A remote environment monitoring method and system based on Internet of Things

Through the remote environmental monitoring method based on the Internet of Things, using Bayesian network model and seasonal division technologies to monitor and analyze environmental parameters in real time, solving the problems of environmental parameter changes and abnormal identification in the existing technology, improving prediction accuracy and Data quality monitoring capability.

CN118520392BActive Publication Date: 2024-11-08SHENZHEN SHENGRUN ENG CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410586516.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-13
Publication Date
2024-11-08
Estimated Expiration
2044-05-13

AI Technical Summary

Technical Problem

Existing remote environmental monitoring technology is difficult to accurately capture dynamic changes and abnormalities in environmental parameters, resulting in large differences in the prediction results from actual conditions, and lacks dynamic monitoring capabilities, making it impossible to identify and handle data quality problems in a timely manner.

Method used

Using remote environmental monitoring method based on the Internet of Things, by building Internet platforms and preset models, we obtain monitoring instructions in real time, collect data and analyze them, use Bayesian network model and seasonal division to calculate confidence intervals, combine time series analysis and sparseness Data processing, generate predicted trend results and adjust equipment operating parameters, and monitor and adjust environmental parameters in real time.

Benefits of technology

It improves the accuracy and reliability of environmental parameter prediction, can more comprehensively consider the impact of multiple factors on environmental parameters, timely identify data quality issues and generate alarm instructions, and provide real-time and accurate environmental monitoring information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118520392B_ABST
    Figure CN118520392B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of environmental data monitoring, and in particular to a remote environmental monitoring method and system based on the Internet of Things, including building an Internet platform for exchanging data with a user end and a device end; obtaining a monitoring instruction from a user end in real time, and analyzing the target data that the monitoring instruction needs to monitor when receiving the monitoring instruction; obtaining the device type and acquisition frequency corresponding to the target data according to the target data; based on the device type and acquisition frequency, controlling the target device to collect data according to the acquisition frequency, and receiving the acquisition results sent by the target device in real time; inputting the acquisition results into a preset model, correlating and analyzing the collection results through the model, and outputting the prediction trend results of the environmental data through the model; and sending the prediction results and the collection trend results to the user end. Correlating and analyzing the collection results through the model can more comprehensively consider the influence of various factors on environmental parameters, thereby improving the accuracy and reliability of the prediction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of environmental data monitoring, and in particular to a remote environmental monitoring method and system based on the Internet of Things. Background Art

[0002] Remote environmental monitoring is a technology that uses sensor technology and communication networks to monitor environmental parameters in real time or periodically. This monitoring method does not require physical access to the monitoring point, but transmits sensor data to the monitoring center or user equipment through remote communication technology.

[0003] Traditional environmental parameter prediction models may predict the results of environmental parameters based on historical data, but in actual environments, environmental conditions may be affected by a variety of factors, such as time and season. These factors may cause fluctuations and changes in environmental parameters (such as temperature, humidity, and air quality), and these changes are often not captured by traditional prediction models. If these environmental change factors are not taken into account, the prediction results may differ greatly from the actual situation, resulting in decision-making errors; secondly, existing technologies may not provide the ability to dynamically monitor changes in environmental parameters over time, resulting in the inability to identify and handle data quality issues in a timely manner. For example, at certain specific time points (such as peak hours of the day or when specific weather events occur), air quality may drop significantly due to human activities, traffic flow, industrial emissions, and other factors. Static thresholds may not be able to adapt to seasonal changes or abnormal environmental parameters at specific time points, which means that in some cases, even normal changes may be mistakenly marked as abnormal.

[0004] Therefore, the prior art has defects and needs to be improved. Summary of the invention

[0005] In order to solve one or several problems in the prior art, the main purpose of this application is to provide a remote environment monitoring method and system based on the Internet of Things.

[0006] In order to achieve the above-mentioned invention object, the present application proposes a remote environment monitoring method based on the Internet of Things, the method comprising:

[0007] Build an Internet platform for exchanging data with users and devices;

[0008] Acquire the monitoring instruction of the user end in real time, and when the monitoring instruction is received, analyze the target data required to be monitored by the monitoring instruction;

[0009] According to the target data, the device type and collection frequency required for acquiring the target data;

[0010] Based on the device type and the collection frequency, control the target device to collect data according to the collection frequency, and receive the collection results sent by the target device in real time;

[0011] Input the collected results into a preset model, perform association and regularity analysis on the collected results through the model, and output the predicted trend results of the environmental data through the model;

[0012] Based on the prediction results and the collected trend results, it is sent to the user end.

[0013] Furthermore, the step of inputting the collected results into a preset model, performing association and regularity analysis on the collected results through the model, and outputting the predicted trend results of the environmental data through the model includes:

[0014] Constructing a Bayesian network model, obtaining historical collection results and classifying the historical collection results by indicators, and training the Bayesian network model with the classified historical collection results to obtain a trained Bayesian network model;

[0015] Inputting the collected results into a Bayesian network model, classifying the collected results by indicators through the trained Bayesian network model, fitting a trend line for the data of each category using a time series, and outputting a trend prediction result through the trained Bayesian network model;

[0016] The historical collection results are seasonally divided, and the confidence interval of the trend prediction is calculated by combining the seasonal division and trend prediction to quantify the uncertainty of the prediction.

