Shulou(Shulou.com)06/01 Report--

Introduction

"Green water and green mountains is Jinshan and Silver Mountain". With our country strengthening legislation and devoting great efforts to environmental control, we have obviously felt that the atmospheric environment around us is constantly improving, in addition to the city-level monitoring data of the National Meteorological Bureau. What is the micro-environment around us like? Where should the next further environmental protection efforts be located?

In order to track the quality of the microenvironment at the community level, Tencent has launched an experimental project: fine-grained distributed atmospheric monitoring. This system is built entirely on Tencent Cloud, and consists of Tencent Cloud-Cloud Database, Tencent Cloud-Tencent Cloud Picture, Tencent Cloud-Internet of things development platform, TencentOS tiny, Tencent Cloud-API gateway, Tencent Cloud-Cloud function, as well as supporting NUCLEO LoRa development kit, Panteng PMS7003 particulate matter sensor and so on.

Our early goal was to deploy terminals in the target area, and to record and track air quality fluctuations with a monitoring frequency of 15 seconds, and to refine the tracking granularity from large environment monitoring to micro-environment monitoring. real-time monitoring of the impact of smoking and kitchen fumes on air quality.

However, after the prototype system was launched, we accidentally tracked the movement process of the large environmental pollution air mass in Shenzhen only in the case of five seed nodes. Next, it will be divided into two parts, namely, "system introduction and tracking Analysis" and "system Technical Architecture". Detailed technical disassembly instructions for each component of the system will be released one after another.

First, system introduction and tracking analysis 1. System data interface

Take a look at the data interface of the system first.

Contains three pages:

-Real-time monitoring data display with geographic location display

-90 minutes of historical data

-30-day historical data.

two。 Monitoring node distribution

The terminal of the prototype system: there are five seed nodes, all of which are placed on the open-air balcony of the volunteers' homes to directly monitor the living environment data. The specific locations are as follows:

-001: Nanshan District-(Qianhai: southwest)

-002: Nanshan District-(Qianhai: northeast)

-003: Nanshan District-(Houhai: coastal City)

-004: Baoan District-(Baoan: West)

-005: Baoan District-(Baoan: East)

Note: the 005 node chooses the balcony outside the kitchen for the prominent display of local real environment data (smoking, lampblack, etc.), so the short-term fluctuation is relatively large in the follow-up display data.

3. Tracking data and analysis

The movement of short-term polluting air masses has occurred many times. Here, for a typical event analysis, let's take a look at the analysis process of the specific situation. The event time is Friday, April 3, 2020.

(1) system data

Real-time monitoring page

From this screenshot, we can see that the air quality in Baoan and Nanshan Qianhai on the west side is OK, but the reading has increased in Houhai and other places on the east side. Is it an endogenous problem? Or is the data abnormal?

Next, take a look at the 90-minute data.

From the screenshots of two successive times, it is obvious that this is an external particle air mass, and there is an obvious moving process.

-for the 1 and 4 nodes on the west side, the monitoring index rises first and decreases first.

-for the 2, 3 and 5 nodes on the east side, the monitoring index increased and decreased later.

Preliminary judgment: the pollution source is from the outside into Shenzhen, which is a moving process from west to east.

Because we have fewer nodes and do not cover areas outside Shenzhen, we then use the data of China Meteorological Administration and http:// (third-party data, credibility unknown, for reference only) for comparison and exogenous analysis.

(2) Open data

Let's first take a look at the data from the National Weather Administration Shenzhen Nanyou Monitoring Station.

National Weather Service data, updated hourly, one data indicator at a time, according to the indicators, the monitoring data did have a rising process and entered an alarm state at 16:00.

In other words, the rise in indicators is real.

Take another look at the http:// data. (third-party reference data, not responsible for its credibility, credible data, please pay attention to the National Weather Service data)

Although http:// also updates data on an hourly basis, it provides the data display effect of geographical dimensions. It can be seen from the picture that higher monitoring indicators have appeared in Shunde and Zhongshan. Combined with the moving data of pollution sources in our own system, it is preliminarily inferred that Shunde and Zhongshan may have been affected first, and the polluted air mass has further crossed the Pearl River and reached Shenzhen.

But Shunde and Zhongshan are located in the northwest and west of Shenzhen, so why does our systematic monitoring seem to be only the source of western transmission? In order to further understand this problem, let's take a look at the geography and path analysis.

(3) Analysis of pollution sources and paths.

First, let's take a look at the topography of the three places.

It can be found that several mountains are separated by the northwest direct path, and because of the elevation of the Pearl River estuary, it provides a permissible transmission channel.

