Fuzzy based Boiler Control
Automation system vendors, universities, research centers and the process industry continually strive to invent new and better methods to keep processes under control. Boiler plays an important role in power generation units and its control is very critical in many applications. In present situation conventional PID control is being implemented for this purpose. These conventional controllers in power plants are not very stable when there are fluctuations and, in particular, there is an emergency occurring. Continuous processes in power plant and power station are complex systems characterized by nonlinearity, uncertainty and load disturbances.
The characteristics of a power plant system change significantly between heavy and light loading conditions. Applying of traditional control methods encounter great difficulties while the process working condition changes within a large operation range. So many studied control methods like expert systems, fuzzy logic control, neural networks and knowledge-based systems, termed Intelligent Control have become common additions to conventional Proportional Integral Derivative (PID)-type control methods in the process industry. This project discusses the possibilities and suitability of an intelligent control method such as fuzzy logic control in controlling and managing combustion performance of an Industrial boiler. This controller demonstrated superiority in controlling the temperature and water level of the boiler.
In this project It has been designed a model with fuzzy control implemented for temperature and level control. The model consists of a process tank along with a reservoir together with a pump and heater connected with it. This heater and pump are controlled using fuzzy logic and the temperature and level conditions are maintained.
The characteristics of a power plant system change significantly between heavy and light loading conditions. Applying of traditional control methods encounter great difficulties while the process working condition changes within a large operation range. So many studied control methods like expert systems, fuzzy logic control, neural networks and knowledge-based systems, termed Intelligent Control have become common additions to conventional Proportional Integral Derivative (PID)-type control methods in the process industry. This project discusses the possibilities and suitability of an intelligent control method such as fuzzy logic control in controlling and managing combustion performance of an Industrial boiler. This controller demonstrated superiority in controlling the temperature and water level of the boiler.
In this project It has been designed a model with fuzzy control implemented for temperature and level control. The model consists of a process tank along with a reservoir together with a pump and heater connected with it. This heater and pump are controlled using fuzzy logic and the temperature and level conditions are maintained.
BLOCK DIAGRAM DESCRIPTION
Boiler is the zone of interest.The boiler parameters are temperature, level, flow and pressure. These parameters are being sensed by the corresponding sensors RTD, Float, Proximity sensor and Diaphragm type pressure sensor respectively. Sensed signal is not compatible with the PC hence necessary signal conditioning (amplification, linearization, etc.,) is provided to obtain the signals in the range of 0-5 V.
The next step is the conversion of these analog signals to digital signals and this is accomplished through ADC.These digital signals simultaneously enter the PC via parallel port. Controlling of these signals is done using fuzzy logic in matlab. Here the real time input signal is compared with the set value and the difference is converted to analog through the DAC and then is given to the control circuitry (heater control and pump control). The control circuitry consists of a TRIAC which switches on and off the heater and pump according to the decision taken by the Fuzzy logic controller. Isolation is provided in the control unit for protection of the circuit as well as the personnel involved.
Boiler is the zone of interest.The boiler parameters are temperature, level, flow and pressure. These parameters are being sensed by the corresponding sensors RTD, Float, Proximity sensor and Diaphragm type pressure sensor respectively. Sensed signal is not compatible with the PC hence necessary signal conditioning (amplification, linearization, etc.,) is provided to obtain the signals in the range of 0-5 V.
The next step is the conversion of these analog signals to digital signals and this is accomplished through ADC.These digital signals simultaneously enter the PC via parallel port. Controlling of these signals is done using fuzzy logic in matlab. Here the real time input signal is compared with the set value and the difference is converted to analog through the DAC and then is given to the control circuitry (heater control and pump control). The control circuitry consists of a TRIAC which switches on and off the heater and pump according to the decision taken by the Fuzzy logic controller. Isolation is provided in the control unit for protection of the circuit as well as the personnel involved.
5 comments:
Thanks for your valuable comments, I am really happy to receive such a nice comment.
hey ur info is really very good......but i have certain doubts to clarify ....how do u think fuzzy logic will help when a c-language program can do the job. in what way is fuzzy logic more flexible....if so it would be nice if u can share ur fuzzy codings....pls...that would be of real help.
My name is tim leopold. i am a boiler super tuner for ABB (I am an old Bailey guy). I would like to see some of the data that you have that can substantiate your claim that fuzzy logic is a better way to control level and temperature than a PID controller with appropriate feed forwards, adaptive tuning, and comparable signal conditioning. You say that it has better controlability, especially in emergency conditions. You must then be aware of runbacks, rundowns, manual interocks, and control tracking schemes that have been de veloped and are standard throughout the industry?
Additionally the situation that you described is somewhat fuzzy (no pun intended) to me. Okay the pun was intended. What temperature are you speaking of? feedwater temp? And the idea that you have the pumps and heaers turning on and off is nothing like the modulating control that are typical in power plants.
I am interested in the application of fuzzy logic to boiler controls, You must understand the complexity of the system. And the robustness of conventional controls on everything from MW units to 850 MW units is something that I have personally experienced. You might want o take a look at my book on boiler controls, available on amazon, 'You Can Tune a Boiler But You Can't Tune a Fish' in order to get some basic grounding in boiler controls if your experince is somewhat limited in the industry. Or check out my blog at http://timthetuner.blogspot.com/
That being said I am interested in pursuing this and would love to be of any assistnce that I can in developing more realistic developments.
Good luck and feel free to contact me.
I think this was a very nice post. Thanks for the great content. I will take this into consideration. Anthony Hunter
Depending on the complexity of the desires of customers for safety and information almost any system can be developed to report sensor activity and relay commands. I would like to suggest a different method, which may be simpler, more accurate over time and precise.The method of employment can be through modulation or on off control. The important thing is to give the boiler a brain so it knows what is needed. The rest is window dressing made complicated to justify unnecessary clerks. I throw in a web site for your library. We are very interested in the system requirements of Industrial applications. www.Exqheat.com
Post a Comment