, the display is more efficient and intuitive, and the deployment is more convenient and stable, ensuring a good display effect of environmental data.

[0056] The above description is only a preferred embodiment of the present invention and does not limit the present invention in other forms. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A real-time environmental monitoring method based on intelligent Internet of Things, characterized in that: Includes the following parts: S1, the perception layer collects environmental data through multiple sensor nodes, each of which includes one or more sensors for detecting different parameters in the environment; S2, the network layer transmits the environmental data collected by the perception layer to the application layer through a wireless communication protocol, and the wireless communication protocol uses a low-power wide area network protocol for data transmission; S3, the application layer analyzes and processes the received environmental data, and predicts and monitors the environmental status through an intelligent algorithm. The intelligent algorithm uses a variational autoencoder to optimize the time series analysis algorithm to monitor changes in environmental quality over a period of time in the future; S4, the user interface module on the application layer realizes environmental data display through front-end development, data visualization, front-end and back-end communication, back-end support, deployment and optimization; The step S3 uses a variational autoencoder to optimize the time series analysis algorithm to monitor changes in environmental quality over a period of time in the future. The implementation steps are: S31. First, define the encoder to map the time series data to the latent variable space. The input layer inputs the time series data x. The hidden layer is one or more fully connected layers for extracting features. The output layer has two outputs, which are the mean μ and standard deviation σ of the latent variable. Define the sampling layer to sample the latent variable z from the latent variable distribution. The sampling formula is: z=μ+σ·ε Where ε is the random noise of the standard normal distribution; the decoder is defined to reconstruct the hidden variable z into time series data. The input layer is the hidden variable z, the hidden layer is one or more fully connected layers for extracting features, and the output layer is the reconstructed time series data S32, then define the loss function, reconstruction loss to measure the reconstructed time series data The difference between the original data x is: Use KL divergence loss to measure the difference between the latent variable distribution q(zx) and the prior distribution p(z): The total loss is the sum of the reconstruction loss and the KL divergence loss: Among them, Γ is the total loss, β is the hyperparameter that weighs the reconstruction loss and KL divergence loss; S33, then compile the model, select Adam optimizer to compile the model, the optimizer formula is: i t+1 =θ t -η·▽ θ C Among them, θ is the model parameter and η is the learning rate; Use training data to train the model and update model parameters. The training formula is: Among them, θ * are the parameters updated after model training; S34. Finally, use the validation data to evaluate the model performance, adjust the model hyperparameters, input new time series data, and use the trained variational autoencoder model for prediction.

2. The method for real-time environmental monitoring based on intelligent Internet of Things according to claim 1 is characterized in that: The sensors described in step S1 include temperature sensors, humidity sensors, air quality sensors, light sensors, etc., and the collected data include environmental parameters such as temperature, humidity, air pollutant concentration, and light intensity.

3. The method for real-time monitoring of environment based on intelligent Internet of Things according to claim 1, characterized in that: The network layer of step S2 uses the low power wide area network protocol for data transmission, and the steps are as follows: S21. First, select NB-IoT, which is suitable for wide coverage, high reliability and low power consumption. After selecting the appropriate protocol, install and configure a gateway device compatible with the LPWAN protocol to receive data from sensor nodes and upload it to the cloud. S22, then perform network configuration, first initialize the LPWAN module, set parameters such as frequency, bandwidth, data rate, etc., to ensure that the sensor nodes can communicate with the LPWAN gateway, then configure the gateway device to connect to the Internet, and set parameters such as network server address and gateway ID to ensure that the gateway can successfully receive data from the sensor nodes; S23, then packing the processed data into a data frame that complies with the LPWAN protocol, the data frame includes a sensor ID, a timestamp, environmental data, etc., and sending the data frame from the sensor node to the gateway device through the LPWAN protocol, using a low-power transmission mode to ensure energy saving; S24. Finally, after the gateway device receives the data frame transmitted by the sensor node, it performs preliminary processing on the data and uploads the processed data to the cloud server through the Internet. It adopts a secure transmission protocol to ensure data security, stores the received environmental data on the cloud server, and establishes a database to index the data according to time, sensor ID, etc.

4. The method for real-time monitoring of environment based on intelligent Internet of Things according to claim 1, characterized in that: The user interface module in step S4 realizes environmental data display through front-end development, data visualization, front-end and back-end communication, back-end support, deployment and optimization. First, HTML, CSS, and JavaScript are used to build the user interface, and the React framework is combined to improve development efficiency and maintainability; Chart.js and other libraries are used to create line charts, bar charts and other charts to dynamically display environmental data; AJAX is used to communicate with the back-end server to obtain and display data; the back-end uses Node.js and other frameworks to provide API interfaces to process and transmit environmental data in the database; finally, the application is deployed to the cloud service or server, and the CI / CD tool is configured for automatic deployment. This series of programming is used to ensure a good display effect of environmental data.

Images