Geological Spatiotemporal Data Model For Dynamic Process Simulating And Real-time Expression

Dynamic simulation of human mining process (metal, nonmetal, coal, oil and gas extraction) and natural geological process (ground collapse, landslide, debris flow, etc.) have characteristics of data form multi-source, multidimensional, multi granularity time, and unequal distribution. Moreover, because of the semantic complexity of spatiotemporal changes, the nonlinear and uncertainty of process modeling, and the "real-time, multidimensional, multi-scale" changes, traditional static data model, that the GIS and geological information system used, is difficult to support the digital expression of the dynamic evolution of mining and geological disaster process.With installation and deployment of various types of geoscience monitoring devices, detection instruments, and the rapid development of wireless sensor networks and the Internet of Things technology, the mining work informatization, automation and intelligent pace accelerated. And now under going through a continual developing period that is from "digital exploration" to "smart exploration", from" digital mining" to "smart mining "from" digital disaster "to" smart disaster". Accordingly, the spatiotemporal variation data, that describing the geology, resources, environment and disaster, is growing. There huge data produced by the geological process, mining, environmental change and natural disaster evolution. So, the traditional data models has been difficult to deal with the issue of store and manage these data, and cannot meet the need of digital expression of variation data of related attributes and model or so on. These models cannot support the semantic expression of the "multi-sensor access", the "multi-granularity changes" and the multi-layered composite spatiotemporal process.According to the general phenomena and processes of the mining and the geological disaster formation, combining with the actual demand of mining and real-time monitoring, we study the digital expression of objects with their variable conditions, properties, geometric spatial relationship, semantic and behavioral models in the geological space. Based on the real time GIS data model, we propose a multi-level (contains spatiotemporal process, geometry, scale, semantic) geological spatiotemporal data model. Its core is a space-time process layer with an event multi factors driven model (EMFDM), and it can be briefly described as follow:In a period of time, the sensors seized geographical or geological object generating event, after the processing of EMFDM (i.e. through constraint rules or simulation algorithms), higher level events can be triggered and then effect to the objects. These objects will response to change their geometric state and property status, so as to promote the space and time information evolution. We use the event (change information) to describe the interactions and the relationships among the various objects. These events also can reflect the change process of object and its state space with time.The model can not only take account of traditional spatial data management, but also support the process visualization management and real-time expression. This model can support multi sensor observation data access, so it is a geological spatiotemporal data model for dynamic process simulation and real-time expression.This paper will be in accordance with three levels (conceptual model, logical model of organization and physical storage model) of the model, to comprehensive study of theory, technology and practice. The main research contents include the following contents:(1) On the basis of analysis of the existing spatiotemporal data model, according to the characteristics of mining and landslide and other geological processes, we present a conceptual spatiotemporal data model which based on the EMFDM. This study takes into account the requirements of dynamic mining process and landslide monitoring process. This research uses some key technologies including the massive spatial geological data storage and management, the geological modeling, and the 3D visualization technology. The research content relates to the advantages, disadvantages and application of the main existing spatiotemporal data models, and the basic theory of models, frameworks, methods and key techniques. We also use cases study to preliminary demonstrate the scientific, practical, real-time, openness and compatibility of the model.(2) Under the guidance of the conceptual model, we carry out analysis on the dynamic mining process with various geological properties, geometry, spatial relationship, semantic and behavior model. The key consideration in our study is to meet the semantic expression of multi-level spatiotemporal variation of complex geological process. We designed its logical organization structure with the idea of object oriented development model. According to the functions, we divided its unit modules into 7 parts:project management, geological spatiotemporal objects and elements management, observation and sensor object management, event management, version management, process management and multi factor driven management.(3) In accordance with the logic organization structure of the model, to satisfy the big data management and the actual geological mine production application requirements, we carried out the system application prototype design. And based on the prototype, we complete the system design and module partition. And further we finished the design of physical structure model and related interfaces.(4) Based on an extensible plug-in development system, through the interfaces of geological spatiotemporal data engine (STSDE), the study implemented the management of geological model, geological features and related data (including geometry, color, texture, materials, indexes, attributes and topological structure) by using the distributed NoSQL database. Based on the distributed database MongoDB, we use the deployment of database fragmentation and copy cluster for the data storage, and with the help of a variety of tools such as Web and controls to monitor and manage the data.(5) With the support of drillings, geological sections, tunnels and other data in the coal area of WangJialing and the data in HuangTupo landslide, we built their 3D geological model based on some geological modeling technology. This paper focus on specific geological modeling work-geological sketch three-dimensional projection modeling, and discuss the functions and roles of geological modeling techniques in the construction of geological temporal data model.(6) Based on the models of WangJialing coal and HuangTupo landslides, we study the multi sensors data access, the storage and management of variation information and digital expression of dynamic simulation process, which including:interactive installation of 3D models of sensors, sensor observation data access, real-time visualization of dynamic monitoring information, dynamic simulation processes constructions based on the OPC interface and the processes construction of based on the multi sensors measurement etc.(7) In the framework of engineering database storage and management, we study the mechanism of events management, mechanism of events transmission and mechanism of spatiotemporal change of geological objects. We also explored and implemented the dynamic process simulation that many events are created by observations and then be transmitted to the related objects to trigger the states changes of geometries and attributes -- by using the simulation OPC servers.(8) Based on the theoretical framework of geological spatiotemporal data model and its key technologies, we developed a prototype system. And with the support of the prototype system, we implement the simulation of dynamic mining and landslides processes. And we also adjusted and modified the model. Its successfully application verified the feasibility, the effectiveness of the geological spatial temporal data model.Our study shows the design of "dynamic process and real time expressing" spatiotemporal model is effective, reasonable and feasible. It not only takes account of traditional GIS and geological information system, adopt for the previously stored data in spatial databases, but also supports the data fragmentation and cluster management by using the distributed database. Moreover, we can increase the data servers to meet the geological spatiotemporal big data management.In conclusions, the main innovation points are as follows:(1) This paper presented a geological spatiotemporal data model that can use for dynamic process simulation and real-time expressing. This model not only can deal with multi-source data and multidimensional temporal, but also support time associated and multi granularities. It can reflect the geological spatial distribution characteristics in the multiple levels spatiotemporal process. This model can also support genetic analysis of complex geological process, dynamic simulation and real-time response.(2) Based on the proposed geological spatiotemporal data model, we implemented the human mining (mining), geological disasters (landslides) dynamic monitoring and process simulation on the conditions of multi sensor access. With the use of a NoSQL database (MongoDB), we stored and managed the geological model, geological features and other related data.