Emergency management system of urban waterlogging based on cloud computing platform and 3D visualization

Abstract

In recent years, the problem of urban waterlogging has been highly valued. The application of information technology and image simulation to emergency management of urban waterlogging can improve urban flood prevention and disaster reduction capabilities and reduce disaster losses. In this paper, the author analyze the emergency management system of urban waterlogging based on cloud computing platform and 3D visualization. Collect data through street monitoring and drones, re-analyze the collected images, and screen cities for easy waterlogging. Researchers can rely on the high-performance computing power of the system and the visualized integrated environment to achieve online monitoring and early warning of waterlogging and 3D visual display. The system can provide early warning services in the form of alarms for monitoring results that exceed the threshold, and use mobile agents to send messages to relevant personnel in a variety of ways, providing fast auxiliary decision-making services. The simulation results show that the system has high simulation accuracy and can provide fast and efficient emergency services.

Keywords

Urban Waterlogging Neural Network cloud computing emergency management

1 Introduction

Emergency management of urban waterlogging is the focus of the current urban flood control system, flood disasters is an important natural disaster. The continuous improvement of the flood control system and the improvement of the accuracy of flood forecasting have effectively reduced disaster losses [1]. However, with the acceleration of urbanization, the frequency and intensity of urban waterlogging have increased, and the forecast period of rainstorms is short, which makes it difficult to meet emergency management and rapid response needs, which increases the risk of waterlogging disasters, which is currently facing global flood management [2, 3]. New challenges and issues. The problem of urban waterlogging is highly valued, and it is of great significance to improve the ability of urban flood prevention and disaster reduction and reduce disaster losses by applying information technology to emergency management of urban waterlogging [4].

Under the influence of the changing environment, the problem of urban waterlogging is becoming increasingly prominent. Cities are suffering from waterlogging disasters, and residents’ personal and property safety is seriously threatened [5, 6]. The suddenness of urban local heavy rain, the complexity of the formation mechanism of waterlogging, the severity of disaster losses, and the uncertainty of waterlogging highlight the urgency of research on urban waterlogging [7]. In recent years, with the rapid development of next-generation information technologies such as big data, cloud computing, and Internet plus, smart water conservancy and smart cities have been applied in different regions and industries [8]. Aiming at the problem of urban waterlogging in a changing environment, drawing on advanced concepts of smart water conservancy, making full use of modern information technologies such as satellite monitoring and big data analysis to collect and integrate information resources, strengthen multi-source information fusion, promote visualization of business applications, and improve management efficiency and effectiveness, Can provide system solutions for urban flood emergency management [9, 10]. Aiming at the problems related to urban flood prevention and disaster reduction in a changing environment, many researchers have designed and developed computer systems, and carried out simulation and visualization applications [11].

Emergency management and scientific response to urban waterlogging under changing circumstances are key and difficult issues. To this end, this article uses fusion GIS, big data, cloud services and comprehensive integration technologies. Design a three-dimensional visualized emergency management information system for urban waterlogging, and provide emergency management services for the entire process of urban waterlogging from front-end monitoring, process early warning, and event response.

2 Related work

China’s climate is characterized by significant rainy and dry seasons throughout the year. The rainy season is generally concentrated in June– August. Heavy rainstorms are concentrated and distributed. In a short time, deep water will form in urban areas. The south of China is also often affected by typhoons. The typhoon-induced rainfall is characterized by heavy rainfall and short time [12 –14]. The greater the amount of precipitation. The more rainwater that needs to be discharged, the greater the pressure on the rainwater drainage system of the city, and the easier it is to form urban waterlogging. In addition, the unreasonable and excessive development of human beings in the natural world has led to the increasing frequency of extreme weather, which has put great pressure on the drainage systems of cities in China [15].

Urban planning is essentially a mechanism for controlling target deviation in the process of urban development. The core of its technology is to control urban land use and its changes, so as to adjust or establish the order of the urban spatial system and make it in line with the interests of society Value-oriented [16 –18]. Urban planning is only part of the overall social plan. It has its own role, nature, and scope of activities. Therefore, the results of urban development, whether ideal or not, should not be captured by urban planning; that is, urban planning has certain limitations [19, 20]. It is inappropriate to judge the failure of urban planning by the negative problems that appear during the development process.

At present, the main reason for urban waterlogging problems is the incomplete flood prevention and drainage system in cities [21]. There are both a series of problems such as low design standards, backward management methods and inadequate government supervision. From the perspective of urban planning, urban construction generally includes residential land, public facilities land and commercial land, as well as municipal land, green land, water land, etc. Urban planning departments should make overall planning and scientific layout [22, 23]. However, most regions of the country have not been able to predict scientifically and accurately the development and changes of cities, and urban planning and design lacks foresight. Some local officials have carried out image projects in order to highlight “political achievements”, which seriously damaged the city’s ecosystem [24, 25]. The end result is that the original natural water storage places such as lakes and ponds in the city have shrunk severely, and their ability to regulate and store urban rainstorms has decreased significantly.