[0017] Further, after the step of calculating the confidence interval of the trend forecast for quantifying the uncertainty of the forecast, the method comprises:

[0018] Determine the confidence interval for each forecast value based on the results of calculating the confidence interval for the trend forecast;

[0019] For each predicted value, calculate the width of the confidence interval, including expressing it as the difference between the upper and lower bounds;

[0020] Comparing the width of the confidence interval for each predicted value with the predicted value itself, analyzing the proportion of the confidence interval width, and the proportion is used to determine the degree of uncertainty in the prediction;

[0021] Analyzing the prediction results according to the proportion of the confidence interval width;

[0022] When the ratio of the confidence interval of the predicted value is greater than a preset threshold, it is judged that the uncertainty of the prediction result is high and an adjustment strategy is generated.

[0023] Furthermore, after the step of outputting the prediction trend result of the environmental data through the model, the method includes:

[0024] When the target data is the monitored temperature, all temperature values ​​collected by the device are obtained, and the temperature change rate is calculated according to all the temperature values;

[0025] Based on the temperature change rate and the predicted trend, adjusting operating parameters of the temperature regulating device to maintain the outdoor temperature range;

[0026] When the target data is to monitor air quality, the concentration of pollutants in the air collected by the device is obtained, including the concentration of suspended particulate matter, carbon monoxide, sulfur dioxide, nitrogen dioxide, ozone and volatile organic compounds;

[0027] All parameters of the pollutant concentration are integrated and the air quality index is calculated;

[0028] According to the air quality index and the predicted trend, the operating parameters of the air purification equipment are adjusted.

[0029] Furthermore, when the ratio of the confidence interval of the predicted value is greater than a preset threshold, it is judged that the uncertainty of the prediction result is high, and an adjustment strategy is generated. The method includes:

[0030] According to time and geographical factors, the historical collection results are grouped to obtain multiple data subsets;

[0031] Performing difference identification on each of the data subsets, and when the data subset presents a long-tail distribution, inputting the data subset with the long-tail distribution into a preset sparse data processing model, and fitting the distribution characteristics of the data subset through the sparse data processing model;

[0032] The importance of the data subset is adjusted according to the distribution characteristics, including strengthening or weakening the importance.

[0033] Furthermore, after the step of sending the prediction result and the collection trend result to the user end, the method further includes analyzing the data quality of the collection result. The method includes:

[0034] According to the time series, thresholds of the collection results at different times are set to determine whether the data at different times meet the environmental quality standards;

[0035] Determine whether the data of the prediction result is discontinuous in the time series, and when the data of the collection result is not discontinuous in the time series, divide the data of the collection result into multiple time periods;

[0036] Calculate the data of each time period using the interquartile range algorithm, and determine whether the calculation result of the corresponding time period is an outlier;

[0037] When the calculation result of the corresponding time period is an abnormal value, it is determined that the data quality of the corresponding time period does not meet the standard, and an alarm instruction message is generated and sent to the user end.

[0038] Furthermore, the data of each time period is calculated by an interquartile range algorithm, and it is determined whether there are outliers in the data of each time period. The method includes:

[0039] The calculation formula of the interquartile range is:

[0040] Calculate the interquartile range for each time period: IQR=Q3-Q1, where Q1 and Q3 are quartiles, and IQR is the interquartile range;

[0041] Lower abnormal threshold = Q1-1.5*IQR;

[0042] Upper abnormal threshold = Q3 + 1.5 * IQR;

[0043] Determine whether the data point in each time period data is within the range of the lower abnormal threshold and the upper abnormal threshold;

[0044] When the data point is not within the range, it is judged as an abnormal data point;

[0045] When the data point is within the range, it is judged as a normal data point;

[0046] The proportion of the abnormal data points to the total data points is analyzed, and when the proportion of the abnormal data points is greater than the proportion threshold of the time period, the data of the time period is determined to be an abnormal value.

[0047] The present application also provides a remote environment monitoring system based on the Internet of Things, including:

[0048] Building modules for building Internet platforms for exchanging data with user terminals and device terminals;

[0049] The analysis module is used to obtain the monitoring instruction of the user terminal in real time, and when the monitoring instruction is received, analyze the target data that needs to be monitored by the monitoring instruction;

[0050] An acquisition module, used for acquiring the device type and acquisition frequency required for the target data according to the target data;

[0051] A control module, used to control the target device to collect data according to the collection frequency based on the device type and the collection frequency, and receive the collection results sent by the target device in real time;

[0052] A prediction module, used to input the collected results into a preset model, perform association and regularity analysis on the collected results through the model, and output the predicted trend results of the environmental data through the model;

[0053] The sending module is used to send the prediction results and the collection trend results to the user end.

