Intelligent water affair edge computing system and method based on industrial Internet of Things gateway

Abstract

The invention relates to the technical field of edge computing gateways, in particular to an intelligent water affair edge computing system and method based on an industrial Internet of Things gateway. The system comprises a data acquisition unit, and the data acquisition unit acquires real-time monitoring data from a monitoring module based on an industrial Internet of Things gateway, and performs preliminary processing and standardization in a data processing module; the data transmission unit transmits the real-time monitoring data processed by the data acquisition unit to a data management layer through a 5G network transmission module; according to the method, the influence of short-term fluctuation can be reduced by setting an exponential smoothing method and using exponential smoothing, so that a clearer trend line is obtained, the basic trend of data can be better understood, and misleading caused by short-term abnormal values is avoided; based on smoothing of past data, exponential smoothing may be used to predict future time series values.

Description

Technical Field

[0001] The present invention relates to the field of edge computing gateway technology, and in particular to a smart water affairs edge computing system and method based on an industrial Internet of Things gateway. Background Art

[0002] Smart water edge computing based on industrial IoT gateway is a modern solution that combines IoT technology, edge computing and water management to improve the efficiency, reliability and sustainability of water resource management and sewage treatment. Here are some explanations of key concepts and components:

[0003] Industrial IoT Gateway: The Industrial IoT Gateway is a key device that connects field devices (such as sensors, actuators, and other smart devices) with cloud platforms or enterprise IT systems. It is responsible for collecting data from various sensors, performing preliminary processing, and forwarding the data to the cloud or other back-end systems. In addition, the gateway can also receive commands and send them to field devices for remote control.

[0004] Edge computing: Edge computing refers to data processing and analysis at or near the source of data generation, rather than sending all data to a central server or cloud for processing. This can reduce network latency, bandwidth consumption, and enhance real-time response capabilities while improving data security and privacy protection.

[0005] Smart water: Smart water uses advanced technologies such as the Internet of Things, big data, cloud computing and artificial intelligence to achieve intelligent management of urban water supply, drainage, sewage treatment and other water facilities. It includes real-time monitoring of water quality, water volume, equipment status, as well as automated control, fault warning and optimized scheduling.

[0006] Smart Water Edge Computing Based on Industrial Internet of Things Gateway In the smart water scenario, the Industrial Internet of Things gateway, as an edge device, undertakes the functions of data collection, preprocessing, local analysis and decision-making. For example, the gateway can collect data from water quality sensors, flow meters, pressure sensors and other devices in real time, perform preliminary cleaning and aggregation of the data, and even execute machine learning algorithms on the edge side to identify abnormal situations or predict equipment failures. This architecture allows for rapid local response to emergencies, such as immediately closing a leaking valve or adjusting the operating parameters of a pump station without waiting for instructions from the cloud platform. At the same time, only necessary data that has been processed and screened will be uploaded to the cloud, saving network resources and improving the response speed and security of the overall system. Edge computing in smart water also supports remote maintenance and software updates, making the management of on-site equipment more flexible and efficient. In general, smart water edge computing based on industrial Internet of Things gateways provides an intelligent, efficient and sustainable way of water resource management; however, in the time series data of existing smart water edge computing systems based on industrial Internet of Things gateways, the original data may contain a large amount of random fluctuations and noise, which will increase the impact of short-term fluctuations and reduce the accuracy and effectiveness of data analysis. Moreover, the existing system cannot quickly update the predicted values ​​according to new data, nor can it adaptively adjust the parameters.

[0007] Therefore, a smart water edge computing system and method based on industrial Internet of Things gateway are provided. Summary of the invention

[0008] The purpose of the present invention is to provide a smart water edge computing system and method based on an industrial Internet of Things gateway to solve the problem raised in the above background technology that the operating conditions and load of the power system may change, and improper setting of word and function constants may lead to misoperation or interference with the normal operating system.