Screen and process the collected urban waterlogging data resources, integrate urban meteorological and hydrological, physical geography, socioeconomic, and waterlogging disaster data, and establish a multi-source information fusion and application method system that combines points, lines, and areas [26]. Source information fusion and other technologies process dynamic urban waterlogging data resources. Integrate spatial geographic information such as administrative divisions, topography, river systems, riverway dikes, waterproof drainage, engineering measures, and other engineering data, as well as disaster data such as meteorology and hydrology, disaster losses, and waterlogging points for waterlogging events [27, 28]. Multi-sector dynamic data sharing mechanism for municipal and public sectors. At the same time, multi-source data integration middleware is used to screen, transform and integrate the flooded data resources to ensure the consistency and integrity of the data resources, and to improve the granularity, reliability and integration of the data resources [29].

Based on the “Water Conservancy Hadoop Big Data Platform”, Spark+Hive, HDFs, Map Reduce and other big data technologies are used to analyze the waterlogged data resources, and the data resources of the entire process of mining and development of waterlogged events are constructed to build a city waterlogged data resource database [30]. Establish data mapping, use component technology to develop standardized data components related to waterlogging, use web service technology to encapsulate components, and establish data resource indexes according to the universal description, discovery and integration directory service of the data to form data federation-based virtual Access channels to access and manage them [31, 32]. According to the idea of priority of big data efficiency and on-demand services, it aims at different application service topics and obtains data from the data resource library through components. Through multi-source information fusion and correlation, process information description finally provides information services, calculation services and decision-making services for flood emergency management and response in the form of standardized components.

3 The theoretical basis

3.1 Data mining theory

The technology of database is becoming more and more perfect. At the same time, the technology demand of exploring potential information is also higher and higher. Although today’s database management tools, in response to massive data sets and other complex situations, can carry out efficient computing processing, but the mining effect of implicit association between data and data is difficult to achieve. Association rules is a classic mining method in the field of data mining. This method finds valuable association between data items through the processing of original data. According to the constraints of different support or confidence, it searches out potential association rules between data items. In the process of mining, it is necessary to set support threshold and confidence threshold in advance. Such a mining model is often used in business, medical or scientific research fields. Urban waterlogging image monitoring system shown in Fig. 1.

Fig. 1

Urban waterlogging image monitoring system.

For the distributed computing framework, Hadoop has the characteristics of high scalability and open source. Meanwhile, it provides a good research platform for the parallel computing model, and is concerned by many researchers. Hadoop is such a big data distributed file system, which encapsulates the complex structure of the bottom layer. Researchers and users only need to use the good transparent interface directly, which provides a lot of convenience for building distributed applications, and also improves throughput. Through zookeeper management, Hadoop can configure multiple standby primary nodes and backup metadata. In case of downtime and other special circumstances, the system will recover the metadata quickly. The actual data on the sub nodes is also greatly guaranteed, which enhances the fault tolerance purpose of the system platform. Hadoop achieves data communication in the sense of controlling file transfer between each point. If necessary, new sub node computers can be added at any time, which has very good scalability. According to the task assigned by the main node to the sub nodes, each node calculates its own task, realizes the theoretical parallel computing architecture, and then comprehensively counts the data results reported by each sub node. Therefore, Hadoop is suitable for mass data storage and calculation.

With the gradual improvement of Hadoop functions, it has developed into the most mainstream big data ecosystem. From HDFS, yarn, map reduce, which was originally located in the underlying data management mode, to now, more functional modules have been derived. For example, hive can use SQL statement operation to simplify Mr programming model, HBase is about NoSQL distributed column storage database in big data ecology; zookeeper can act as the general service manager in Hadoop cluster, coordinate the working mode of each framework module in Hadoop ecology, and manage each node task.

In terms of structure and technology, spark provides RDD functional programming mode interface to reduce the cost of algorithm writing, that is, method oriented programming, memory based cluster computing, and RDD to achieve parallel operation. Obviously, this effectively improves the efficiency of data processing. From this, we can see that Mr distributed computing mode, function oriented programming and DAG have been used for reference and integrated by spark.the optimization task can be modelled as the follows. $x^{*} \in \cap X_{i} = X_{0}$

Map reduce is widely used in clusters to achieve iterative computing and deal with some large datasets. For work distribution and fault tolerance, the algorithm often uses a group of higher-level operators into the parallel algorithm. However, this has exposed its own shortcomings. For example, when the system performs two Mr Jobs separately, it needs to reuse some data, which is generally stored in the peripheral storage system (here HDFS) to achieve shared data. Obviously, such an implementation mode can no longer meet the needs of complex logical business. Considering the shortcomings of map reduce iterative operation, the time efficiency required in the whole algorithm execution process can reach 90% or more. Spark can use RDD interdependence and broadcast variables to share data, which can monitor variable data in real time and control the data consistency of the whole work. In addition, spark also supports cache and persist to persist RDD operations on multiple nodes according to the clustering algorithm strategy, or to communicate data with each other across multiple work sub nodes. $min f (X) \begin{matrix} s . t . & g_{i} (X) ⩽ 0 \end{matrix}$ $min \sum_{k} w_{k} β_{k} \begin{matrix} s . t . & {\begin{matrix} f (X) ⩽ β_{k} \\ g_{j} (X) ⩽ 0 \end{matrix} \end{matrix}$

In the Hadoop distributed environment, multi computing nodes have significant effect in parallel processing and scalability, and become an important research hotspot in today’s society. However, Hadoop cluster computing architecture also has its own shortcomings. With the increasing demand for big data, the way of data processing is becoming more and more complex. These need to adopt a lot of iterative work processes. Each computing node will produce a lot of intermediate results and save them in HDFS during these processes, resulting in a lot of load-based I / O Operation, in other words, this kind of work efficiency will be much lower. the overall goal to achieve the optimal as shown in the Fig. 2. If the map reduce computing mode of Hadoop distributed platform is adopted, the final result is quite different from the expected efficiency standard. In order to build an efficient distributed computing environment, it is more necessary to choose a suitable distributed computing environment that can work iteratively. This computing environment uses the spark distributed computing framework proposed in the previous part.

