CHINHOYI UNIVERSITY OF TECHNOLOGY
MSC IN ENGINEERING PROJECT MANAGEMENT
Project Management Information Systems (MEPM611)
ASSIGNMENT 1
STUDENT: DESIRE BUNJIRA
REG No: C15124430U
Introduction
The internet of things is a term that encompasses everything connected to the internet. This includes
devices like smartphones, sensors, wearables etc combined with automated systems for a specific
purpose that will be useful to the end user and impact a wider economy.
Tokyo Electric Power Co. Holdings or TEPCO, is a Japanese company which is in the business of
generation and distribution of electric power. The company runs several power plants around Japan. In
2016, the company engaged General Electric and the two companies signed a memorandum of
understanding to jointly develop and introduce IoT solutions throughout TEPCO’s thermal plants.
General Electric has been tasked with improving efficiency at power plants for Japanese electric power
company, TEPCO, using the Internet of Things (IoT).
System Operation
Thermal powerplants basically work using the same principle. A fuel is burned to produce heat which
will then be used to heat water to steam. The steam is then directed into the turbine thereby causing
rotational motion. The turbine will be coupled to a generator which will generate electricity through
induction. The type of fuel to heat the water differs from plant to plant, from fossil fuels, to nuclear and
some even use solar thermal energy.
Hardware and sensors
The furnace is one of the critical components of the power plant. The inside flame and temperature
need to be constantly monitored to ensure complete combustion of the fuel. To do so, soft computing
methods using Artificial Neural Network (ANN) model with Black Propagation Algorithm (BPA) and Ant
Colony Optimization (ACO) will be used for this purpose.
The turbine is also another essential component of the power generation process. Through the use of
IoT, data obtained from IoT is converted into actionable information to improve the turbine
performance. To monitor blade vibration within a turbine, an RF based sensor is used. The monitoring
system will also incorporate magnetic sensors to assess the degree of tip deflection of the turbine
blades.
A thermal power plant consists of valves throughout the plant. The valves are used to regulate the flow
of fluids through a pipe. The valves are traditionally controlled by actuators, usually pneumatic and
electric ones. The pneumatic actuators are used to control processes which require a quick and accurate
response, as these processes do not need a large amount of motive force. As for electric actuators, they
apply an external energy source and their designs are based on the intended tasks within the process.
IoT can be integrated with the actuators to efficiently and effectively control the valves through an
intelligent system.
On the distribution side of the business, TEPCO’s reaction to arrival of smart grid-era has been
impressive. The company installed ‘smart meters’ or advanced metering infrastructure (AMI) that
enhances the efficient and automated monitoring of energy flow across the grid. The company currently
have 10million AMIs installed along the grid. This makes the company one of the world’s leading smart
utilities.
Lastly, to optimize the plant performance and reliability the company is investing in a monitoring center.
The center will make use of an analytic tool called Engineering Consultant Group (ECG) Predict-it
predictive analytics. Predict-it delivers value through the early detection and prevention of equipment
health degradation and also provides advanced diagnostics capabilities for troubleshooting.
Data collected
When it comes to flame and temperature monitoring in the furnace, the most important data to be
collected is the flame image. This is done by detecting, recognizing and understanding the color image
processing for flame color analysis. The principle behind is that the color of the flame images is
dependent on the temperature. The conception of flame detection and classification is conceded to
compute the temperature from its color. The whole specimen flame images will be recognized and
classified based on the temperatures corresponding to the core of the fire ball. The IoT and the
proposed temperature sensor will be connected to the embedded computing system to monitor the
fluctuation in flame temperature with respect to color changes in order to ensure complete combustion
in the furnace.
The RF sensors used to monitor blade vibration collects a lot of data so as to assess the health of the
turbine blade. The vibration monitoring system aims to assess changes in vibration frequency of blades
over time. The sensor hardware will measure the received signal waveform and that will provide
information about the position of the blade with respect to aperture. The magnetic sensors will sense
time of arrival for each blade and the difference between actual and expected times represents tip
deflection. After processing, the data will present a clear picture on the state of the turbine enabling the
plant engineers to schedule a maintenance if there is need. This will play an important role in making
the of informed decisions when it comes to turbine operations and maintenance.