[0009] To achieve the above objectives, on the one hand, the present invention aims to provide a smart water edge computing system based on industrial Internet of Things, including:

[0010] A data acquisition unit, which collects real-time monitoring data from the monitoring module based on the industrial Internet of Things gateway and performs preliminary processing and standardization in the data processing module;

[0011] A data transmission unit, wherein the data transmission unit transmits the real-time monitoring data processed by the data acquisition unit to the data management layer through a 5G network transmission module;

[0012] A data analysis unit, wherein the data analysis unit analyzes abnormal data points in the monitoring data based on the abnormality detection model;

[0013] An intelligent control unit, which performs intelligent control processing based on the analysis result of the data analysis unit;

[0014] The data analysis unit performs in-depth analysis based on the real-time monitoring data transmitted by the data transmission unit, and the in-depth analysis steps are specifically as follows:

[0015] S4.1. Import the received real-time monitoring data into the database;

[0016] S4.2. Clean the real-time monitoring data;

[0017] S4.3, extract features that are helpful for analysis;

[0018] S4.4. Identify the changing trends of various indicators in the water system over time through exponential smoothing methods;

[0019] S4.5. Use the isolation forest algorithm to detect abnormal data points.

[0020] As a further improvement of the technical solution, the data acquisition unit includes a monitoring module and a data processing module;

[0021] Wherein, the monitoring module is used to collect real-time monitoring data of the water system;

[0022] The data processing module is used to convert the real-time monitoring data into digital signals and convert them into a unified CSV format.

[0023] As a further improvement of the present technical solution, the data transmission unit includes a 5G network transmission module, and the 5G network transmission module is used to transmit the processed real-time monitoring data network to the data management layer.

[0024] As a further improvement of the present technical solution, in S4.3, the features extracted to facilitate analysis are specifically: water quality parameters, flow parameters, pressure data, temperature data, and timestamp.

[0025] As a further improvement of the technical solution, in S4.4, the exponential smoothing method is specifically:

[0026] S t =αX t +(1-α)S t-1 ;

[0027] Among them, S t represents the exponential smoothing value at time point t; α represents the smoothing coefficient, which is between 0 and 1; X t Represents the observed value at the current time point; S t-1 Represents the smoothed value at the previous time point.

[0028] As a further improvement of the technical solution, in S4.5, the isolation forest algorithm is specifically:

[0029] Path Length:

[0030]

[0031] Where x represents a single data sample; E[h(x)] represents the average path length of sample x in the isolation forest; E[h(l)] represents the expected value of the path length of the subtree; c(n) represents the average path length of the full binary tree; l represents the number of child nodes of x;

[0032] Anomaly Score:

[0033]

[0034] Where n represents the number of samples; the larger the value of s(x,n), the more likely x is an outlier.

[0035] As a further improvement of the technical solution, the intelligent control unit includes a data visualization module, a remote control module, and a remote maintenance module;

[0036] The data visualization module is used to present data to operation and maintenance personnel in the form of charts and dashboards;

[0037] The remote control module is used to allow operation and maintenance personnel to realize automatic adjustment in combination with preset control logic;

[0038] The remote maintenance module is used by operation and maintenance personnel to remotely repair or update the system.

[0039] On the other hand, the present invention also provides a smart water edge computing method based on an industrial Internet of Things gateway, which is used in any one of the smart water edge computing systems based on an industrial Internet of Things gateway described above, comprising the following steps:

[0040] S9.1. Collect real-time monitoring data in the monitoring module and perform preliminary processing and standardization in the data processing module;

[0041] S9.2. The processed real-time monitoring data is transmitted to the data management layer through the 5G network transmission module;

[0042] S9.3, importing the real-time monitoring data into the database and analyzing the abnormal data points in the monitoring data through the anomaly detection model;

[0043] S9.4. Based on the analysis results of the data management layer, intelligent operations such as visualization, remote control, and remote maintenance are performed.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention sets an exponential smoothing method. By using exponential smoothing, the impact of these short-term fluctuations can be reduced, thereby obtaining a clearer trend line, which helps to better understand the basic trend of the data without being misled by short-term abnormal values; based on the smoothing of past data, exponential smoothing can be used to predict future time series values; exponential smoothing is often used as a data preprocessing technology. Through smoothing, unnecessary noise can be eliminated, making subsequent analysis more accurate and effective; exponential smoothing can quickly update the predicted value according to the new data while maintaining a certain memory of the historical data. The exponential smoothing method can also realize the adaptive adjustment of parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flowchart of the overall process of the present invention.