Fig. 2

The Demonstration of the Optimization Strategy.

3.2 Improved data mining algorithms

In this paper, the technology of multi-source Information Fusion and data cache is used to build an efficient and expressive integrated visualization environment for the system, we will provide basic information analysis, monitoring, early warning and emergency management services for urban waterlogging. The Tile pyramid technology is used to tile the massive spatial data, and the tile rules are used to store and schedule the data, thus realizing the seamless mosaic of the massive terrain and image data, data cache technology is used to realize the efficient management and scientific scheduling of spatial data and support the Application Service of process visualization. Through standardized data communication protocol and spatial information standard, the database of waterlogging data is connected to the system, and the spatial geographic information and waterlogging data are scientifically organized and managed, so as to improve the ability of data access, it provides an efficient and expressive visual integrated environment to realize the functions of on-line monitoring and early warning, real-time early warning and scene response.

The monitoring and warning information of different spatial and temporal scales is extracted from the urban waterlogging data resource database, and the online monitoring and warning and 3D visualization of waterlogging are realized by the high performance computing power and visual integrated environment. Based on the system, on-line monitoring, calculation, analysis and visualization of Meteorological and hydrological factors such as rainfall, runoff and water level in the study area are carried out, and early warning services are provided in the form of warnings for the monitoring results above the threshold, mobile agent is used to send short message and other ways to relevant people to provide rapid decision-making assistance services.

From the construction of FP Tree to the mining of the most frequent items, it is necessary to calculate and process the type relationship of data items. If the data volume is large and there are many different data items, the time efficiency of the two scan transaction database has been greatly reduced. In order to solve this problem, we first design a more portable and flexible data set conversion mode, introduce the prime number theory, improve the efficiency of operation, and enhance the security.The algorithm based on the SQP theory can be expressed as follows:

$\begin{matrix} f^{(α)} (x 0) = {\frac{df (x)}{{dx}^{α}} |}_{x = x_{0}} \\ = lim_{δ x \to 0} \frac{Δ^{α} (f (x) - f (x_{0}))}{(x - x_{0})^{α}} \end{matrix}$ (1) for 0 < a ⩽ 1 where $Δ^{α} (f (x) - f (x_{0})) ≅ Γ (1 + α) lim_{x \to \infty} Δ (f (x) - f (x_{0}))$ (2)

And local fractional integral of f (x) defined by Equation 3. $\begin{array}{l} _{a} I_{b}^{(α)} f (t) = \frac{1}{Γ (1 + α)} \int_{a}^{b} f (t) {(d t)}^{α} \\ = \frac{1}{Γ (1 + α)} \lim_{Δ t \to 0} \sum_{j = 0}^{j = N - 1} f (t_{j}) {(Δ t_{j})}^{α} \end{array}$ (3)

If f (x) is defined on the real line -∞ < x < ∞, its local fractional Hilbert transform, denoted by $f_{x}^{H, α} (x)$ is defined by $\begin{matrix} H_{α} {f (t)} = {\hat{f}}_{H}^{α} (x) \\ = \frac{1}{Γ (1 + α)} \oint_{R} \frac{f (t)}{(t - x)^{α}} (dt)^{α} \end{matrix}$ (4)

Some iterates (and, indeed, the solution of the problem) may lie at other points on the boundary or interior of the feasible region. that is: $\begin{matrix} \frac{1}{Γ (1 + α)} \oint_{R} \frac{f (t)}{(t - x)^{α}} (dt)^{α} \\ = lim_{ɛ \to 0} [\frac{1}{Γ (1 + α)} \int_{- \infty}^{x - ɛ} \frac{f (t)}{(t - x)^{α}} (dt)^{α} + \\ \frac{1}{Γ (1 + α)} \int_{x + ɛ}^{\infty} \frac{f (t)}{(t - x)^{α}} (dt)^{α}] \end{matrix}$ (5)

To obtain the inverse local fractional Hilbert transform, write again Equation (4) as $\begin{matrix} {\hat{f}}_{H}^{α} (x) = \frac{1}{Γ (1 + α)} \int_{- \infty}^{\infty} \frac{f (t)}{(t - x)^{α}} (dt)^{α} \\ = \frac{1}{Γ (1 + α)} \int_{- \infty}^{\infty} f (t) g (x - t) (dt)^{α} \\ = f (x) * g (x), \end{matrix}$ (6)

The equation of motion is as follows: $\partial_{j} (C_{ijkl} \partial_{k} u_{l} + e_{kij} \partial_{k} φ) - ρ {\ddot{u}}_{i} = 0$ (7)

