Environmental big data application "budding"

Environmental Big Data Application "Germination" _ Data Analyst Examination

Recently, the "beautiful image" of big data has frequently appeared in various plans and guidance issued by the state, and it has also become a hot topic in environmental protection systems.

It is understood that data resource centers in Beijing, Wuhan and other provinces and cities have already been built. Up to now, nearly two-thirds of the provinces (autonomous regions and municipalities directly under the Central Government) have built or are building environmental data resource centers.

Is the environmental data resource center ready for the tornado of big data? What stage has the application of environmental big data reached?

How does the data "speak"?

Compared with other provinces and cities, Beijing's environmental informatization construction started earlier. The construction of Beijing Environmental Data Center began on 20 12 and was officially put into use on 20 13. At present, the main business data of each business office of Beijing Environmental Protection Bureau has entered the environmental data center.

Chen Haining, director of Beijing Environmental Information Center, said that the construction of data center has greatly promoted environmental protection. For example, in the past, environmental monitoring personnel needed to consult a large number of paper materials and manually record the site situation. Now, as long as the data in the data center is directly retrieved through the monitoring terminal or mobile client, the relevant data of pollution sources can be consulted and recorded, which saves time and effort in environmental monitoring.

But not long after, Chen Haining's confusion also appeared. "The data center has been built, but the help to various business departments is still very limited. How to develop better big data products to serve environmental management has become a bottleneck that data centers need to break through. "

Wuhan, which is also ahead, is also facing similar problems. Zhan Wei, director of Wuhan Environmental Information Center, said that for a long time, Wuhan Environmental Data Center has integrated the data of various business departments. "But now these data are only used for some simple statistics and reports, and the data has not really spoken."

More environmental information centers of provincial environmental protection departments are in a state of continuous construction and improvement. It is understood that the Inner Mongolia Environmental Data Center has been basically completed and has entered the third phase of construction; Yunnan Province is in the second stage of "digital environmental protection", focusing on building resource data centers.

The Information Center of the Ministry of Environmental Protection also began to pay attention to the development of data resources very early, hoping that big data can become the key to solving environmental management problems. Since last year, Xu Fuchun, deputy director of the Information Center of the Ministry of Environmental Protection, has taken the lead in exploring the application of big data, and cooperated with Microsoft (China) Co., Ltd. to develop a big data model for the temporal and spatial distribution of major air pollutants in cities-U-AIR. This cooperation also opened the "first spring" of big data applications.

The rise of big data applications represented by U-Air

At present, the ground air quality monitoring stations in China are set up according to administrative regions, and the number of stations is very limited, which can not reflect the spatial distribution of pollutants, and is easily affected by the environment near the stations, causing audiences in different regions to question the reliability of data.

As far as Beijing is concerned, there are 35 ground air quality monitoring points in Beijing, but they cannot fully reflect the air quality of the whole city. "U-Air can simulate fine-grained urban air quality (1km× 1km) by collecting air quality data from ground monitoring stations, combining with big data such as traffic flow, road structure, distribution of points of interest, meteorological conditions, and people flow laws, and can predict air quality in the next 24 hours with an algorithm based on machine cognition." Xu Fuchun said that this not only makes up for the geographical limitation of ground monitoring and monitors every inch of land in real time, but also solves the problem of how to predict urban air quality without a pollution source emission list and unclear emission boundaries.

"U-Air is a typical case of using big data to solve environmental problems, which has been verified in science and technology." Xu Fuchun told reporters that big data applications need to meet three conditions. The first is that the amount of data should be large enough, the second is that the types of data should be diversified, and the third is that the learning methods and algorithms should be based on machine cognition. In addition to industrial structured data, U-Air also has urban road data, road network structure, meteorological data, geographical location, urban feature distribution and other data, and the background also has the conditions for machine learning and cognitive computing.

Luo Tong, a senior architect of Microsoft (China) Co., Ltd., said that in the future, U-Air will be able to predict the air quality of 1 ~5 hours in advance to help people plan their lives better, such as when and where to jog, when to close the window and when to wear a mask.

In fact, in addition to Microsoft's exploration of big data applications, the environmental protection industry has been eager to try in the face of the big cake of big data.

