Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
11Activity
0 of .
Results for:
No results containing your search query
P. 1
boiler in thermal power plant

boiler in thermal power plant

Ratings: (0)|Views: 546 |Likes:
Published by vishalkalma

More info:

Published by: vishalkalma on Mar 31, 2011
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

06/15/2013

pdf

text

original

 
A New Method For Controlling Boiler of Thermal Power PlantUsing Fuzzy Logic
HAMID BENTARZI*, RABAH AMR CHENTIR* and
 
NIKOS E.MASTORAKIS**
*Signals and Systems Laboratory (SiSyLAB)DGEE, FSI, Boumerdes UniversityALGERIA**Technical University of Sofia,English Language Faculty of EngineeringIndustrial EngineeringBULGARIA
Abstract:
- This work presents a new method for controlling fuel and air fed boiler of thermal power plant using Fuzzylogic. The control strategy is based on supervisor level using Fuzzy logic that is required to determine automaticallythe optimal process set points of regulations level. Besides, this paper describes thermal power plant simulatordeveloped by Matlab-Simulink.
Key-Words: -
 
Thermal power plant, Boiler, Coordinated Control Mode, Optimal Controller, Fuzzy Logic.
1 Introduction
The number of construction of a new thermal powerplant is relatively reduced. However, thermal powerplants account for approximately 65% of the world’spower supply. Recently, there have been concernsregarding the deterioration in the efficiency of existingthermal power plants. The efficiency of such type of power plant is very low and great amount of loss inthermal energy may be noticed [1]. In order to generate arequired electric energy, the turbine needs an equivalentamount of thermal energy in addition to the loss.Minimizing the loss leads to a reduction of pollutants inthe environment as well as production cost. Steam ismerely a vehicle which transports the energy from theburners to the turbine shaft. The flow of steam is notonly the parameter that should be determined, but alsothe temperature and pressure must be taken intoconsideration for a given value of power.Nowadays, for improving the efficiency of the thermalplants, the control strategies with different hierarchicallevels have been used. In the high levels, the controlstrategies are coordinated and the process set-points aredetermined. However, this work presents a new methodof control strategy based on supervisor level that isrequired to determine automatically the optimal processset points of regulations level using Fuzzy logic. Thesupervisor level is based on an energy optimizer andconsidering the process operational constraints. Sincecombustion control in a utility boiler is one the mostimportant control loops of power plant, for this reasonthe work focus has given to the boiler controllers.
2 A Simulator of Thermal Power Plant
2.1 Process Description
In a thermal power unit, steam is produced inboiler, is expanded in the prime mover (Turbine) andcondensed before feeding it into the boiler again. Theturbine shaft is coupled with alternator, which is used toproduce electricity. The power plant can be divided intofive main systems, namely a) Fuel and Ash system b)Air and Flue Gas system c) Fed water and steam systemd) Condenser and e) Cooling water system. Schematicdiagram of a typical thermal power plant is shown in theFig.1. The feed-water is supplied to the drum, where thethermal energy of combustion products is transferred tothe condensed water. Then, the feed-water passesthrough risers, where the furnace heat is used to increasethe water temperature and eventually it causes theevaporation.Thus, the circulation of the water and steam take place inthe drum and risers. Steam generated in the risers isseparated in the drum, from where it follows through thesuperheater to the high pressure turbine. Then, this steamis recycled to the boiler in the reheater where its energycontent is increased.
2.2 Process Modeling
The main assumptions and equations of thethermal power plant components are presented [2-4].The boiler equations are based on the followingassumptions:
COMPUTING and COMPUTATIONAL INTELLIGENCEISSN: 1790-5117 31 ISBN: 978-960-474-088-8
 
 Fig.1 Schematic layout of a typical power plant.-Polynomials approximations of steam tables are used toestablish the relations between the steam parameters,such as the enthalpy, density, temperature and pressure.- The superheated and the exhaust gases of the furnaceare considered as ideals.The following physical thermodynamic equations aresatisfied.Heat balance
)(
ououououininin
hdt hwhwQ
ρ 
+=+
(1)Where, Q
in
is incoming heat flow, win inlet massflow, h
in
inlet specific enthalpy, w
ou
outlet mass flow, h
ou
 outlet specific enthalpy.Mass balance
)(
ououin
dt ww
ρ 
=
(2)Then, the main equations of steam turbine are given by :Mass balance for steam
)(
ooi
dt ww
ρ 
=
(3)where, w
i
is inlet steam flow, w
o
outlet steam flow,Heat balance for steam
)(
ooooii
hdt hwhw
ρ 
=+
(4)Where h
i
inlet steam specific enthalpy and h
o
is outletsteam specific enthalpy.
3
 