[0054] The present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0055] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0056] The remote environment monitoring method and system based on the Internet of Things of the embodiment of the present application can monitor the changes of environmental parameters in real time and make predictions by acquiring monitoring instructions and collecting data in real time, and analyzing them in combination with preset models, which helps to grasp the environmental status more accurately. Considering that environmental conditions may be affected by various factors and cause fluctuations and changes, this method can more comprehensively consider the influence of various factors on environmental parameters by establishing a model to correlate and analyze the laws of the collected results, thereby improving the accuracy and reliability of the prediction. Considering that environmental conditions may be affected by various factors and cause fluctuations and changes, this method can more comprehensively consider the influence of various factors on environmental parameters by establishing a model to correlate and analyze the laws of the collected results, thereby improving the accuracy and reliability of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a flow chart of a remote environment monitoring method based on the Internet of Things according to an embodiment of the present application;

[0058] Figure 2 This is a flow chart of a remote environment monitoring method based on the Internet of Things according to an embodiment of the present application;

[0059] Figure 3 This is a schematic block diagram of the structure of a remote environment monitoring system based on the Internet of Things according to an embodiment of the present application;

[0060] Figure 4 A schematic block diagram of the structure of a computer device according to an embodiment of the present application.

[0061] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0063] Reference Figure 1 In an embodiment of the present application, a remote environment monitoring method based on the Internet of Things is provided, and the method comprises:

[0064] S1. Build an Internet platform for exchanging data with the user end and the device end;

[0065] S2. Acquire the monitoring instruction of the user terminal in real time, and when the monitoring instruction is received, analyze the target data required to be monitored by the monitoring instruction;

[0066] S3. According to the target data, obtaining the device type and collection frequency required for the target data;

[0067] S4. Based on the device type and the acquisition frequency, control the target device to collect data according to the acquisition frequency, and receive the acquisition result sent by the target device in real time;

[0068] S5, inputting the collected results into a preset model, performing association and regularity analysis on the collected results through the model, and outputting the predicted trend results of the environmental data through the model;

[0069] S6. Send the prediction result and the collected trend result to the user end.

[0070] As described in steps S1-S2 above, an Internet platform can be built using Web services or API interfaces, where users can submit monitoring instructions through a Web browser or application, and the device sends data through the IoT device management platform. The platform needs to have data transmission, storage and processing capabilities. After the user submits the monitoring instruction through the platform, the system will receive and parse the instruction content in real time. The instruction may include information such as the parameters to be monitored and the monitoring range, and the system determines the target data to be monitored based on this information.

[0071] As described in step S3 above, the system queries the pre-configured device type and acquisition frequency information in the database. This information is stored in the device management database, and the system determines which devices should be controlled for data acquisition and the frequency of acquisition by querying this information. With the pre-configured device type and acquisition frequency information, the system can automatically determine the best data acquisition strategy, improving the efficiency and accuracy of data acquisition. This helps to overcome the manual operation and inefficiency of traditional systems in device selection and acquisition frequency setting.

[0072] As described in step S4 above, the system sends a control message to the target device through an IoT protocol (such as MQTT, CoAP, etc.), and the device starts to collect data at a specified frequency after receiving the message. At the same time, the system monitors the data sent by the device to ensure that the collected results are received and processed in real time. Through IoT technology, the system can remotely control and monitor the device and receive data in real time, which improves the automation of the system and the real-time nature of data collection. This helps to overcome the limitations of traditional systems in device control and data transmission.

[0073] As described in the above steps S5-S6, the collected data is input into a preset machine learning model through an API interface or a file transfer method. The model will associate and analyze the data and predict the trend of future environmental data. The analysis results are usually output in the form of charts or reports. The predicted trend results and the collected trend results obtained by the model analysis are packaged into reports or notifications and sent to the user end through the Internet platform. Users can receive and view this information through mobile phones, computers and other devices. The sending method can be an email, text message, instant message or application push notification. By analyzing and predicting the collected results through a machine learning model, the changes and trends of environmental parameters can be captured more accurately, improving the accuracy and reliability of the prediction. This helps to overcome the limitations of traditional systems in predictive models and analysis methods. The predicted results and the collected trend results are sent to the user end to provide users with real-time and accurate environmental monitoring information, which helps users make data-based decisions.

[0074] Specifically, by acquiring monitoring instructions and collecting data in real time, combined with preset models for analysis, it is possible to monitor changes in environmental parameters in real time and make predictions, which helps to grasp the environmental status more accurately. Considering that environmental conditions may be affected by a variety of factors, resulting in fluctuations and changes, this method can more comprehensively consider the impact of various factors on environmental parameters by establishing a model to correlate and analyze the patterns of the collected results, thereby improving the accuracy and reliability of the prediction. Considering that environmental conditions may be affected by a variety of factors, resulting in fluctuations and changes, this method can more comprehensively consider the impact of various factors on environmental parameters by establishing a model to correlate and analyze the patterns of the collected results, thereby improving the accuracy and reliability of the prediction.

[0075] Reference Figure 2 In one embodiment, the inputting of the collected results into a preset model, performing association and regularity analysis on the collected results through the model, and outputting the predicted trend results of the environmental data through the model includes:

[0076] S41, constructing a Bayesian network model, obtaining historical collection results and classifying the historical collection results by indicators, and training the Bayesian network model with the classified historical collection results to obtain a trained Bayesian network model;

[0077] S42, inputting the collection results into a Bayesian network model, classifying the collection results by indicators through the trained Bayesian network model, fitting a trend line for the data of each category using a time series, and outputting a trend prediction result through the trained Bayesian network model;

[0078] S43, seasonal division is performed on the historical collection results, and the confidence interval of the trend prediction is calculated in combination with the seasonal division and the trend prediction, so as to quantify the uncertainty of the prediction.