[0047] The meaning of each number in the figure is:

[0048] 1. Data acquisition unit;

[0049] 2. Data transmission unit;

[0050] 3. Data analysis unit;

[0051] 4. Intelligent control unit. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] On the one hand, the present invention aims to provide a smart water edge computing system based on industrial Internet of Things, including:

[0054] A data acquisition unit 1, which collects real-time monitoring data from the monitoring module based on the industrial Internet of Things gateway and performs preliminary processing and standardization in the data processing module;

[0055] A data transmission unit 2, wherein the data transmission unit 2 transmits the real-time monitoring data processed by the data acquisition unit 1 to the data management layer through a 5G network transmission module;

[0056] A data analysis unit 3, wherein the data analysis unit 3 analyzes abnormal data points in the monitoring data based on an abnormality detection model;

[0057] The intelligent control unit 4 performs intelligent control processing based on the analysis result of the data analysis unit 3 .

[0058] As a further improvement of the technical solution, the data acquisition unit 1 includes a monitoring module and a data processing module;

[0059] Wherein, the monitoring module is used to collect real-time monitoring data of the water system;

[0060] The data processing module is used to convert the real-time monitoring data into digital signals and convert them into a unified CSV format.

[0061] As a further improvement of the present technical solution, the data transmission unit 2 includes a 5G network transmission module, and the 5G network transmission module is used to transmit the processed real-time monitoring data network to the data management layer.

[0062] As a further improvement of the present technical solution, the data analysis unit 3 performs in-depth analysis based on the real-time monitoring data transmitted by the data transmission unit 2, and the in-depth analysis steps are specifically as follows:

[0063] S4.1. Import the received real-time monitoring data into the database;

[0064] S4.2. Clean the real-time monitoring data;

[0065] S4.3, extract features that are helpful for analysis;

[0066] S4.4. Identify the changing trends of various indicators in the water system over time through exponential smoothing methods;

[0067] S4.5, use the isolation forest algorithm to detect abnormal data points;

[0068] As a further improvement of the present technical solution, in S4.3, the features extracted to facilitate analysis are specifically: water quality parameters, flow parameters, pressure data, temperature data, and timestamp.

[0069] As a further improvement of the technical solution, in S4.4, the exponential smoothing method is specifically:

[0070] S t =αX t +(1-α)S t-1 ;

[0071] Among them, S t represents the exponential smoothing value at time point t; α represents the smoothing coefficient, which is between 0 and 1; X t Represents the observed value at the current time point; S t-1 Represents the smoothed value at the previous time point.

[0072] The above exponential smoothing methods mainly serve the following purposes: Trend smoothing: In time series data, the original data may contain a large amount of random fluctuations and noise. By using exponential smoothing, the impact of these short-term fluctuations can be reduced, resulting in a clearer trend line. This helps us better understand the basic trend of the data without being misled by short-term outliers; Predicting future values: Based on the smoothing of past data, exponential smoothing can be used to predict future time series values. This is because the smoothed sequence reflects the long-term trend of the data, which can serve as the basis for predicting future behavior; Data preprocessing: Exponential smoothing is often used as a data preprocessing technique before more complex data analysis or modeling. Through smoothing, unnecessary noise can be eliminated, making subsequent analysis more accurate and effective; Real-time data analysis: In real-time monitoring and control applications, exponential smoothing can quickly update the predicted value based on new data while maintaining a certain memory of historical data. This is particularly useful for scenarios that require immediate response; Parameter adaptive adjustment: By adjusting the smoothing factor αα, the sensitivity of the model to new information can be adjusted according to actual needs. For example, in an environment where water conditions change rapidly, a larger αα can be set to respond to new information more quickly; whereas in a relatively stable environment, a smaller αα can be set to maintain the stability of the prediction.

[0073] As a further improvement of the technical solution, in S4.5, the isolation forest algorithm is specifically:

[0074] Path Length:

[0075]

[0076] Where x represents a single data sample; E[h(c)] represents the average path length of sample x in the isolation forest; E[h(l)] represents the expected value of the path length of the subtree; c(n) represents the average path length of the full binary tree; l represents the number of child nodes of x;