Under the linear theory, that is: $\partial_{j} (e_{ijkl} \partial_{k} u_{l} - η_{kij} \partial_{k} φ) = 0$ (8)

The linear equation can be expressed into the following simplified forms: $\begin{matrix} L (\nabla, ω) f (x, ω) = 0, \\ L (\nabla, ω) = T (\nabla) + ω^{2} ρ J \end{matrix}$ (9)

Consider the Lagrangian expression for the equality constrained problem depicted in (9), In which, $\begin{matrix} T (\nabla) = ∥ \begin{matrix} T_{ik} (\nabla) & t_{i} (\nabla) \\ t_{k}^{T} (\nabla) & - τ (\nabla) \end{matrix} ∥, \\ J = ∥ \begin{matrix} δ_{ik} & 0 \\ 0 & 0 \end{matrix} ∥, f (x, ω) = ∥ \begin{matrix} u_{k} (x, ω) \\ φ (x, ω) \end{matrix} ∥ \end{matrix}$ (10)

Consider delay, the L can be expressed as: $L^{0} = ∥ \begin{matrix} C_{ijkl}^{0} & e_{kij}^{0} \\ e_{ikl}^{0 T} & - η_{ik}^{0} \end{matrix} ∥$ (11)

These functions can be expressed in the following form: $\begin{matrix} C (x) = C^{0} + C^{1} (x), e (x) = e^{0} + e^{1} (x), \\ η (x) = η^{0} + η^{1} (x), ρ (x) = ρ_{0} + ρ_{1} (x) . \end{matrix}$ (12)

The value with superscript of 1 represents the difference below: $\begin{matrix} C^{1} = C - C^{0}, e^{1} = e - e^{0}, η^{1} = η - η^{0}, \\ ρ_{1} = ρ - ρ_{0} . \end{matrix}$ (13)

The whole function can be simplified into the following integral equation set:

$\begin{matrix} f (x, ω) = & f^{0} (x, ω) + \int_{V} S (x - x^{'}) (L^{1} F (y^{'}) \\ + ρ_{1} ω^{2} g (R {) T}_{1} f (y^{'})] S (y^{'}) {dy}^{'} \end{matrix}$ (14)

In addition, we can introduce the abbreviated formula: $\begin{matrix} g (x, ω) = ∥ \begin{matrix} G_{ik} (x, ω) & γ_{i} (x, ω) \\ γ_{k} (x, ω) & g (x, ω) \end{matrix} ∥, \\ s (x, ω) = ∥ \begin{matrix} G_{ik, l} (x, ω) & γ_{i, k} (x, ω) \\ γ_{k, l} (x, ω) & g_{, k} (x, ω) \end{matrix} ∥, \\ L^{1} (x, ω) = ∥ \begin{matrix} C_{ijkl}^{1} & e_{kij}^{1} \\ e_{kij}^{1 T} & - η_{ik}^{1} \end{matrix} ∥, \\ F (x, ω) = ∥ \begin{matrix} u_{(i, j)} (x, ω) \\ φ_{, i} (x, ω) \end{matrix} ∥ . \end{matrix}$ (15)

Although there are some researches on parallel computing and the optimization of FP growth algorithm’s cluster architecture idea, when dealing with the situation of large amount of data, it is often because of the large scale of building FP Tree, the long time of traversing head table and recursive pattern base, and the unequal task grouping that affects the mining efficiency and final results. Based on the spark distributed computing platform, the optimization idea of association mining is designed, the grouping strategy is improved, and the storage resource occupancy ratio is compressed to carry out efficient mining. Different from parallel computing mining, each sub task is independent of each other, and the mining results do not affect each other, effectively reducing the amount of shuffle of remote access. Spark is determined to be a distributed computing platform because it is more suitable for frequent iterative main memory computing than Hadoop

Although the classical PFP (parallel association mining algorithm) shows a very good association mining effect in the cluster environment, there are some shortcomings: relying on the head table, it produces frequent Shuffle, together with traversal search, takes a lot of time to build a local frequent pattern tree; and when the size of the data mined is large, there are a large number of parallel subtasks with unbalanced computing groups.Waterlogging control system is shown in Fig. 3.

Fig. 3

Waterlogging control system.

3.3 Web geographic information system

The working principle of GIS is to abstract the real world into a combination of layers linked with different characteristics, which is used to solve various complex geographic problems. According to the characteristics of the real object, it can be divided into three typical layers: point, line and surface. For each feature level, we can further use vector and grid format to store geographic data, through which we can carry out relevant operations and analysis.

The representation of the road network is based on its spatial geometry data, which is the location of the solid geometry information, such as the coordinates of C/ nodes and the spatial position of the road segments in the road network. The relationship between spatial objects in the network is embodied by their smart Park relationship, which mainly includes the smart node arc segment and arc segment node relationship. The description of network related information is completed by the attribute table corresponding to the node and the road segment. With the maturity of GIS for road network data weaving and storage, the efficiency of road network analysis can be improved by using GIS technology to model road network chain. GIS deals with spatial problems by abstracting spatial information into three basic elements: point, line and surface. Complex elements are composed of these h basic elements. In the aspect of data organization, G technology manages and stores the spatial information in layers, and realizes the structure visualization of road network data in a layered way, and effectively manages the attribute information of spatial objects. Because of the different ways of recording spatial data and attribute data, different GIS softwares adopt different ways of storage. Some of them use the way of separating sister organization to store spatial data and attribute data. They use GIS database to store spatial data, DBMS to store attribute data, and usually use the primary key of database table to contact spatial data and attribute. Geographic information system as shown in Fig. 4.