[0079] As described in the above steps, a Bayesian network is a graphical model for modeling probabilistic relationships between variables. It consists of nodes (variables) and directed edges (representing dependencies between variables). By observing the data, the conditional probability distribution between nodes can be estimated to construct a Bayesian network. Building a Bayesian network model can capture the potential correlation between environmental data, such as the dependency between PM2.5 concentration and temperature and humidity. This helps to more accurately understand the interaction between environmental data and improve the accuracy of prediction. Classify the historical collection results according to different indicators, such as classifying PM2.5 concentration, temperature, humidity and other data into different indicator categories. By classifying the indicators of the historical collection results, we can more clearly understand the impact of each indicator on the environmental data, which helps to more accurately establish the dependency between nodes in the Bayesian network model. Use the classified historical collection results to train the Bayesian network model, that is, to estimate the conditional probability distribution between nodes. By training the Bayesian network model, the association rules between environmental data can be learned based on historical data, providing a basis for subsequent predictions. The real-time collected environmental data is classified according to indicators and input into the trained Bayesian network model. By inputting real-time collected data and classifying indicators, inferences can be made in the Bayesian network model to predict future trends of different indicator categories. Combining seasonal division and trend prediction, the confidence interval of the predicted trend is calculated through statistical methods. Calculating the confidence interval can quantify the uncertainty of the prediction results, provide decision makers with more comprehensive information, help them evaluate the reliability of the prediction results, and make corresponding countermeasures.

[0080] In one embodiment, after the step of calculating a confidence interval for the trend forecast to quantify the uncertainty of the forecast, the method comprises:

[0081] Determine the confidence interval for each forecast value based on the results of calculating the confidence interval for the trend forecast;

[0082] For each predicted value, calculate the width of the confidence interval, including expressing it as the difference between the upper and lower bounds;

[0083] Comparing the width of the confidence interval for each predicted value with the predicted value itself, analyzing the proportion of the confidence interval width, and the proportion is used to determine the degree of uncertainty in the prediction;

[0084] Analyzing the prediction results according to the proportion of the confidence interval width;

[0085] When the ratio of the confidence interval of the predicted value is greater than a preset threshold, it is judged that the uncertainty of the prediction result is high and an adjustment strategy is generated.

[0086] As described above, confidence intervals are calculated for the forecasted trend using statistical methods for seasonality and trend forecasting. This may involve using time series analysis, statistical modeling, or machine learning methods to forecast future trends and quantifying the uncertainty of the forecast using confidence intervals. Calculating confidence intervals can help understand the reliability of the forecast results. The narrower the confidence interval, the more reliable the forecast result; the wider the confidence interval, the higher the uncertainty of the forecast result. For each forecast value, determine its confidence interval based on the results of calculating the confidence interval for the trend forecast. This often involves using statistical methods, such as confidence interval calculations based on the normal distribution or other distributions. Determining the confidence interval for each forecast value can provide a basis for subsequent analysis and help us understand the degree of uncertainty in each forecast value. For each forecast value, calculate the width of its confidence interval, which is the difference between the upper and lower bounds. Calculating the width of the confidence interval can quantify the degree of uncertainty in the forecast result. A wider confidence interval means higher uncertainty, while a narrower confidence interval means lower uncertainty. Compute the proportion of the width of the confidence interval for each forecast value compared to the forecast value itself. Analyzing the proportion of confidence interval width can help users understand the degree of uncertainty in the forecast results. A larger proportion indicates higher uncertainty, while a smaller proportion indicates lower uncertainty. The forecast results are analyzed based on the proportion of confidence interval width. When the proportion is greater than the preset threshold, the forecast result is judged to have high uncertainty. By analyzing the proportion of confidence interval width, the degree of uncertainty in the forecast results can be determined, and adjustment strategies can be generated as needed to deal with high uncertainty forecast results.

[0087] It is worth mentioning that the use of seasonal partitioning is to better capture the cyclical changes in time series data, thereby improving the accuracy of trend forecasting. In some cases, the data may show seasonal fluctuations, such as we are monitoring the air quality of a city, and we hope to predict the air quality index (AQI) in the next few months. In this case, the air quality may be affected by seasonality, such as higher ozone levels in summer and higher particulate matter concentrations in winter. If we ignore this seasonal effect and directly analyze the overall time series, the accuracy of the forecast results may decrease. For example, suppose we start monitoring during the summer when the ozone content is high. If we directly use the data during this period to predict the air quality in the next few months, we may overestimate the overall trend because the ozone content may decrease with the change of seasons, resulting in a lower-than-expected air quality index in the next few months. Therefore, by using seasonal partitioning, we can better capture the changing patterns of air quality in different seasons and more accurately predict the future air quality index. In this way, after considering seasonal factors in the forecast process, we can better cope with the changes in air quality brought about by different seasons and improve the accuracy of the forecast results.