Zhongke Yutu Xia Tian technology co., ltd

The department has also established its own big data center. According to Yao Xin, president of Zhongke Yutu, the company's big data center includes not only local environmental protection business data, but also industry data related to environmental protection. At present, they have also developed an air quality early warning and forecasting model.

Beijing Liding Chuangruan Technology Co., Ltd. put forward the overall framework of "smog calculation", combining traditional air monitoring data, pollution source data, meteorological data, traffic data, restaurants, schools, gas stations and other information to make early warning and prediction of air quality.

Beijing Siwei Innovation Technology Co., Ltd. launched the environmental protection quick search function by using big data, and realized the integrated query of various businesses. In addition, environmental quick search can also automatically recommend environmental data that users may need according to the analysis of users' preferences. For example, the pollution prevention department needs the enterprise data exceeding the standard every quarter, and the system can push it for users at regular intervals.

The core of big data analysis is the research of algorithms and models. IBM China Research Institute has made great efforts in the application mode of big data. According to Wang Haifeng, senior director of IBM China Research Institute, IBM has established a super model of air quality self-regulation by integrating various types and multi-source models (physical model, learning model and expert model) with Watson cognitive computing technology, which can be used for environmental pollution simulation modeling and weather forecasting and early warning. Wang Haifeng also introduced the unique advantages of big data in monitoring pollution sources. By calculating satellite observation data, pollution source information provided by mobile users and ground pollution source monitoring data, we can identify abnormal data and provide basis for environmental law enforcement.

The development of big data industry provides new opportunities for cooperation between environmental protection departments and enterprises. Ordos met IBM at the seminar on the application of environmental information resources held by the Information Center of the Ministry of Environmental Protection in March.

Ordos big data platform is just the beginning.

The reporter learned that the environmental protection work in Ordos has entered a new stage of information-driven environmental management, and the environmental data resource center has also been completed, and all of them will be arranged in the Ordos Big Data Center in the future.

Dong, director of the Environmental Protection Bureau of Ordos City, said that the Ordos Big Data Center will have municipal, transportation, agriculture, meteorology, energy and other social data. Soon, environmental data will really connect with big data. Developing big data platform and using big data to serve environmental management is an urgent problem to be solved by Ordos Environmental Protection Bureau. Therefore, with the support and guidance of the Information Center of the Ministry of Environmental Protection and the Environmental Protection Department of Inner Mongolia Autonomous Region, Ordos Environmental Protection Bureau cooperated with IBM to build an environmental protection big data platform and put it into use initially.

At present, in addition to the traditional ground observation data, emission list, monitoring data of key pollution sources, meteorological data and topography, the big data platform also collects data resources such as satellite remote sensing data, enterprise production data, mobile sensor data, public opinion and enterprise credit status.

"A major feature of the big data platform is its accurate traceability. When the data of the automatic observation station is abnormal, the platform will automatically trigger traceability analysis, then conduct meteorological analysis and air quality analysis in turn, then start pollution traceability, and then conduct centralized production analysis of enterprises. " Dong told reporters that the traceability big data platform can give quantitative analysis results, thus providing data support for accurate law enforcement and helping to improve law enforcement efficiency.

In the past, the traceability was basically based on manual observation, and it was impossible to estimate the specific contribution, which was very prone to deviation. Etuoke Banner is located at the junction of Wuhai City and Erdos City in Inner Mongolia, adjacent to several industrial parks, and the pollution situation has been unclear. After the calculation of the big data platform, Otog Banner has completely figured out the pollution source. "The traceability function of the big data platform is convenient for our environmental protection department to carry out targeted environmental governance." Dong said to him.

In addition to traceability, the big data platform can also predict and warn the heavily polluted weather, and even predict the day when the heavily polluted weather will appear, thus helping the environmental protection department to control pollution sources in a targeted manner and avoid the occurrence of heavily polluted weather. After accessing online monitoring data and air automatic station data in Ordos Information Center, the reporter saw a dynamic picture of color blocks. Green represents the best air quality, and the darker the color, the worse the air quality.

The big data platform of Ordos Environmental Protection Bureau can be said to be the representative of the application of big data in environmental protection system, and the construction of big data will also become the focus of the next work of the Ministry of Environmental Protection.

The above is what Bian Xiao shared about the "green shoots" of environmental big data applications. For more information, you can pay attention to Global Ivy and share more dry goods.