Fig. 4

Geographic information system.

4 Urban waterlogging system and optimization

4.1 Rainwater system

With the change of global climate, extreme weather is increasing. The frequent occurrence of rainstorm once in ten years or even once in a hundred years is far beyond the design return period of rainwater pipe network in China. The excessive rainwater stays in the city in the form of surface runoff, which is an important cause of urban rainwater flood disaster. In addition, the first is that the high-density and high-intensity development of the city often exceeds the maximum bearing capacity of the originally planned municipal drainage facilities; the second is the lack of green space and the increase of hard paving, which increases the rainwater runoff invisibly; the third is that the drainage facilities are aging with the increase of service life, and the main network of urban drainage pipes is laid in the city underground, which is difficult to manage and manage after completion Maintenance, easy to lead to problems such as congestion. These factors further affect the drainage capacity of rainwater pipe network and aggravate the problem of urban rainwater and flood. The reconstruction of rainwater pipe network is very difficult and expensive due to the complexity of the underground pipe network caused by many times of urban construction, and even the lack of data in some areas; moreover, it is also very wasteful to improve the drainage capacity of the pipe network only to cope with extreme weather.

Through the reasonable construction of GIS network to achieve the ecological restoration, protection and function improvement, and through the ecological technology, to alleviate the problem of urban rain and flood in urban areas. The infrastructure layer of cloud big data platform as show in Fig. 5.

Fig. 5

The infrastructure layer of cloud big data platform.

4.2 Inundation emergency plan model

Adopt the idea of comprehensive integration, apply knowledge management to the urban flood emergency response plan management process, establish a scenario-based urban flood emergency response plan model, and systematically organize, integrate, share, The application uses graphs to correlate knowledge, and through discussion and learning, knowledge integration, dynamic adjustment and continuous improvement of emergency response to knowledge can be achieved, and the capacity of urban flood emergency management can be improved. Using scenario analysis and comprehensive integration methods, according to the characteristics of urban waterlogging, an emergency plan model for urban waterlogging is proposed. The model structure includes: 1. Digital processing of the plan On the basis of the standardized plan template prepared by the state, the theme of urban waterlogging is collected. Historical data related to it, real-time monitoring data and results data are collected, and the data is preprocessed to form the initial data of the urban waterlogging plan. Digital plan. 2. Component development plan. All data, information, models, and methods of constructing plans are considered as components, and component plans are developed to make the digital plans form a loosely coupled structure, and the developed components are stored in the component library for the development of similar plans. 3. Reconstruction of scenarios. According to the characteristics of urban waterlogging, select similar scenarios from the historical scenario database for reconstruction, reconstruct the response plan that is closest to the current event in the shortest time, and use component visualization to develop the component visualization to form different waterlogging scenario combinations. Scenario plan. 4. Scenario evaluation. According to the characteristics of flooding, the historical scenario response strategy is incorporated into the scenario plan. The scenario plan is comprehensively integrated and discussed. From data, information to knowledge and decision-making, qualitative discussion and quantitative analysis are combined, and the expert’s thinking and experience are used to carry out the scenario plan Correction and optimization, and cyclical execution until a better and optimal plan is generated, and this is used as an implementation plan for emergency response to urban waterlogging. At the same time, the theme and implementation plan of urban waterlogging feature scenarios are stored in the scenario set and plan library respectively, providing basic scenario and plan resources for rapid reconstruction of similar scenarios of urban waterlogging and rapid organization of emergency plans.

4.3 Emergency plan handling

Structuring, informatizing, and intelligentizing the urban waterlogging emergency plan, realizing the operability, visualization, and quantification of the emergency plan, improving the efficiency and effectiveness of urban waterlogging emergency plan management. The processing flow includes: 1, structured processing of text plans. Structured processing of information, classification, organization, responsibilities, monitoring and early warning, emergency response, resources and other data resources of urban waterlogging events, using component methods to develop relatively independent and related program modules, and packaging through Web Services Reusable. 2. Analyze the process of emergency planning for urban waterlogging and draw up a flowchart of the emergency plan. According to the characteristics of the event and the evolution process, the digital plan is described in a process. Information reception, information transmission, plan start, early warning level determination, response level determination, etc. are regarded as key nodes. The flow of information between different nodes is linked by arrows. Draw an emergency plan flowchart, and use knowledge management methods to develop a plan process module that can be automatically analyzed and dynamically revised. 3. Realize the association of urban flood event information resources with flowcharts. The waterlogging plan structural element module, event basic information, evolution information, emergency resources, forecasting and early warning, monitoring and monitoring information resources are associated with the waterlogging plan flowchart module to form a highly simulated urban waterlogging emergency plan. 4. Develop an emergency plan management platform for urban waterlogging. Comprehensively integrate typical cases and emergency knowledge of similar flood events, develop urban flood emergency response knowledge bases and case bases, formulate inference rules for emergency strategies and disposal plans based on the evolution mechanism of flood events, and intelligentize emergency disposal plans for key nodes during the emergency response Analysis and evaluation.