[0088] In one embodiment, after the step of outputting the predicted trend result of the environmental data through the model, the method includes:

[0089] When the target data is the monitored temperature, all temperature values ​​collected by the device are obtained, and the temperature change rate is calculated according to all the temperature values;

[0090] Based on the temperature change rate and the predicted trend, adjusting operating parameters of the temperature regulating device to maintain the outdoor temperature range;

[0091] When the target data is to monitor air quality, the concentration of pollutants in the air collected by the device is obtained, including the concentration of suspended particulate matter, carbon monoxide, sulfur dioxide, nitrogen dioxide, ozone and volatile organic compounds;

[0092] All parameters of the pollutant concentration are integrated and the air quality index is calculated;

[0093] According to the air quality index and the predicted trend, the operating parameters of the air purification equipment are adjusted.

[0094] As mentioned above, all the temperature values ​​collected by the device are obtained, and then the temperature change rate is calculated based on these values. The purpose of this step is to predict future temperature changes by monitoring the temperature change trend. By calculating the temperature change rate, it is possible to understand whether the temperature is rising or falling, so that the operating parameters of the temperature control device can be adjusted according to the predicted trend to maintain the outdoor temperature within the required range. The pollutant concentrations in the air collected by the device are obtained, including the concentrations of suspended particulate matter, carbon monoxide, sulfur dioxide, nitrogen dioxide, ozone, and volatile organic compounds, and then these parameters are fused to calculate the air quality index. By monitoring the concentrations of multiple pollutants and fusing the data, the air quality situation can be evaluated more comprehensively, not just limited to a single pollutant. Calculating the air quality index can transform complex pollutant data into a more intuitive and comparable indicator, which helps to evaluate the quality of air quality in real time.

[0095] In one embodiment, when the ratio of the confidence interval of the predicted value is greater than a preset threshold, it is judged that the uncertainty of the prediction result is high, and an adjustment strategy is generated. The method includes:

[0096] According to time and geographical factors, the historical collection results are grouped to obtain multiple data subsets;

[0097] Performing difference identification on each of the data subsets, and when the data subset presents a long-tail distribution, inputting the data subset with the long-tail distribution into a preset sparse data processing model, and fitting the distribution characteristics of the data subset through the sparse data processing model;

[0098] The importance of the data subset is adjusted according to the distribution characteristics, including strengthening or weakening the importance.

[0099] As described above, the historical collection results are divided into multiple data subsets according to time and geographical factors. Then, the differences are identified for each data subset, especially for the data subset with long tail distribution. By dividing the historical data into multiple subsets and identifying the data subset with long tail distribution, the distribution and characteristics of the data can be better understood. The data with long tail distribution usually represents important information in a small number of samples, so special processing of it can improve the accuracy and robustness of the model. The data subset with long tail distribution is input into the preset sparse data processing model, and the distribution characteristics of the data subset are fitted through the model. The sparse data processing model can help the system better understand and process the data with long tail distribution and extract the effective information therein. By modeling the data with long tail distribution, the system can more accurately evaluate the uncertainty and adjust the confidence interval of the prediction result accordingly. According to the distribution characteristics of the fitted data subset, the emphasis on each data subset is adjusted, including strengthening or weakening the emphasis. By adjusting the emphasis according to the distribution characteristics of the data subset, the system can more finely control the way to deal with uncertainty. Strengthening the emphasis on key data subsets can improve the system's grasp of important information; reducing the emphasis on general data subsets can reduce unnecessary interference and misjudgment.

[0100] In one embodiment, after the step of sending the prediction result and the collection trend result to the user end, the method further includes analyzing the data quality of the collection result. The method includes:

[0101] According to the time series, thresholds of the collection results at different times are set to determine whether the data at different times meet the environmental quality standards;

[0102] Determine whether the data of the prediction result is discontinuous in the time series, and when the data of the collection result is not discontinuous in the time series, divide the data of the collection result into multiple time periods;

[0103] Calculate the data of each time period using the interquartile range algorithm, and determine whether the calculation result of the corresponding time period is an outlier;

[0104] When the calculation result of the corresponding time period is an abnormal value, it is determined that the data quality of the corresponding time period does not meet the standard, and an alarm instruction message is generated and sent to the user end.

[0105] As mentioned above, thresholds for collection results at different times are set according to the time series to determine whether the data meets the environmental quality standards. The implementation principle is to set predetermined thresholds and compare the collection results with these thresholds to determine whether the data quality meets the requirements. This helps to detect data anomalies or non-compliance in advance, so that timely measures can be taken to correct or warn. Detect whether the prediction result data is discontinuous in the time series and divide the collection result data into multiple time periods. The implementation principle is to detect whether there is discontinuity by analyzing the continuity between data points in the time series. If the data continuity is poor, it may indicate that anomalies or failures occurred during the data collection process. Dividing the data into time periods helps to analyze the data quality in each time period more finely. The interquartile range algorithm is calculated for the data in each time period, and it is determined whether the calculation result is an outlier. The interquartile range algorithm is used to identify outliers in the data. It divides the data into four equal parts, calculates the first quartile (Q1) and the third quartile (Q3), and then calculates the interquartile range (IQR) based on these two values ​​to determine whether the data is an outlier. When an abnormal value is detected, the system will generate an alarm instruction message and send it to the user end. This helps users to understand the abnormal data quality in a timely manner so that they can take appropriate measures to deal with or adjust it.

