These four key technologies are necessary to realize the unified management of urban governance through one network

Reading guide ： To realize the unified management of urban governance , You must have the following four key technologies ： One network perception of urban state 、 City data is shared by one network 、 Three screen linkage of information flow 、 Virtual real mapping digital twins .

author ： Zheng Yu     source ： big data DT（ID：hzdashuju）

01 One network perception of urban state

A network of city perception is the source of mastering the five senses of city state and generating city data , For cities Six kinds of perceptual content , utilize Four modes of perception and Three data acquisition methods , Realize the global perception of the city 、 Precise control 、 Rational layout .

1. Six kinds of perceptual content

According to the practice of a large number of smart city applications , The following six categories are the most common , And most importantly . Pictured 4.1 Shown , These six categories include ： Visitors flowrate 、 Traffic flow 、 Environmental Science 、 energy 、 Economy and public safety , Each category is further subdivided into several subcategories , For example, the environment includes meteorology 、 Air quality 、 soil 、 Water quality, etc , Each subclass contains several indicators , For example, air quality covers PM2.5、PM10 Concentration of .

▲ chart 4.1 Six categories of urban perception

Only first define these six kinds of perceptual content , In order to further standardize the perception of each type of content , Including how to select the corresponding sensing device standard 、 data format 、 sampling frequency 、 Access form and layout principle ; In order to coordinate and connect various stocks 、 Incremental awareness infrastructure ; To accurately grasp the status of these contents , Truly realize one network perception of urban state .

2. Four modes of perception

Pictured 4.2 Shown , The patterns of city perception can be divided into two categories: sensor centered and human centered . among , The sensor centered sensing mode can be further subdivided into Fixed perception and Mobile perception ; Human centered perception is further subdivided into Active and passive group perception . therefore , There are four perception modes .

▲ chart 4.2 Four modes of city perception

• People centered active group perception ： If grid members find problems in the community, they should report them in time 、 Residents pass 12345 Ask about something 、 Citizens pass “ Readily take ” And other applications to find hidden dangers in the city , All of these are actively helping to perceive the state of the city , It is called people-centered active group perception .

• People centered passive group perception ： There is another mode , It is also human centered perception , But people who participate in the perception mode do not know the existence of the perception task , The original intention of generating these perceptual data is not to complete this task , It is called people-centered passive group perception .

3. Three data acquisition methods

When four perception modes are used to complete the perception of the state of the city , The generated perceptual data will pass through as shown in the figure 4.3 The three acquisition methods shown in are imported into the data base ：

▲ chart 4.3 Three data acquisition methods

• Timely push mode ： When the subsystem completes the perception task , Actively push sensing data to the digital base at the first time . People centered active sensing mode usually adopts this kind of data acquisition .

• Timed pull mode ： Perception data is continuously imported into each subsystem for storage , Wait for the upper application to use , Pull relevant perceptual data from each subsystem in batches through the digital base . This kind of data acquisition method is often used in people-centered passive perception mode .

• Geographical convergence ： In this way, a single sensor near the space location is locally networked , Gather the readings locally first , Then pass through a certain channel in time 、 Push to the digital base uniformly , Each sensor is not directly connected to the digital base , Reduce transmission costs , But the data transmission is still timely 、 active . This data acquisition mode is often used for sensor centered fixed sensing .

02 City data is shared by one network

One network sharing of urban data is the digital cornerstone of building an intelligent city and the digital intelligence base of innovative applications , Facing the structure in the city 、 Unstructured and spatiotemporal data , Realize real-time data aggregation 、 Efficient management 、 Deep integration 、 Intelligent analysis and cross domain learning .

1. Categories of city data

• According to the structure of the data, the data can be divided into Structured data 、 Unstructured data 、 Spatiotemporal data .

• City data can be divided into Government data 、 Third party enterprise data and Internet public data .

2. Data access

The data of a city covers all walks of life 、 Produced in different systems 、 From different channels 、 Have different data structures , There is a new system , There are also a large number of systems left over from history , Different systems are usually developed by different companies using different technologies in different years .

therefore , How to connect the complex data in a city from each isolated system to the data base , It has become the first problem to be solved to realize the one network sharing of urban data . Here we should consider solving Access costs 、 transmission efficiency 、 Security Three challenges .

3. Data management

Due to different types of data , Used in different ways , therefore , Different management methods should also be adopted for three different types of data .

1） Structured data

For structured data represented by government data , A collection library needs to be established , Then establish a topic library according to the application field , Finally, a special database is established for a special application .