By comparing the results of as each index selection, the factors of participation were selected rationally, and the weights were determined according to the influence factors. The calculated based form of the function can be expressed as the follows. $Y_{general} = b_{0} b_{1} x_{1} + b_{2} x_{2} + . . . + b_{m} x_{m}$

At present the evaluation factors classification for qualitative methods and a factor of the maximum and minimum values, with a uniform grade of tolerance. In the following table one, we show the parameters under consideration.

5 Simulation analysis

On the basis of dealing with the emergency plan of urban waterlogging, the emergency management process is described according to the emergency process of urban waterlogging. According to the process organization data, information, model and method, a visual logical arrangement is formed to realize the waterlogging emergency response process with the characteristics of component, visualization, loose coupling and so on. Based on the theory and method of knowledge management, the data, information, models and methods related to the emergency response of waterlogging are organized and managed through Knowledge Map Association and relevant information of urban waterlogging management, through the description of the relationship between the subject and the data to form information, using the “focus” box to customize the components, through the “link” Arrow data flow to describe the hierarchical logical relationship between different information sources, combining the historical situation to deal with the pre-plan, in the knowledge layer and in the decision-making process, it forms the response plan under the different situation, according to the early warning information to determine whether to enter the response state, according to the different city waterlogging level to determine the emergency response level. Based on the collaborative working environment of metasynthetic platform, different emergency management subjects customize several scenarios according to the theme of waterlogging in the city, and form scenario scenarios under different scenarios combination according to their own understanding and needs, and the knowledge map is stored in the knowledge map database, and through the study of different scenarios, the Implementation Scheme under the storm waterlogging rank of the specific scenario theme is finally formed, which can be used by decision makers to respond quickly to the specific storm waterlogging rank, urban Flood Control and Disaster Mitigation and prevention programs shall be formulated to minimize flood losses. Sample of urban waterlogging images is shown in Fig. 6.

Fig. 6

Urban waterlogging images.

Prediction and simulation results as show in Fig. 7. Training simulation curve as show in Fig. 8. The results show that the system can effectively simulate and predict urban flood nodes.

Fig. 7

Prediction and simulation results.

Fig. 8

Training simulation curve.

The hazards of urban waterlogging include river overflow, large-scale water accumulation, traffic paralysis, collapse of dangerous old buildings, water inflow of underground facilities and secondary disasters such as power and water cut-off. According to the depth of the water can be divided into four warning level, as show in Table 2.

Table 1

The parameters taken into consideration by the system

Class No.	Comfort level	Sub level	Comfort unit
–	S1	–	–
–	–	S2w	S(2k-1)
S	S2	S2w	–
–	–	S2v	S(2k-1)
–	S3	–	–
N	–	–	–

Table 2

Warning levels for urban waterlogging

Warning level	Rainfall
Rainstorm blue warning sign	Within 12 hours, the rainfall will reach 50 mm or more, or it may reach 50 mm and the rainfall may continue.
Rainstorm yellow warning signal	In 6 hours, the rainfall will reach 50 mm or more, or it may reach 50 mm and the rainfall may continue.
Rainstorm orange warning sign	Within 3 hours, the rainfall will reach 50 mm or more, or it may reach 50 mm and the rainfall may continue.
Rainstorm red warning sign	Within 3 hours, the rainfall will reach 100 mm or more, or the rainfall may continue.

When urban waterlogging occurs, the municipal party committee and government coordinate various departments to establish an emergency management command center, and the emergency management command center is responsible for issuing orders to various departments, so that when the waterlogging occurs, it is prepared, busy, and uncluttered. Urgent and orderly, there are rules to follow and comprehensive consideration in disaster reconstruction.

6 Countermeasures for emergency management of urban waterlogging disaster

6.1 Adopting correct principles and standards to guide urban construction

Urban construction must be considered from all dimensions and perspectives. “The foundation is not strong, the earth is shaking.” Only when the underground infrastructure is secure can the prosperity and development of the ground be guaranteed. For the prosperity and harmonious development of a city, a scientific and reasonable urban planning must be formulated. The most effective way is to incorporate the construction of the city’s underground pipeline network into the performance evaluation of local officials. Encourage them to use the correct principles and standards for urban construction. Form a long-term, stable and effective management mechanism. The development of the city cannot rely on the flashy appearance. The construction of urban underground infrastructure is more important, only the above-ground and underground parts of the city develop simultaneously. It is development in the real sense.