[0106] In one embodiment, the method of calculating the data of each time period by an interquartile range algorithm and determining whether there are outliers in the data of each time period includes:

[0107] The calculation formula of the interquartile range is:

[0108] Calculate the interquartile range for each time period: IQR=Q3-Q1, where Q1 and Q3 are quartiles, and IQR is the interquartile range;

[0109] Lower abnormal threshold = Q1-1.5*IQR;

[0110] Upper abnormal threshold = Q3 + 1.5 * IQR;

[0111] Determine whether the data point in each time period data is within the range of the lower abnormal threshold and the upper abnormal threshold;

[0112] When the data point is not within the range, it is judged as an abnormal data point;

[0113] When the data point is within the range, it is judged as a normal data point;

[0114] The proportion of the abnormal data points to the total data points is analyzed, and when the proportion of the abnormal data points is greater than the proportion threshold of the time period, the data of the time period is determined to be an abnormal value.

[0115] As mentioned above, using the interquartile range algorithm, first calculate the first quartile (Q1) and the third quartile (Q3) of the data in each time period, and then calculate the interquartile range (IQR) from these two values. The interquartile range can well reflect the distribution of the data, and the calculated IQR can be used to identify the abnormality of the data, thereby helping to determine the threshold range of the abnormal value. According to the calculation results of the interquartile range, the lower limit abnormal threshold and the upper limit abnormal threshold are calculated respectively. The lower limit abnormal threshold is Q1 minus 1.5 times the IQR, and the upper limit abnormal threshold is Q3 plus 1.5 times the IQR. By determining the abnormal threshold range, normal data points and abnormal data points can be effectively distinguished, and then it can be determined whether the data has abnormal conditions. For the data in each time period, check whether the data point is within the range of the lower limit abnormal threshold and the upper limit abnormal threshold. If the data point is not within the range, it is judged as an abnormal data point; otherwise, it is judged as a normal data point. By judging whether the data point is within the abnormal threshold range, the abnormal data point can be effectively identified, thereby helping to process and correct the abnormal data. The proportion of abnormal data points in each time period to the total data points is analyzed. If the proportion is greater than the preset proportion threshold, the data in this time period is determined to be an outlier. By setting the proportion threshold, the abnormality of the data in the entire time period can be judged according to the proportion of abnormal data points, so as to more objectively evaluate the data quality and take corresponding measures. It is worth mentioning that the interquartile range is a non-parametric statistical method that does not depend on the specific distribution form of the data. This makes it suitable for environmental indicator data that does not conform to the normal distribution, ensures the versatility and robustness of the analysis, and is insensitive to extreme values ​​of the data. In the data analysis of environmental indicators, there may be outliers or outliers, and the interquartile range can effectively eliminate the interference of these extreme values ​​and more accurately reflect the true situation of the data. Although the interquartile range method can better identify anomalies in the data, its static threshold may not be able to fully adapt to the dynamic nature of environmental parameters changing over time. Environmental parameter anomalies at a specific time point may be mistakenly marked as anomalies, resulting in false positives or false negatives. For example, when environmental parameters change due to specific weather events or human activities, the static threshold may not accurately reflect the actual situation, thus affecting the accuracy and reliability of data quality analysis.

[0116] In a feasible embodiment, if the time series data is temperature data recorded hourly, a day can be divided into 24 time periods, and then the IQR of each time period is calculated. The formula is:

[0117] For the nth hour (n=1, 2, ..., 24), Q1 and Q3 are calculated.

[0118] Calculate IQR=Q3n-Q1n.

[0119] Determine the abnormal threshold: lower abnormal threshold = Q1n-1.5*IQRn, upper abnormal threshold = Q3n+1.5*IQRn;

[0120] When the data point is not within the range, it is judged as an abnormal data point;

[0121] When the data point is within the range, it is judged as a normal data point;

[0122] The proportion of the abnormal data points to the total data points is analyzed, and when the proportion of the abnormal data points is greater than the proportion threshold of the time period, the data of the time period is determined to be an abnormal value.

[0123] From the above analysis, a day is divided into 24 time periods because the temperature may change with different time periods of the day. For example, the temperature may be significantly different during the day and at night. The introduction of dynamic monitoring technology collects environmental parameter data in real time and dynamically adjusts the abnormal threshold to better adapt to the situation where environmental parameters change over time.

[0124] In another embodiment, information from different data sources is integrated, including various measurement data such as temperature, humidity, air quality, wind speed, rainfall, etc.; missing values, duplicate values ​​and outliers are processed for the information from the different data sources; time series analysis is performed on the data to analyze the changing trend and periodicity of the data in the time dimension, and the time series and spatial correlation features are combined to establish a joint modeling framework; abnormal data points are analyzed through the joint modeling framework, and the abnormal time and location are determined based on the abnormal data points; the cause of the abnormality is determined based on the device, abnormal time and abnormal location corresponding to the abnormal data point.

