Intelligent Coal Fired Thermal Power Plants - Next generation automation to accommodate high renewable penetration into the grid
The requirement for making coal fired stations intelligent arises from the beed for highly efficient plant operation and asset management on one hand and wide scale penetration of renewable energy (solar photovoltaic and wind) into the grid which results in rapid cycling of the thermal plants, on the other hand. Ramping up or down of thermal plants hitherto was through manual load setting or through the free governor mode of operation. With the entrance of rapidly varying renewable generation, the need for automatic load setting on the thermal plants through intelligent features arises...
- M. Siddhartha Bhatt,
The requirement for making coal fired stations intelligent arises from the wide scale penetration of renewable energy (solar photovoltaic and wind) into the grid which results in rapid response cycling of the thermal plants. Ramping up or down of thermal plants hitherto was through manual load setting or through the free governor mode of operation. With the entrance of rapidly varying renewable generation, the need arises for automatic load setting on the thermal plants through intelligent features to respond to variations in renewable generation. Manual load setting or free governor mode has limitations of response time. Also the ramping rate is also to be automatically set. When the load on the thermal plants goes below its technical minimum load, coal cannot be fired into the plant because of flame separation but fuel oil has to be fired to maintain the load. The other issue with cycling plants is that the plant machinery especially of the power block are designed for a specific number of cyclic operations.
Figure 1: View of a coal fired thermal power plant
When the rate of ramping and cycling goes up, the plant component life will be reduced due to figure and creep mechanism of degradation. The power plant assets (boiler, turbine, generators, major auxiliaries, etc.) are designed for an operational life of 3,00,000 (3 lakh) operating hours or around 35 years of service under normal operating regime. The factors which affect the operational life are both the physical running hours as well as cyclic (on/off) operations. Each on/off or start/stop operation can be taken as an expenditure of 20 h of steady operational life. The allowable starts of base load units are 10 hot starts/year, 5 warm starts/year and 3 cold starts/year. For peaking units the starts are much higher. For all units, starts and stops are factored into the life expenditure @ 20 h/start on an average.
Typical (normal) number of starts in the life of a unit over 200 MW unit are as 5000 hot starts, 1000 warm starts and 500 cold starts. The design step load change is + 15% /min and the ramping rate is + 5%/min.
The plants are designed for continuous operation at near full load with annual PLF of around 80 %. Under high renewable injection, the thermal plant loading will have to be ramped up and down resulting in cycling transient operations with low load factors. When the rate of cycling goes up, the plant component life will be reduced due to degradation mechanisms of fatigue and creep rupture initiation.
Figure 2: View of a Solar PV Power Plant based on Crystalline Silicon Modules
Figure 3: View of a Solar PV Power Plant based on Thin Film (Amorphous Silicon) Technology.
Power plant operation involving intelligence features through IT applications involves automation of manual operations, control of systems, data acquisition and logging of information to fulfil requirements of versatility, user friendliness and cost competitive.
Intelligent Power Plants
Intelligent power plant automation is defined as a composite system of components which enables the operation, monitoring, control, co-ordination, security in real time mode from remote locations to fulfill the objectives of load management, energy efficiency, environmental control and resource conservation. Intelligent power plants help to save resources of primary energy (coal and solar photovoltaic power) by avoiding shut downs to balance the system.
The components of a coal fired station which respond dynamically are:
• Steam turbine
• Feed water heaters
• Lube oil coolers
• Heat recovery devices: gland steam condenser, stack steam condenser, vent steam condenser
• Auxiliaries: Pumps, Fans, Mills
To handle this dynamic behavior, power plant automation started with supervisory control and data acquisition (SCADA), Energy management systems (EMS), distributed control systems (DCS) and field control stations (FCS) with controls to maintain the operating safety band.
Later controls for performance optimization were also included. Presently power plant automation encompasses control and monitoring systems for meeting the requirements of load control, fluctuating grid conditions, system security, communication, safety and operational energy efficiency.
The starting point of intelligent power plant designs is the digitization of the existing instrumentation and control loops which open the floodgates of vast array of technologies hitherto inaccessible: Soft instrumentation, Expert systems, Logic based systems, Intelligent instruments & devices, Optimization based on regression data, Optimization based on deterministic calculations of 3-d profiles, Fuzzy logic models, Computational models, ANN models, Multimedia systems, Web interactions, Telemetry based gateways and Diverse external connectivity through communication technologies.