6.2 Raising awareness of disaster prevention

Cities should strengthen publicity on knowledge about flood prevention and disaster reduction. News media and government propaganda departments should explain to the public the relevant knowledge about disaster avoidance and survival through existing propaganda channels. And form a long-term effective city management mechanism. Developed countries in Europe and the United States attach great importance to disaster prevention in the whole society. Disaster prevention and rescue knowledge has been popularized in elementary and middle schools, and China should learn from its reasonable management experience. Popularize disaster prevention and mitigation knowledge as soon as possible in the whole society. At present, the majority of cities in China have adopted the combined pipe system. Industrial sewage has been eroding the urban underground pipe network for a long time, making the drainage of drainage pipes more difficult. To this end, we can learn from Japan. Implement garbage classification management. To ensure the normal operation of underground drainage facilities. Simultaneously. The state should strengthen its legislative work on pollution discharge and increase the cost of illegal sewage discharge by enterprises. Currently. Because the cost of illegal business is low. Some companies start from their own interests without considering the impact on the urban environment. Discharge of industrial waste water and industrial waste residues through urban rain drain pipes without any worries, resulting in severe blockage of rain drain pipes. In recent years. The state attaches great importance to the pilot use of paid use and trading of emission rights. Provinces and cities across the country have successively established systems for the paid use of pollution rights. Accelerate the progress of emissions trading. The introduction of this policy can effectively control the emission of pollutants. therefore. Relevant management departments need to increase punishment for illegal sewage discharge on this basis, and not only have to collect huge fines. The legal responsibility of the legal person of the company must be investigated. Urban emergency management departments should also increase their inspections of illegal sewage discharge in cities. Form a long-term effective monitoring mechanism.

6.3 Strengthen the construction of various professional rescue teams

The rescue capabilities of various professional rescue teams are directly related to the rescue efficiency in the event of a disaster. The first is to strengthen the construction of medical rescue personnel. Increase the number of medical staff and hospital beds. Strengthen the emergency response capacity of medical personnel for emergencies. Strengthen the construction of medical assistance system. The second is to strengthen the team construction of post-disaster psychological counseling experts. After the flood disaster. Disaster victims will have various psychological problems. Therefore, it is necessary to strengthen the psychological assistance work after the disaster. Psychologists need more than a wealth of psychological knowledge. Must also have strong psychological qualities and good communication skills. Only the experts themselves can adapt to the special environment of the disaster area. Adjust your own mentality. In order to do a good job of psychological adjustment of the victims. Simultaneously. Experts should also have good communication skills. Can truly open the heart of the victims. Eliminate the troubles of the victims. The third is the training of managers in crowded places. All personnel engaged in management work shall be adequately trained. Make sure they have the appropriate ability to perform work tasks. When a waterlogging disaster occurs. Poorly managed crowded places tend to exacerbate the impact of disasters. This requires managers to be vigilant at all times and be able to detect hidden dangers early. When a flood disaster occurs, it can respond quickly and handle it properly. Through training. Let managers fully recognize the serious consequences that may arise if the situation worsens and the importance of crowd mobilization; let managers know clearly the work tasks, roles and Responsibilities; ensure that management personnel have a more accurate judgment of potential safety issues, and try to avoid derivative disasters caused by disasters.

7 Conclusion

Affected by the dual effects of climate change and the acceleration of urbanization, urban waterlogging incidents have occurred frequently in recent years, with severe damage, and have been highly valued by the state and experts and scholars. Especially since 2014, the state has invested a lot of funds to build a “sponge” nationwide “Cities”, from the perspective of engineering measures to provide ideas for urban waterlogging response. However, the construction of the “sponge city” did not happen overnight. In recent years, urban rainstorm and waterlogging incidents are still frequent, disaster losses are still serious, and the situation is not optimistic. From the perspective of non-engineering measures, this paper applies modern information technology such as big data and cloud services to urban flood emergency management and response by constructing a three-dimensional visual emergency management system for urban flood. Through information analysis, monitoring and early warning, and response plan modules, Provide visible, fast and efficient emergency response services to reduce the loss of waterlogging disasters, and provide a reference for scientifically responding to urban waterlogging and disaster reduction in a changing environment.

Footnotes

Acknowledgments

This paper was supported by (1) Project funded by China Postdoctoral Science Foundation; (2) Project funded by the Project(017/2018/A)of FDCT; (3) Project funded by the Project of Macao Foundation; (4) Project funded by the Project of Innovation research and application of cross domain multi-dimensional cloud platform technology in Guangdong and Macao cross-border big data.

References

Siponen

, Mahmood

M.A.

and Seppo

, Employees’ adherence to information security policies: An exploratory field study, Information & Management 51 (2014), 217–224.

Mundher

, et al., Digital watermarking for images security using discrete slantlet transform, Applied Mathematics & Information Sciences 8(6) (2014), 2823.

Hove

, et al., Information security incident management: identified practice in large organizations. IT Security Incident Management & IT Forensics (IMF), 2014 Eighth International Conference, (2014), 23–25.

Nassif

, Modeling and optimization of HVAC sys-tems using artificial neural network and genetic algorithm, Building Simulation 7(3) (2014), 237–245.

Huang

, Li

and Xiao

, A proposed iteration optimization approach integrating backpropagation neural network with genetic algorithm, Expert Systems with Applications 42(1) (2015), 146–155.

Shao

, Fu

, Li

, Altan

and Cheng

, Remote sensing monitoring of multi-scale watersheds impermeability for urban hydrological evaluation, Remote Sensing of Environment 232 (2019), 111338.

, Wang

, He

and Shao

, User Identification across Asynchronous Mobility Trajectories, Sensors 19(9) (2019), 2102–2111.

Shao

Zhenfeng

, Wu

Wenjing

, Wang

Zhongyuan

, Du

Wan

, Li

Chengyuan

, SeaShips: A Large-Scale Precisely Anno-tated Dataset for Ship Detection, IEEE Transactions on Multimedia 20(10) (2018), 2593–2604.