[0125] In another embodiment, the seasonal division of the historical collection results includes:

[0126] Obtaining the latitude and longitude information and climate type of the historical collection results, wherein the climate type includes tropical, temperate and frigid zones;

[0127] Analyze the change sequence of environmental data of historical collection results according to the latitude and longitude information and climate type;

[0128] Determine the start and end rules of each season according to the change sequence of the environmental data;

[0129] Determine whether the start and end patterns of each season are periodic;

[0130] When the start and end rules of each season are periodic, the seasonal division of the historical collection results is output.

[0131] The remote environment monitoring method based on the Internet of Things of the present application can monitor the changes of environmental parameters in real time and make predictions by acquiring monitoring instructions and collecting data in real time, and combining with preset models for analysis, which helps to grasp the environmental status more accurately. Considering that environmental conditions may be affected by various factors and cause fluctuations and changes, this method can more comprehensively consider the impact of various factors on environmental parameters by establishing a model to correlate and analyze the patterns of the collected results, thereby improving the accuracy and reliability of the prediction. Considering that environmental conditions may be affected by various factors and cause fluctuations and changes, this method can more comprehensively consider the impact of various factors on environmental parameters by establishing a model to correlate and analyze the patterns of the collected results, thereby improving the accuracy and reliability of the prediction.

[0132] Reference Figure 3 In an embodiment of the present application, a remote environment monitoring system based on the Internet of Things is also provided, including:

[0133] Building module 1, used to build an Internet platform for exchanging data with a user end and a device end;

[0134] Analysis module 2, used for obtaining the monitoring instruction of the user terminal in real time, and when receiving the monitoring instruction, analyzing the target data required to be monitored by the monitoring instruction;

[0135] An acquisition module 3 is used to acquire the device type and acquisition frequency required for the target data according to the target data;

[0136] A control module 4, for controlling the target device to collect data according to the collection frequency based on the device type and the collection frequency, and receiving the collection result sent by the target device in real time;

[0137] Prediction module 5, used for inputting the collected results into a preset model, performing association and regularity analysis on the collected results through the model, and outputting the predicted trend results of the environmental data through the model;

[0138] The sending module 6 is used to send the prediction result and the collection trend result to the user end.

[0139] As described above, it can be understood that the various components of the remote environment monitoring system based on the Internet of Things proposed in this application can realize the functions of any one of the remote environment monitoring methods based on the Internet of Things described above, and the specific structure will not be repeated.

[0140] Reference Figure 4 In an embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 4As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as monitoring data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a remote environment monitoring method based on the Internet of Things is implemented.

[0141] The above-mentioned processor executes the above-mentioned remote environment monitoring method based on the Internet of Things, including: building an Internet platform for exchanging data with the user end and the device end; obtaining the monitoring instructions of the user end in real time, and when receiving the monitoring instructions, analyzing the target data that needs to be monitored by the monitoring instructions; according to the target data, obtaining the device type and collection frequency required for the target data; based on the device type and collection frequency, controlling the target device to collect data according to the collection frequency, and receiving the collection results sent by the target device in real time; inputting the collection results into a preset model, correlating and analyzing the collection results through the model, and outputting the predicted trend results of the environmental data through the model; and sending them to the user end based on the predicted results and collection trend results.

[0142] The above-mentioned remote environmental monitoring method based on the Internet of Things can monitor the changes of environmental parameters in real time and make predictions by acquiring monitoring instructions and collecting data in real time, and combining with preset models for analysis, which helps to grasp the environmental status more accurately. Considering that environmental conditions may be affected by various factors and cause fluctuations and changes, this method can more comprehensively consider the impact of various factors on environmental parameters by establishing a model to correlate and analyze the patterns of the collected results, thereby improving the accuracy and reliability of the prediction. Considering that environmental conditions may be affected by various factors and cause fluctuations and changes, this method can more comprehensively consider the impact of various factors on environmental parameters by establishing a model to correlate and analyze the patterns of the collected results, thereby improving the accuracy and reliability of the prediction.

[0143] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a remote environment monitoring method based on the Internet of Things is implemented, including the following steps: building an Internet platform for exchanging data with a user end and a device end; obtaining monitoring instructions from the user end in real time, and when the monitoring instructions are received, analyzing the target data that needs to be monitored by the monitoring instructions; obtaining the device type and collection frequency required for the target data according to the target data; based on the device type and collection frequency, controlling the target device to collect data according to the collection frequency, and receiving the collection results sent by the target device in real time; inputting the collection results into a preset model, correlating and analyzing the collection results through the model, and outputting the predicted trend results of the environmental data through the model; and sending the predicted results and collection trend results to the user end based on the predicted results and collection trend results.

[0144] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0145] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, device, article or method including the element.