[0077] The Isolation Forest algorithm is an efficient anomaly detection method that is mainly used to identify anomalies or outliers in a data set. In the iForest algorithm, the calculation of path length plays a core role, and its significance is mainly reflected in the following aspects: Anomaly score calculation: Isolation Forest detects anomalies by constructing a variant of a random forest of multiple trees. Each tree is constructed by randomly selecting features and data subsets. Path length refers to the number of edges required from the root node of the tree to isolate a data point. Anomalies tend to have shorter average path lengths because their characteristics are different from most normal data points and are more easily "isolated". Therefore, the average path length can be used as an anomaly score to identify which data points are more likely to be outliers. Fast detection of anomalies: Compared with other distance-based anomaly detection methods, a significant advantage of Isolation Forest is that it can detect anomalies quickly. This is because the calculation of path length can be performed in parallel during the process of building the tree, and there is no need to calculate the distance matrix between all data points, which greatly reduces the computational complexity. Adaptation to high-dimensional data: The Isolation Forest algorithm remains efficient when processing high-dimensional data. In high-dimensional space, distance-based methods may be affected by the curse of dimensionality, while isolation forests effectively alleviate this problem by randomly selecting features and data subsets to build trees, making it perform well on high-dimensional data sets. Unsupervised learning: Isolation forest is an unsupervised learning algorithm, which means that it does not require labeled data to train the model. This is very important in practical applications because obtaining large amounts of labeled data is often difficult and expensive, and isolation forests can automatically detect anomalies by observing the distribution characteristics of the data itself. Interpretability: By observing which features and data subsets lead to the "premature isolation" of data points, it can provide some clues as to why the anomalies are considered anomalies, increasing the interpretability of the algorithm results;

[0078] Anomaly Score:

[0079]

[0080] Where n represents the number of samples; the larger the value of s(x,n), the more likely x is an outlier.

[0081] The anomaly score in the Isolation Forest algorithm is a key concept that quantifies the degree of abnormality of a data point relative to the rest of the points in the dataset. The high or low anomaly score can help us identify which data points are more likely to be outliers or outliers. Here are a few important meanings of the anomaly score in the Isolation Forest algorithm: Quantifying the degree of abnormality: The anomaly score provides a continuous numerical value for each data point, which reflects the degree of abnormality of the point in the dataset. Generally speaking, the higher the anomaly score, the more likely the data point is an outlier. This is because anomalies tend to be more easily isolated than normal points, and therefore have a shorter average path length in the isolation forest, which in turn leads to a higher anomaly score. Decision threshold: Users can set a threshold to mark data points with anomaly scores above this threshold as anomalies. The choice of this threshold usually depends on business requirements, the characteristics of the dataset, and the specific goals of anomaly detection. For example, in fraud detection, users may want to further review any transactions with anomaly scores above a certain threshold. Feedback for unsupervised learning: In the unsupervised learning framework, Isolation Forest can detect anomalies without prior labeled data. The anomaly score provides users with a quantitative feedback to help them understand potential abnormal patterns in the dataset, even in the absence of explicit labels. Feature importance: Anomaly scores can also reveal which features are most critical for distinguishing normal points from anomalies. In the isolation forest construction process, if certain features are often used to segment data early, this may mean that these features are particularly important for identifying outliers. Dynamic threshold adjustment: In some application scenarios, the distribution of anomaly scores may change over time. Therefore, anomaly scores can be used to dynamically adjust the threshold of anomaly detection to adapt to changes in data distribution. Interpretability of anomaly detection: Anomaly scores provide interpretability of anomaly detection results. By analyzing which data points have obtained high anomaly scores and why they have obtained such scores, we can gain a deep understanding of abnormal patterns in the data;

[0082] As a further improvement of the present technical solution, the intelligent control unit 4 includes a data visualization module, a remote control module, and a remote maintenance module;

[0083] The data visualization module is used to present data to operation and maintenance personnel in the form of charts and dashboards;

[0084] The remote control module is used to allow operation and maintenance personnel to realize automatic adjustment in combination with preset control logic;

[0085] The remote maintenance module is used by operation and maintenance personnel to remotely repair or update the system.

[0086] On the other hand, the present invention also provides a smart water edge computing method based on an industrial Internet of Things gateway, which is used in any one of the smart water edge computing systems based on an industrial Internet of Things gateway described above, comprising the following steps:

[0087] S9.1. Collect real-time monitoring data in the monitoring module and perform preliminary processing and standardization in the data processing module;

[0088] S9.2. The processed real-time monitoring data is transmitted to the data management layer through the 5G network transmission module;

[0089] S9.3, importing the real-time monitoring data into the database and analyzing the abnormal data points in the monitoring data through the anomaly detection model;

[0090] S9.4. Based on the analysis results of the data management layer, intelligent operations such as visualization, remote control, and remote maintenance are performed.