Shao

Zhenfeng

, Yang

, Zhou

Weixun

, Performance Evaluation of Single-Label and Multi-Label Remote Sensing Image Retrieval Using a Dense Labeling Dataset, Remote Sensing 10(6) (2018), 964–977.

10.

Shao

Zhenfeng

, Cai

Jiajun

, Remote Sensing Image Fusion with Deep Convolutional Neural Network, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 11(5) (2018), 1656–1669.

11.

Zhang

Lei

, Weng

Qihao

, Shao

Zhenfeng

, An evaluation of monthly impervious surface dynamics by fusing Landsat and MODIS time series in the Pearl River Delta, China, from 2000 to 2015, Remote Sensing of Environment 201 (2017), 99–114.

12.

Huyan

, Shao

Zhenfeng

, Fu

Peng

, Cheng

Qimin

, The Dynamic Analysis between Urban Nighttime Economy and Urbanization Using the DMSP/OLS Nighttime Light Data in China from 1992 to 2012, Remote Sensing 9(5) (2017), 416–420.

13.

Wang

Zhongyuan

, Hu

Ruimin

, Jiang

Junjun

, Han

Zhen

, Shao

Zhenfeng

, Heteroskedasticity tuned mixed-norm sparse regularization for face hallucination, Multimedia Tools and Applications 75(24) (2016), 17273–17301.

14.

Ghaedi

, Zeinali

, Ghaedi

A.M.

, et al., Artificial neural network-genetic algorithm based optimization for the adsorption of methylene blue and brilliant green from aqueous solution by graphite oxide nanoparticle, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 125 (2014), 264–277.

15.

Song

, Wang

and Lei

, Evidence combination based on the degree of credibility and falsity, Journal on Communications, 36(5) (2015), 1041–1046.

16.

Kim

, Roh

and Ha

, Expert system based on the arrangement evaluation model for the arrangement design of a submarine, Expert Systems with Applications 42(22) (2015), 8731–8744.

17.

Ding

Lin

, Shao

Zhenfeng

, Zhang

Hanchao

, Xu

Cong

, Wu

Dewen

, A Comprehensive Evaluation of Urban Sustainable Development in China Based on the TOPSIS-Entropy Method, Sustainability 8(8) (2016), 746–758.

18.

Yin

, Cheng

Qimin

, Wang

Zhenxin

, Shao

Zhenfeng

, ‘Big data’ for pedestrian volume: Exploring the use of Google Street View images for pedestrian counts, Applied Geography 63 (2015), 337–345.

19.

Niu

Ruiqing

, Zhang

Liangpei

, Shao

Zhenfeng

, Cheng

Qimin

, Web-based geological hazard monitoring in the three gorges area of China, Photogrammetric Engineering and Remote Sensing 73(6) (2007), 709–719.

20.

Deren

, Yao

Yuan

, Shao

Zhenfeng

, Wang

, From digital Earth to smart Earth, Chinese Science Bulletin 59(8) (2014), 722–733.

21.

Shao

Zhenfeng

, Li

Deren

, Image City sharing platform and its typical applications, Science China-Information Sciences 54(8) (2011), 1738–1746.

22.

Deren

, Shao

Zhenfeng

, The new era for geo-information, Science in China Series F-Information Sci-ences 52(7) (2009), 1233–1242.

23.

Shidpour

, Cunha

and Bernard

, Group multi-criteria design concept evaluation using combined rough set theory and fuzzy set theory, Expert Systems with Applications 64 (2016), 633–644.

24.

Vinodh

, Kamala

and Jayakrishna

, Application of fuzzy axiomatic design methodology for selection of design alternatives, Journal of Engineering, Design and Technology 13(1) (2015), 2–22.

25.

Kannan

S.R.

, et al., Robust fuzzy clustering algo-rithms in analyzing high-dimensional cancer databases, Applied Soft Computing 35 (2015), 199–213.

26.

Pourjabbar

, et al., Fuzzy hierarchical cross-clustering of data from abandoned mine site contaminated with heavy metals, Computers & Geosciences 72 (2014), 122–133.

27.

George

and Binu

, DRL-Prefixspan: A novel pattern growth algorithm for discovering downturn, revision and launch (DRL) sequential patterns, Open Computer Science 2(4) (2012), 426–439.

28.

Sadeghzadeh

, Ehyaei

M.A.

and Rosen

M.A.

, Techno-economic optimization of a shell and tube heat ex-changer by genetic and particle swarm algorithms, Energy Conversion and Management 93 (2015), 84–91.

29.

El-Emam

N.N.

, New data-hiding algorithm based on adaptive neural networks with modified particle swarm optimization, Computers & Security 55 (2015), 21–45.

30.

Martin

and Choi

, Nonlinear Electrical Impedance Tomography reconstruction using Artificial Neural Networks and ParticleSwarm Optimization, IEEE Transactions on Magnetics (2015), 1–1.

31.

Zhao

, Jin

, Huang

, et al., An objective predic-tion model for typhoon rainstorm using particle swarm optimization: neural network ensemble, Natural Hazards 73(2) (2014), 427–437.

32.

Xiangkui

, Yongqing

and Wan

, BP Neural Network Camera Calibration Based on Particle Swarm Optimization Genetic Algorithm, Journal of Frontiers of Computer Science & Technology 8(10) (2014), 1254–1262.