Let's assume that there are two directions of transmission, and guess the effect of the spread:

A) spread from the northwest

Due to the barrier of many mountains, it is more difficult to get there quickly. On the contrary, it will gather in Zhongshan to form a detention point in the west.

B) spread from the West

Across the Pearl River, from west to east, the passage is smooth, which is theoretically qualified.

In other words, no matter whether the polluted air mass is spread from the northwest or from the West, it will produce the data effect monitored in our system. Our monitoring system is real and effective.

Note: this experimental project is only to verify the technical feasibility, the relevant discussion is only simulated reasoning, does not represent the actual situation.

II. System technical architecture

Next, explain the composition of the system, as well as the principle of architecture.

1. System architecture diagram

The core part of this system is completely implemented by Tencent Cloud-related products, and the system is built together with Tencent Cloud compatible products. The following figure describes the composition of the architecture and the cloud products used.

two。 Terminal module and function

(1) Monitoring node

The monitoring node consists of three parts:

-firmware: software layer, using TencentOS Tiny, you can use encrypted communication to connect with Tencent Cloud Internet of things development platform.

-main control board: hardware layer, using TencentOS Tiny-compatible hardware products, equipped with firmware, reading sensors, and reporting data to Tencent Cloud.

Sensors: hardware layer, where the more common climbing sensors are used to monitor data.

(2) Local gateway

Gateways are divided into two types, which are used for access in different environments. Monitoring nodes access the network through gateways and communicate with Tencent Cloud:

-LoRa gateway: supports the access of LoRa protocol terminals, and a single gateway supports 1km~10km coverage. Because of the long-distance communication capability, low-orbit satellite access can also be used when the gateway cannot be used.

-WiFi gateway: supports the access of WiFi protocol terminals, which is used for access to home environment.

3. Cloud modules and functions

(1) Internet of things access layer

Use Tencent Cloud-Internet of things development platform to achieve encrypted terminal access to Tencent Cloud, with multi-product management, multi-protocol access, data analysis, basic storage, status monitoring and other core capabilities of the Internet of things.

In this project, the complex logic processing is transferred to other cloud products, and the forwarding function of the Internet of things development platform is used to transfer the data to the application access module.

(2) Application access layer

Use Tencent Cloud-API gateway to achieve fast REST service interface, comprehensive PaaS access service, back-end connection with logic layer, and import and export functions of formatting monitoring data and displaying data composition.

(3) logical layer

Use Tencent Cloud-Cloud function to perform real-time and scheduled logical processing. There are three applications:

-Real-time storage: the data forwarded by the API gateway is disassembled, analyzed and entered into the database.

-timing analysis: use your own periodic task capabilities to summarize and analyze the data on an hourly, daily basis.

-Real-time query: provide display data to the display layer based on strict authentication. So that the display data has a better ability to cache and combine with complex data.

(4) Storage layer

Tencent Cloud-Cloud Database is used to record data directly based on the high performance and high stability of the cloud database and the current lack of seed node data.

(5) display layer

Use Tencent Cloud-Tencent Cloud Image, and take advantage of the powerful map, visual effects and multi-interface compatibility of cloud image to quickly build rich display effects.

On the display page, all the map data are drawn in real time. The data effect is displayed according to the specified longitude and latitude through the interface data provided by Tencent Cloud-API gateway and Tencent Cloud-Cloud function. The 2D and 3D provincial and municipal display effects of the cloud image will be applied to the follow-up functions of the project.

III. Conclusion

Due to the large number of project architecture points and limited space, this paper only shows the core effect and architecture overview of the experimental project, and the technical implementation and processing details of each module will be explained one by one in the follow-up articles. There will also be a detailed interpretation of how to apply Tencent Cloud database and Tencent Cloud Image to this project. Please pay attention to the following articles.

Note: this project is an experimental project, and the relevant data and inferences are simple analysis. For accurate monitoring data, please also pay attention to the data of the National Weather Service.

Reference:

[1] Tencent Cloud-Internet of things development platform: https:// ct/iotexplorer https:// entOS-tiny https:// ct/apigateway https:// ct/scf https:// ct/cdb https:// ct/tcv

A brief introduction to the author

Gao Shulei, Senior Ecological Product Manager, Tencent Cloud

Gao Shulei, senior ecological product manager of Tencent Cloud, joined Tencent in 2015. He has served as basic product manager and ecological product manager of Tencent Cloud. He owns 9 national patents and is now responsible for the construction of Tencent ecological joint laboratory and related standards.

Recommended reading

Welcome the rebirth, Weimeng database is on the cloud in a comprehensive way.

Special experience of cloud database

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.