Figure 4: Typical view of a power plant automation system
In a digital environment, information from all sources can be freely exchanged and utilized in any form creating a uniform digital currency which enable intelligent features. Some of the expected functions of an intelligent distributed control power plant automation system are:
• Data acquisition: Logging of data, Event report, report generation
• MMI: Graphical User Interface and Cross Platform Portability
• Computation: Computation of gross performance parameters and indices, computation of 3-d system profiles, neuro computing, computation for pattern recognition.
• Decisions: Logical operations, fuzzy logic based decisions, neuro-fuzzy decisions, pattern recognition decisions, expert system based decisions
• Monitoring: System operating point data, Component specification, Customization, Alarm generation (audio / video)
• Control: Switching, interlocking, modulation
• System Information: System variables and status (CB, LBS, isolator, auto trip, etc.) Component health,
• Security: authentication , intruder alarm, access control, surveillance
• Watchdog: diagnostic and rectification tool, CCTV
• Network Generation: Topological information, Graphical representation, Editing, validation, printing
Intelligent automation is achieved by a network of the following elements:
• Primary sensors and their associates transmitters which acquire data and control components
• Control elements such as actuators for valves
• Analog to digital (A/D) and (D/A) converters
• Processors of digital data
• Software & drivers
A view of a coal fired thermal power plant is shown in Figure 1. Views of solar pv power plants are given in Figure 2 & 3. A detailed list of components are given in Annexure 1.
Automation is based on the principle of converting all inputs and outputs into digital forms. Figure 4 gives the basic configuration of a power plant automation system.
The automation system can be designed and developed using information technology/embedded systems and integrating the same into the existing power plant systems. Components such as computers, Remote Terminal Units (RTUs), actuator control of motorized valves, breakers, switched capacitor banks, on-load tap changing transformers, load break/make switches, auto re-closures, sectionalizers, and communication systems can be integrated into the automation system. Integration with Automated Mapping (AM) and Geographical Information System (GIS) Software packages is widely used at present.
An intelligent integrated distributed control, automation system enables power plants to have real time on-line control of energy efficiency, system security, safety systems at an improved ENERGY efficiency which ultimately results in lower costs, better reliability, planned control & optimum resource utilization.
The power plant automation systems aims at achieving:
• Intelligent operations and control of the plant
• Reactive power control through switched capacitors banks
• Fault detection and isolation
• Control of DTRs, pumps, fans, mills, heaters and their motors
• Control of air conditioning systems, water heating systems, lighting systems
• Operation of auto re-closures, lines sectionalizers, motorized actuators,
• Data acquisition from load end CTs and PTs
• Operation of CBs
• Issue, control and receiving back of line clears, introduce interlock and safety algorithm, password protected operating environment.
• Real time logging of data / archived records.
• Equipment data base
• Interactive voice based security systems
The major components are described below:
Some of the sensors are:
• Fibre optic photoelectric sensors: thru-beam, retro-reflex, diffuse -reflective and definite-reflective types
• Photoelectric sensors: self contained, focused beam, transparent target and one touch calibration types
• Proximity sensors
• Displacement sensors
• Pressure sensors
• Area sensors: Laser thru-beam and retro-reflex
• Electromagnetic sensors
• Chemical sensors: zirconia, electrochemical
• Temperature sensors: RTDs, thermistors and IR sensors
• Optical CTs & PTs.
• IR, UV & optical sensors
• Accoustic sensors
The sensor technology is moving into the area of non-interactive and non-intrusive measurement systems.
MEMS is the integration of mechanical elements, sensors, actuators and electronics on a common silicon substrate through micro-fabrication technology. While electronics are fabricated using IC process (CMOS, bipolar, BICMOS processes), micromechanical components are fabricated by micro-machining by etching silicon wafer or adding new materials to form mechanical and electromechanical devices.
Smart and Intelligent Sensors
Sensors are primary sensing elements in the distributed control automation system.
Smart sensors are those that have additional secondary sensors which measure secondary variables and then use them to correct the calibration of the main sensor. A smart pressure transducer is one where the deviation in calibration due to changes in atmospheric temperature and pressure are sensed (as secondary variables). These are used as inputs to calibrate the pressure transducer.
An intelligent sensor is one which has a logic based internal control system which performs certain programming operations for optimizing the required tasks. Fuzzy logic and neural network based controls are most popular.