Since the actuator will be used to control the opening and closing of the valve so as to regulate flow, the
main data that has to be collected and recorded is the flowrate of the fluid. The actuators enable the
interaction with the physical world by taking electrical input and transforms the input into tangible
action. A wireless sensor and actuator Network to support the intelligent control of efficient energy
usage will be developed. These technologies collect a high amount of data, which can be very valuable
and useful to an enterprise once it has been stored, organized and processed.
The millions of advanced metering infrastructure installed throughout the grid reads electrical
consumption at a high frequency. The information is then processed in real time and signals sent to
mange demand. The meters also provide the opportunity to capture voltage feedback of termination
points. The smart meters attend to issues to do with data fluctuation that may occur due to load
variations in quasi real-time. With these capabilities, the Automatic Infrastructure Meters give grids a
wide range of functionalities such as remote consumption control, time-based pricing, consumption
forecast, fault and outage detection, and theft detection.
Predict-it collects data by monitoring critical equipment in real time process environments and uses
advanced pattern recognition methods to recognize potential faults. If the monitoring software
identifies a deviation outside of a tolerance limit, it will initiate an event for a subject matter expert to
review and investigate. The software leverages machine learning and advanced equipment diagnostics
reasoning to provide advanced assessment of the equipment health and potential causes.
Benefits of the system
The Internet of Things systems installed by General Electric has helped improve the plant efficiency by 1-
3% reducing costs by about 10 billion yen (US$98.9 million) annually. The technology also averts
unexpected interruption of operations and require fewer inspections than before. This will result in a
reduced plant downtime thereby increasing the number of units of electricity generated over a period of
time. The increased number of units being generated will mean more revenue for Tepco in the long run.
The technology will also improve the plant reliability and that will be good for the business in terms of
competitive advantage.
The introduction of IoT to the plant management will also present an opportunity to monitor the plant
in real time. This will allow decision makers to make critical decisions based on real time data. When it
comes to power plant management, there might be need to make critical decisions within a small
window of period. That window of opportunity can be as little as a couple of minutes. An IoT data
management system provides more instances of actionable data through real time coverage.
Also, when it comes to managing a power plant, data gathering and monitoring systems should be
inherently flexible. With the introduction of the IoT in the plant system by GE, TEPCO will easily monitor
its power plants with flexibility. The IoT technology being introduced creates a mesh network of sensors
throughout the plant that will transmit data to a cloud-based analytics software. The flexibility will allow
management teams to stay mobile throughout the plant while staying up to date at a moment’s notice.
An improvement in the safety aspect of the plant is another important benefit of IoT to the business.
The IoT allows a data center solution provider to create a more efficient network, with a more robust
safety net for detecting malfunctioning equipment before it becomes a serious problem. This enables
the management of the power plant to be more cost efficiently with greater safety redundancy. With
tools like Pedict-it, TEPCO will raise the bar on service, reliability, efficiency and compliance beyond the
levels currently achievable using all other conventional best practices.
On the generation side of the business, optimal operation modes using IoT technology could reduce the
volume of excessively burned fuel, which in some case amounts to half of the expenses. The efficient
combustion of fuel through the use of IoT technology will subsequently reduce the carbon emissions by
the plant. Japan introduced a carbon tax in 2012 to enhance measures to reduce carbon dioxide
emission. Currently the tax rate is at about US$2.60/ton of emitted carbon dioxide and with the
introduction of IoT the emission will significantly reduce. In the long run, this will translate to savings
which will be a positive thing for the business.
The integration of IoT technology in the plant will also benefit the company in the knowledge transfer
management. The system automatically accumulates actual data about troubles and failures to develop
training plans and contents at an individual’s own level. Since solutions are recorded paired with the
relevant troubles and problems, knowledge can be automatically put down into manuals. The company
can now possibly provide a combination of three learning methods based on actual case studies. These
methods include classroom lecture including e-learning, training using a simulator and on the job
training for actual work.