[0146] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A remote environment monitoring method based on the Internet of Things, characterized in that: The method comprises: Build an Internet platform for exchanging data with users and devices; Acquire the monitoring instruction of the user end in real time, and when the monitoring instruction is received, analyze the target data required to be monitored by the monitoring instruction; According to the target data, the device type and collection frequency required for acquiring the target data; Based on the device type and the collection frequency, control the target device to collect data according to the collection frequency, and receive the collection results sent by the target device in real time; Input the collected results into a preset model, perform association and regularity analysis on the collected results through the model, and output the predicted trend results of the environmental data through the model; Sending to the user terminal based on the predicted trend result and the collected trend result; The step of inputting the collected results into a preset model, performing association and regularity analysis on the collected results through the model, and outputting the predicted trend results of the environmental data through the model includes: Constructing a Bayesian network model, obtaining historical collection results and classifying the historical collection results by indicators, and training the Bayesian network model with the classified historical collection results to obtain a trained Bayesian network model; Inputting the collected results into a Bayesian network model, classifying the collected results by indicators through the trained Bayesian network model, fitting a trend line for the data of each category using a time series, and outputting a trend prediction result through the trained Bayesian network model; Perform seasonal division on the historical collection results, combine seasonal division with trend prediction, calculate the confidence interval of trend prediction, and use it to quantify the uncertainty of prediction; After the step of calculating a confidence interval for the trend forecast to quantify the uncertainty of the forecast, the method comprises: Based on the results of calculating the confidence interval for the trend forecast, determine the confidence interval for each forecast value; For each predicted value, calculate the width of the confidence interval, including expressing it as the difference between the upper and lower bounds; Comparing the width of the confidence interval for each predicted value with the predicted value itself, analyzing the proportion of the confidence interval width, and the proportion is used to determine the degree of uncertainty in the prediction; Analyzing the prediction results according to the proportion of the confidence interval width; When the ratio of the confidence interval of the predicted value is greater than a preset threshold, it is judged that the uncertainty of the prediction result is high, and an adjustment strategy is generated; When the ratio of the confidence interval of the predicted value is greater than a preset threshold, it is judged that the uncertainty of the prediction result is high, and an adjustment strategy is generated. The method includes: According to time and geographical factors, the historical collection results are grouped to obtain multiple data subsets; Performing difference identification on each of the data subsets, and when the data subset presents a long-tail distribution, inputting the data subset with the long-tail distribution into a preset sparse data processing model, and fitting the distribution characteristics of the data subset through the sparse data processing model; The importance of the data subset is adjusted according to the distribution characteristics, including strengthening or weakening the importance.

2. The remote environment monitoring method based on the Internet of Things according to claim 1 is characterized in that: After the step of outputting the prediction trend result of the environmental data through the model, the method includes: When the target data is the monitored temperature, all temperature values ​​collected by the device are obtained, and the temperature change rate is calculated according to all the temperature values; Based on the temperature change rate and the predicted trend, adjusting operating parameters of the temperature regulating device to maintain the outdoor temperature range; When the target data is to monitor air quality, the concentration of pollutants in the air collected by the device is obtained, including the concentration of suspended particulate matter, carbon monoxide, sulfur dioxide, nitrogen dioxide, ozone and volatile organic compounds; All parameters of the pollutant concentration are integrated and the air quality index is calculated; According to the air quality index and the predicted trend, the operating parameters of the air purification equipment are adjusted.

3. The remote environment monitoring method based on the Internet of Things according to claim 1 is characterized in that: After the step of sending the prediction result and the collection trend result to the user end, the method further includes analyzing the data quality of the collection result. The method includes: According to the time series, thresholds of the collection results at different times are set to determine whether the data at different times meet the environmental quality standards; Determine whether the data of the prediction result is discontinuous in the time series, and when the data of the collection result is not discontinuous in the time series, divide the data of the collection result into multiple time periods; Calculate the data of each time period using the interquartile range algorithm, and determine whether the calculation result of the corresponding time period is an outlier; When the calculation result of the corresponding time period is an abnormal value, it is determined that the data quality of the corresponding time period does not meet the standard, and an alarm instruction message is generated and sent to the user end.

4. The remote environment monitoring method based on the Internet of Things according to claim 3 is characterized in that: The data of each time period is calculated by an interquartile range algorithm, and it is determined whether there is an abnormal value in the data of each time period. The method includes: The calculation formula of the interquartile range is: Calculate the interquartile range for each time period: IQR = Q3 - Q1, where Q1 and Q3 are quartiles, and IQR is the interquartile range; Lower anomaly threshold = Q1 - 1.5 * IQR; Upper anomaly threshold = Q3 + 1.5 * IQR; Determine whether the data point in each time period data is within the range of the lower abnormal threshold and the upper abnormal threshold; When the data point is not within the range, it is judged as an abnormal data point; When the data point is within the range, it is judged as a normal data point; The proportion of the abnormal data points to the total data points is analyzed, and when the proportion of the abnormal data points is greater than the proportion threshold of the time period, the data of the time period is determined to be an abnormal value.

5. A remote environment monitoring system based on the Internet of Things, used in the method according to any one of claims 1 to 4, characterized in that: include: Building modules for building Internet platforms for exchanging data with user terminals and device terminals; The analysis module is used to obtain the monitoring instruction of the user terminal in real time, and when the monitoring instruction is received, analyze the target data that needs to be monitored by the monitoring instruction; An acquisition module, used for acquiring the device type and acquisition frequency required for the target data according to the target data; A control module, used to control the target device to collect data according to the collection frequency based on the device type and the collection frequency, and receive the collection results sent by the target device in real time; A prediction module, used to input the collected results into a preset model, perform association and regularity analysis on the collected results through the model, and output the predicted trend results of the environmental data through the model; The sending module is used to send the predicted trend results and the collected trend results to the user end.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.