Big data in bulk solids handling and transportation

Big Data in Bulk Solids

Handling & Transportation

G. Lodewijks, W. Li, Y. Pang, X. Jiang

The 12th _{International Conference on Bulk Materials Storage,}

Contents

1.  Introduction

2.  The Internet of Things

3.  The Technology behind the IoT

4.  Big Data

5.  Application of the IoT in BMHT

1.

In the past decades, the development of the Information Technology (IT) has enabled a continuous growth of entities and devices that are connected to the Internet. According to Evans [3] the term ‘the Internet of Things’ (IoT) was born around 2009 when more devices than people were connected to the web.

The IoT aims to collect and share data from everyday objects via autonomous means and to use this data to enhance the quality of life [4]. In order to be able to access the value of the IoT for the Bulk Material Handling and Transportation industry the following questions will be answered in this paper:

1) What is the Internet of Things (IoT)? 2) What technologies enable the IoT?

3) What is Big Data and how is it related to the IoT?

4) What is the current role of the IoT in the Bulk Material Handling and Transportation industry and what may be expected in the future?

2.

IoT - Definitions

•  In [5], the IoT is defined as “things or objects, which through

addressing schemes interact with each other and cooperate with their neighbors to reach common goals”.

•  In [6] the IoT are “interconnecting physical objects with

computing and communication capabilities across a wide range of services and technologies”.

•  In [7] the IoT is perceived as “Interconnection of sensing and

actuating devices providing the ability to share information across platforms through a unified framework...with Cloud

The ‘Internet Oriented’ Vision

•  Each I0 device uses IP as the connection standard, which removes the

need for costly and complex translation interfaces.

•  Communication protocol software is simplified and non-segregated.

•  Each set of I0-deviced is able to work independently, thus there are no

client/server relations.

•  An I0-device keeps track of its own identity.

•  I0 uses bits which are bigger than the network.

•  Big bits allows data to be represented in the same way, no matter what

medium conveys them.

The ‘Internet Oriented’ vision incorporates the requirement for a standardized communication architecture that should allow the IoT to become widespread.

The ‘Things Oriented’ Vision

•  Anything identifies itself: smart objects (SO’s) are identified with a unique

digital name to establish relationships in that domain.

•  Anything communicates: smart objects form ad hoc networks of

interconnected objects.

•  Anything interacts: smart objects interact through sensing and actuation with

their environment.

The ‘Things Oriented’ vision sees the IoT as a network of smart physical or virtual objects with extended Internet technologies and at the same time with a set of technologies that realizes this. The focus lies on the physical embodiment of the IoT.

Things….

•  Connected devices should be heterogeneous

•  Scalable addressing and information management

•  Data exchange should be ubiquitous through proximity

wireless technologies

•  Energy-optimized solutions, minimizing the energy spent for

communication and computation

•  Devices should be track- and traceable

•  Device should be autonomous, the network should distribute

intelligence

•  Data formats should be standardized

The ‘Semantic Oriented’ Vision

The ‘Semantic Oriented’ vision deals with representing, storing, interconnecting, searching and organizing information

generated by the IoT.

It promotes the use of smart connectivity and context-aware computation features. These features should allow the

technology to ‘disappear’ from the consciousness of the user. Raw sensor data obtained through data acquisition has no real value if there is no understanding of its context and meaning.

The challenge lies in the translation of the collected data into information.

3.

•  Data acquisition

•  Identification and tracking

•  Communication and networking

•  Middleware

•  Data storage and analytics

•  Applications

The IoT holds several disciplines and consists of

multiple technologies.

•  DAQ: the key in IoT applications is the fact that objects or entities

can gather data and provide information. The kind of information depends entirely on the application.

•  T&T: several technologies with means for identifying and tracking

SO’s are researched, improved and used in the IoT. Key technologies are Radio Frequency Identification (RFID) and Wireless Sensor Networks (WSN). Three limiting factors:

•  Expected heterogeneous support of any device

•  The need for equipping a battery to SO’s and their limited or short

battery life

•  Current state-of-the-art electronics dimensions are insufficient for

•  Since the IoT assumes the situation where “anything communicates”,

only bidirectional communication technologies are being used (PAN, LAN, WAN).

•  The IoT needs a software platform in order to perform the required

functions like self-configuration, scalability and interoperability. Middleware is defined as “..a software layer or a set of sub-layers interposed between the technological and the application levels” and it will play a “major role in simplifying the development of new

services and the integration of legacy technologies into new ones”. The following layers can be distinguished:

· Sensing layer – integrated with existing hardware to sense and control

· Networking layer – Basic networking and data transfer support

· Service layer – Creates and manages services for the end-user

•  When more and more devices that are connected to the IoT

start to deliver data, the management of this data has to become more efficient and effective in order to stay

manageable. Policies concerning storage, ownership and expiry of the data will become critical during the growth of interconnectivity. The main challenge lies in the amount of noise in the generated data. The data is not standardized and should be filtered first before it can be categorized.

•  Cloud Computing will be used to store and analyze Big Data.

Big data implies a combination of databases too large and/or too diverse to maintain by regular database management systems.

4.

Big Data the can be characterized by the following four V’s:

• Volume: the massive data volumes that are processed

• Variety: the data is collected from a great variety of sources in

multiple formats

• Velocity: data is acquired, sent and analyzed with high data

transfer rates

• Value: value is found in, first considered, unstructured and

5.

Organize/design

network arrangement

A E D B C F

A E D B C F

PC

Modbus Master

Enterprise or Automation

PC

Modbus Master

Enterprise or Automation

PC

Modbus Master

Enterprise or Automation

Network

To equip the belt conveyor with a system made of:

1. procedures and devices for system set-up and configura7on

(roller installa7on and replacement);

2. roller with an embedded device able to deliver signal in real-7me

outside of the roller through wireless communica7on;

3. a network of data aggregators distributed along the conveyor able to collect signals coming from rollers and to deliver them in real-7me to a data storage system (on a server)

through either wireless communica7on or cables;

4. a server (database) for data logging of the informa7on coming from the conveyor;

5. a user interface, to be installed on a PC, able to access the database and manage the data according to user requirements (display, analysis, modelling, messages management, …).

Rulmeca solution

Data

aggregator aggregator Data aggregator Data TERMINAL SERVER

System topology

LEVEL Rollers

LEVEL Section of Belt Conveyor

(Semi-Distributed)

LEVEL Site (Concentrate)

LEVEL Belt Conveyor (Discrete)

• •

• •

•

• •

•

Control Room

• Completed PCB layout

• Main challenge - Very size constrained

•  Difficult to fit all components on PC board without compromising

performance

•  Close packing of components may increase fabrication cost

• 45 sensor nodes completed

Prototype Sensors Installed in three

PSV7 Rollers

• Six rollers – 3 PSV7 rollers, 3 PSV3 rollers

• Realistic test conditions

• Speed controlled – 100 to 1000RPM

• Can be moved outdoors for radio range testing

System Level Test Platform

6.