Figure 5: Components of an automation system
Remotely Operable Devices
These include Air break type Load Break Switch (LBS) for 6.6 kV-11 kV operation ( operating time: 80 ms); Moulded Case Circuit Breaker (MCCB) unit for 415 V operation (operating time: 80 ms); Load change over switches for 6.6 kV-11 kV operation (operating time: 80 ms) or Motorized control valves for flow quantity control, flow isolation, flow bypass, etc.(operating time:4-5s)
Closed Circuit TVs
CCTVs play a role in digitization of process operations as well as surveillance. Typically CCTVs help in flame monitoring, fire ball monitoring, leak detection in coal and water piping prone to leakage.
Static Electronic Devices
Some of the new trends in power plants is the constant and continuous movement from mechanical and electro mechanical components to static electronic digitally controlled systems such as the following:
• Numerical control relays
• Turbine governing
• Turbine start-up
• Generator excitation
• ESP control
• Generator synchronization with the grid
Remote Terminal Units
Remote Terminal Units (RTU) designed through standard off-the-shelf cards are used for acquisition of data parameters in the field and transmitting these to the control centre along with capability to exchange the data with IEDs, microcontrollers and DCS/FCS/SCADA. The RTU are designed to be flexibile and expandable, modular at signal conditioning and communication interface level, low cost, rugged, intelligent microprocessor based with bi-directional data communication with 24/48/54 analog and 24/48/96 digital I/O channels.
The processor is the heart of the automation technology. Power plant automation is generally build around one of the following core technologies:
• Multi-medial technology
• Transputer based instrumentation and control systems.
• SCADA (Supervisory control and data acquisition) systems.
• EMS (Energy management systems)
• Field control stations (FCS)
• Distributed control system (DCS)
Distributed Control Systems (DCS)
Distributed control system (DCS) is a system whereby all control processing is decentralized, built up of stand-alone controllers and independent of a central computer to avoid damage to system due to controller failure.
Remotely Operable Micro Controllers & Field Control Stations
Microcontrollers operate as per microprocessor based control logic, derives control logic operate on current and voltage sensors, samples the data on continuous basis, conditions the signals and converts to digital form for processing, decision making and logging.
Intelligent Electronic Devices (IEDs)
Intelligent Electronic Devices (IEDs) (such as Intelligent transformer) have inbuilt intelligence into them depending on their functions. IEDs available at the site such as relays, electronic energy meters, flow controllers, motor controllers directly communicate the data with the DCS. Analog quantities and digital information, which are not available directly from the IEDs, are extracted through the I/O interfaces of the RTUs. RTUs have provision to send the control command to the actuator of a switching element through the IED relay (if available). Few analog and digital signals, which are not handled by any of the IEDs, are directly connected with the RTU that communicates with DCS.
PC based Intelligent Power Plant Automation Systems
The various components (software and hardware) needed for PC based power plant automation consist of Industrial work stations, server-slave configurations, Signal conditioning, Modular industrial controllers, Industrial communications and Application software.
Unlike traditional communication solutions, the approach in power plant automation is to have a core communication controller at the control centre, which can support diverse choices of communication media options such as the following radio, satellite, hardwire, wireless Ethernet, etc.
One-way VHF radio can be used for load control because low cost load control switches are available for this technology. VHF radio switches can also be used for such applications as capacitor control instead of more expensive RTUs.
The components of an automation system are indicated in Figure 3 below.
The present techniques of data communication among the IEDs supplied by different vendors involve adherence to standard protocols such as IEC-60870-5 (International Electro-technical Commission).
Integration of Components & Networking
Integration is the ability of control system components from independent manufacturers to inter-connect and facilitate coordinated real-time data exchange and control through common communication data exchange protocol and man machine interface or integrator interfaces.
Network is the linking of a system of distributed control units on a communication highway. A network enables central monitoring and control of the entire system from any distributed control unit location and permits sharing of point information between all control units. First tier networks provide "Peer-to-Peer" communications while Second tier networks provide "Peer-to-Peer", Master-Slave or Supervised Token Passing communications.
Intelligence features can be incorporated in the most of the components. Building a digital utility by development of intelligent IT architecture of the power plant suitable for seamless interface for automatic process control and e-business and Web based interactions is gaining importance.