[0091] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A smart water edge computing system based on industrial Internet of Things, characterized by: include: A data acquisition unit (1), wherein the data acquisition unit (1) collects real-time monitoring data from the monitoring module based on the industrial Internet of Things gateway, and performs preliminary processing and standardization in the data processing module; A data transmission unit (2), wherein the data transmission unit (2) transmits the real-time monitoring data processed by the data acquisition unit (1) to the data management layer through a 5G network transmission module; A data analysis unit (3), wherein the data analysis unit (3) analyzes abnormal data points in the monitoring data based on an abnormality detection model; An intelligent control unit (4), wherein the intelligent control unit (4) performs intelligent control processing based on the analysis result of the data analysis unit (3); The data analysis unit (3) performs in-depth analysis based on the real-time monitoring data transmitted by the data transmission unit (2), and the in-depth analysis steps are specifically as follows: S4.

1. Import the received real-time monitoring data into the database; S4.

2. Clean the real-time monitoring data; S4.3, extract features that are helpful for analysis; S4.

4. Identify the changing trends of various indicators in the water system over time through exponential smoothing methods; S4.

5. Use the isolation forest algorithm to detect abnormal data points.

2. According to claim 1, a smart water affairs edge computing system based on industrial Internet of Things gateway is characterized in that: The data acquisition unit (1) comprises a monitoring module and a data processing module; Wherein, the monitoring module is used to collect real-time monitoring data of the water system; The data processing module is used to convert the real-time monitoring data into digital signals and convert them into a unified CSV format.

3. According to the smart water edge computing system based on industrial Internet of Things gateway according to claim 1, it is characterized by: The data transmission unit (2) comprises a 5G network transmission module, and the 5G network transmission module is used to transmit the processed real-time monitoring data to the data management layer through the network.

4. The smart water edge computing system based on industrial Internet of Things gateway according to claim 1 is characterized in that: In S4.3, the features that are useful for analysis are specifically extracted: water quality parameters, flow parameters, pressure data, temperature data, and timestamp.

5. The smart water edge computing system based on industrial Internet of Things according to claim 4 is characterized in that: In S4.4, the exponential smoothing method is specifically: S t =αX t +(1-α)S t-1 ; Among them, S t represents the exponential smoothing value at time point t; α represents the smoothing coefficient, which is between 0 and 1; X t Represents the observed value at the current time point; S t-1 Represents the smoothed value at the previous time point.

6. The smart water edge computing system based on industrial Internet of Things gateway according to claim 4 is characterized by: In S4.5, the isolation forest algorithm is specifically: Path Length: Where x represents a single data sample; E[h(x)] represents the average path length of sample x in the isolation forest; E[h(l)] represents the expected value of the path length of the subtree; c(n) represents the average path length of the full binary tree; l represents the number of child nodes of x; Anomaly Score: Where n represents the number of samples; the larger the value of s(x,n), the more likely x is an outlier.

7. The smart water edge computing system based on industrial Internet of Things gateway according to claim 1 is characterized in that: The intelligent control unit (4) comprises a data visualization module, a remote control module, and a remote maintenance module; The data visualization module is used to present data to operation and maintenance personnel in the form of charts and dashboards; The remote control module is used to allow operation and maintenance personnel to realize automatic adjustment in combination with preset control logic; The remote maintenance module is used by operation and maintenance personnel to remotely repair or update the system.

8. A smart water edge computing method based on an industrial Internet of Things gateway, used in a smart water edge computing system based on an industrial Internet of Things gateway as claimed in any one of claims 1 to 7, characterized in that: The steps include: S9.

1. Collect real-time monitoring data in the monitoring module and perform preliminary processing and standardization in the data processing module; S9.

2. The processed real-time monitoring data is transmitted to the data management layer through the 5G network transmission module; S9.3, importing the real-time monitoring data into the database and analyzing the abnormal data points in the monitoring data through the anomaly detection model; S9.

4. Based on the analysis results of the data management layer, intelligent operations such as visualization, remote control, and remote maintenance are performed.

Images