A Proposed Control Strategy
The most modern thermal generation plants use a controlscheme that is usually named an integrated orcoordinated control system. This type of controllersimultaneously adjusts firing rate, pumping rate, andturbine throttling in order to follow changes in loaddemand. Such a coordinated control mode is shown inFig.2. In this type of control, both pressure andgenerated output are fed back for the control of bothboiler and turbine. In this manner, it is possible toachieve the stable and smooth load changes of theturbine-following mode and still enjoy the promptresponse of the boiler-following mode. This isaccomplished by making maximum use of the availablethermal storage in the boiler. Both pumping and firingrates are made proportional to the generation error sothat these efforts are stabilized as the load approachesthe required value. Pressure deviation is controlled as afunction of the thermal storage and the generation error.Nowadays, the control strategies for the thermal plant isorganized using different hierarchical levels. In the highlevels, the control strategies are coordinated and theprocess set-points are determined. The proposedcontroller is based on supervisor level that is required todetermine automatically the optimal process set points of regulations level using an economic optimizer andconsidering the process environmental and operationalconstraints. After that the supervisor level makescorrection continuously through the use of Fuzzy logicto the existing conventional regulations.
Fig.2 Coordinate control mode.
3.1
 
Implementation using Matlab/Simulink
Simulator has been developed for steam turbinegenerator using the powerful software programMatlab/Simulink. It includes one boiler steam turbineand generator as shown in Fig.3. Parameters that havebeen used for testing this simulator have been taken fromsteam turbine generator (SIEMENS-KWU-SGP) of 176MW manufactured by consortium of Austria-Germane.The used conventional controllers which are simples instructure, are not suitable for non-linear, higher order,time delayed and complex systems that have no precisemathematical models. Besides, these controllers needfrequent tuning that is not easy task and is also timeconsuming.Several methods have recently been developed, most of them are based on controllers with fuzzy system withoutincorporating any on line adaptive structure and itsapplications particularly for combustion control of boiler.
COMPUTING and COMPUTATIONAL INTELLIGENCEISSN: 1790-5117 32 ISBN: 978-960-474-088-8
 
 Fuzzy systems, as an artificial intelligence approach,emerge in power plant as a complement to mathematicalapproaches. For a comprehensive survey of fuzzy settheory in power plants, the reader is referred to book[6].Fuzzy-logic controllers (FLC) can be classified asknowledge based systems (KBS). There are principallytwo classes of KBSs; namely: supervisory expert controlsystems (SECS) and direct expert control system(DECS) [6,7]. SECS’s use FL to tune the controller inthe main loop.
 3.2 Fuel and Air Controller Using Fuzzy Logic
The proposed fuzzy control for fuel and aircontrol system is shown in Fig.3. The characteristics of thermal power plant (SIEMENS-KWU-SGP) have beenused which are shown in two response graphs, one givesload as function of fuel and the other load versus airflow.The fuel and the airflow set points can be derived fromthe graphs in our simulator from these characteristics.The difference between the set point and the actual air orgas flow is computed as error signal.The proposed DECS implemented for combustionprocess using fuzzy logic control has two inputs and oneoutput for fuel and airflow respectively. The inputs areerror signal and change in error signal. The universe of discourse of the controller variables are e,
Δ
e and Urespectively. The following are the range of databaseconsidered. Error (e) =-50% to +50%,
Δ
e = -25% to+25%, control valve position (U)=10% to 100%. Thenumber of linguistic terms for each linguistic variable is5. (e) for fuel = {MN, N, Z, P, MP}, (
Δ
e) for fuel ={VS,S,M,L,VL}, (U) for fuel ={VS, S,M,L,VL}, (e) forair = {MN, N,Z,P,MP}, (
Δ
e) for air = {VS,S,M,L,VL},(U) for air = {VS,S,M,L,VL}.
Fig.3 Complete system of power plant using Simulink.
The triangular membership functions are used torepresent the linguistic terms. (VS=very small, S=small,M=medium, L=large, VL=very large, MN=mediumnegative, N=negative, Z=zero, P=positive, MP=mediumpositive).The processed signal from the fuzzy controller isdefuzzyfied and applied to the respective adder circuit.In the proposed FLC based SECS, the differencebetween feed-forward error and the adaptive error. Theerror and the change of the error are taken as inputs of the proposed SECS controller. The processed signal isdefuzzyfied and applied to the respective adder circuit.The sum of the DECS and SECS output signal issupplied to the fuel control valve as well as the airdamper.
3.3 Adaptive Error Factor
To improve the performance, an adaptivevariable error set point has been developed to the SECScontroller for fuel and air flow. When there is dynamicchange due to load demand change the adaptive errorcalculation is proposed by taking into consideration thedifference between the actual specific heat consumptionand the required one. An adaptive error, which isequivalent to the required (adaptive) at particular load, iscalculated as follows:
COMPUTING and COMPUTATIONAL INTELLIGENCEISSN: 1790-5117 33 ISBN: 978-960-474-088-8

Activity (11)

You've already reviewed this. Edit your review.
1 hundred reads
1 thousand reads
Azra Avdić liked this
prabhjot singh1 liked this
SKBABU liked this
Amit Hadke liked this
selvam2010 liked this
Subhabrata Saha liked this

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->