Some of the areas of intelligent automation in power plants are :
Boiler and its associated system
• Coordinated boiler-turbine control
• Boiler stress level monitoring in certain areas
• Fatigue cycling monitoring
• Residual stress monitoring
• Boiler tube cleanliness monitoring and intelligent soot blowing
• On-line combustion diagnostics
• Oil gun operation
• Furnace safeguard supervisory system (FSSS) and burner management system (BMS).
• Soot blower operation
• Operation of continuous/intermittent blow down (CBD/IBD)
• Burner tilt operations
• Combined O2/CO based control strategy for combustion control
• Fireball centering controls
• Flame scanning through CCTV and control and fireball
• Condition monitoring of failure prone tubing
• Tube leak monitoring
• Monitoring of corrosion prone boiler components
• Slag monitoring
• Noise monitoring inside boiler
Turbine, Condenser and its Associated System
• Turbine stress level monitoring
• Turbine supervisory instrumentation for:
Relative shaft vibration
Absolute vibration in the bearing pedestals (horizontal and vertical)
Differential expansion of rotor and cylinders
Positions of emergency stop valve and control valve
Turbine metal temperatures
Bearing metal temperatures
Drain oil temperatures
• Fatigue cycling monitoring
• Electronic governing and synchronization
• Heat rate and heat balance monitoring
• Process optimization
• Maintenance management
• Condition monitoring
• Noise monitoring
• Endoscopes for examination of blading without opening of the casing.
• Expert system for signature analysis and fault diagnosis
• Expert system for heat rate and efficiency monitoring.
• Generator remaining life monitoring
• Stress level monitoring
• Fatigue cycling monitoring
• Electronic synchronization.
• Numerically controlled generator excitation.
• Numerically controlled generator protection.
• Bearing vibration, eccentricity, axial shift monitoring
• Speed monitoring
Auxiliaries- in-house and out-lying
• Bus transfer schemes for change over from station transformer to auxiliary transformer.
• Variable pressure operation of BFP
• Operation of FDFs, IDFs & PAFs at optimal efficiency point
• Operation of CEPs, CWPs at optimal efficiency point
• Operation of mills at optimal loading
• Operation of conveyors in coal handling plant
• Logic based operation of cooling towers based on ambient conditions and cooling requirements.
• Operation of crushers
• Fuel management system
• Operation of water treatment plant
• Logic based operation of river water pumps
• Logic based operation of ash handling pumps
Many of the manual controls, semi-automated open loop controls can be converted into closed loop controls by the integration of some of the following with the DCS:
Plant Outages Analysis
Power plants have basically five types of outages:
• Forced outages
• Pre-arranged outages
• Planned annual overhauls
• Planned capital overhauls
• Planned outages associated with renovation/modernization/up gradation/ revamping/ retrofitting/life extension programs.
Outage management involves a combination of modern technology, econometric techniques and data management. The ITs find application in minimization of deployment time, event management, knowledge application, use of public information to achieving reduction in cost and improvement in efficiency. The present technology of authentication, confidentiality, integrity and non-repudiation is suitable for e-business during outages.
Intelligent operation of thermal power stations involves:
• Cold start
• Warm start / hot start
• Black start (grid has failed)
• Absolute cold start
Intelligent start-up of a thermal power plant unit consisting of:
• Water filling
• Fuel oil support
• Steam generation
• Turbine rolling as per the heating ramps
• Steam ejectors-condenser vacuum
• Turbine synchronization with the grid
• Turbine loading as per the heating ramps
Dynamics of a power plant involves:
• Load setting on the unit and associated control of Mill throughput, Coal flow, Steam flow, Turbine output, Generator output
• Predictive emission monitoring
• Performance monitoring
• Process optimization
• Maintenance management
• Compliance reporting
• Coordinated boiler-turbine control
By and large the most important feature expected out of an intelligent power plant is the load fixing based on intelligent features in the event of a transient from the paralleled renewable generation in the grid.
The main conclusions are as follows:
i. Intelligent power plants help to save resources of primary energy. Intelligent features can be integrated into the plant DCS based on 3-d computations, logic learning algorithms, neural computing, pattern recognitions, expert systems, etc., for accommodating variation arising from integration of sizeable renewable power.
ii. Present concepts of intelligent power plant automation involve operation and control based on optimal performance or optimum decisions emanating from on-line computation.
iii. New developments in non-interactive sensor technology and MEMS now enable very sophisticated solutions at reasonable costs to meet the objectives of optimal operation of the power plant through intelligent systems.
If you want to share thoughts or feedback then please leave a comment below.