2） Spatiotemporal data

Establish six data models for spatiotemporal data , Store thousands of data ; Combine Spatiotemporal Indexing Technology with distributed computing technology , Use less computing resources to provide faster query efficiency ; Provide a variety of spatio-temporal query methods , Meet the rigid need to aggregate data according to space and time in the process of urban governance .

▲ chart 4.4 Six spatiotemporal data models

3） Unstructured data

Video 、 Images 、 Unstructured data represented by voice and text , Later, most of them need to be analyzed and processed into structured data before they can be used and queried . For image data , First extract colors from unstructured files 、 shape 、 texture 、 Spatial relations and other characteristics . According to the nature of the business , The object features concerned by specific applications can also be extracted from the image , Like face 、 human body 、 License plate 、 Structured information such as vehicles , Support the use of upper algorithms and functional modules .

4. Intelligent analysis of data

In the one network sharing of urban data , In addition to using data to provide good services , Deep use of the knowledge behind the data to solve problems will bring greater value . In intelligent analysis of data, we should focus on multi-source data fusion 、 space-time AI And building block components .

5. Cross domain learning of data

In view of user privacy 、 Data security, laws and regulations and other factors , Data in a city is impossible 、 There is no need to physically converge to the same digital base .

A new generation of data sharing technology , Support the retention of raw data in various departments and enterprises , Install components such as federal digital gateway in each department , Through federal learning and privacy Computing , Realize data without going out , Algorithms run more , That is, it integrates the knowledge of different data sources 、 Created value , It also avoids the disclosure of original data . This new way of data sharing is also known as cross domain learning .

03 Three screen linkage of information flow

Three screen linkage is the flow of information between different positions , It is also the guarantee of consistent action between different levels . Big screen view situation 、 Medium screen tube disposal 、 Small screen for execution , Let the business make efficient decisions 、 Efficient execution of instructions 、 Event efficient closed loop .

1. Big screen view situation

The large screen is generally configured in the Municipal Urban Operation Center 、 A hall like the District Command Center , Through the big screen , Leaders and staff can observe the overall situation of urban operation at ordinary times 、 Command and execution of emergency events in wartime .

2. Medium screen tube disposal

Most of the events are through the medium screen （ That is, the desktop computer screen of the staff ） To complete the allocation 、 Management 、 Dispatch and report .

3. Small screen for execution

The small screen is the handheld terminal device of front-line staff , Such as mobile phones . The small screen is the entrance for grass-roots staff to enter the one network unified management system , Use the small screen to receive and perform assigned tasks , At the same time, it also finds and reports grassroots problems .

04 Virtual real mapping digital twins

Digital twins are the bridge between the physical world and information systems 、 Human machine cooperation interface , Implement physical mapping 、 Dynamic stacking 、 Fusion analysis 、 Four links of interactive feedback .

1. Physical mapping

In the virtual world, a very realistic digital model and image of the physical world are built based on data . for example , There is a bridge in the physical world , There will be detailed digital modeling of this bridge in the virtual world , The associated information is specific to each pier 、 The volume of each brick 、 texture of material 、 Manufacturer and construction date .

Pictured 4.5 Shown , The commonly used digital twin models in the field of intelligent city include three-dimensional model of primitive city 、 3D model of fine texture urban design 、 Three dimensional model of urban landform 、 3D model of underground pipeline, etc .

▲ chart 4.5 City model commonly used in digital twin system

 

2. Dynamic stacking

The dynamic information of the physical world , Such as traffic flow 、 Pedestrian flow 、 meteorological 、 Energy consumption, etc , Superimposed on the digital world model , Achieve a more realistic presentation . This is not a simple display rendering problem , There must be a strong support of the underlying system and algorithm capabilities .

for example , There is a large amount of space-time data in the city , If you want to show any area smoothly 、 Some kind of data at any time , Most of the original big data platforms cannot be realized , Face cannot catch 、 Can't see 、 Use bad challenges .

3. Fusion analysis

In addition to dynamically loading various city data , The digital twin system also needs to use AI And big data model , Go deep into the data 、 Fusion analysis , Produce decisions that guide the actions of the physical world .

for example , According to the real-time traffic data, find the traffic jam 、 Analyze the impact scope and diffusion trend , And suggest the dredging scheme and detour path . Reporting through residents 、 Logistics Express 、 Take out ordering and community housing basic information to dynamically explore hidden dangers of group rental , And timely coordinate with relevant government departments for troubleshooting and disposal 、 Reasonable relief .

4. Interactive feedback

The resolution is applied to the physical world through the digital twin system , Guide people's behavior and program implementation . The resolution can be used for immediate response , It can also be used as long-term feedback for future planning .