Development of Neuro-fuzzy Controller for Alcohol Separation From Aqueous Solution

Project Title

Development of Neuro-fuzzy Controller for Alcohol Separation From Aqueous Solution

Project Area of Specialization Electrical/Electronic EngineeringProject Summary

The project aims to solve the local Sugar industry problem of ethanol-water separation at low cost and efficiently. We would develop the ethanol-water separation system based on nuero-fuzzy algorithm which would later be implemented on Programmable Logic Controller (PLC).  Ethanol-water separation mixture is a very common and usually found during the fermentation of sugarcane, so this mixture would be taken as an object. In proposed system the technique that author would utilize is the fractional distillation. Since Alcohol is lighter than water along with lower vapor pressure, so it evaporates. The distillation operation is totally dependent on temperature of miscible solution.  The task can be efficiently accomplished through electronic automation, it has now become a basic requirement for industries because it can reduce the maintenance, generates the quality products and improves productivity Fuzzy logic controller is an intelligent toll to moderate and control the system process, it intelligently control the temperature of the system distillation by comparing with its pressure. The whole system would be controlled based on PLC and performance can be monitored on Human Machin Interface (HMI)

Project Objectives

(1) The main objective is to get the separated solvents from the mixture of two miscible liquids using Nuero-fuzzy controller implemented on PLCs. We aim to take alcohol-water homogenous mixture. As it is commonly found during the fermentation of sugarcane. The separation would be done using distillation technique.

(2) To provide the continuous monitoring and controlling of the distillation column parameters on the HMI screen and manual control through touchscreen of HMI.

(3) To provide easiness with less human efforts and also make the whole process faster.

(4) To Install the proposed system in local industry

(5) Commercialize the proposed system

Project Implementation Method

Project Implementation Method

Figure below shows the block diagram of the proposed system.
We would use Nuero-Fuzzy controller, because it is simple to implement, and solve classes of previously intractable problems.  The input set point command would fuzzifies the value and would compared with set-point and error signal would be generated according to the Nuero-Fuzzy Sugeno Model.  The Model would provide the defuzzified value which works as
the control input to the SCR based AC Voltage controller. Hence fine control of temperature in the distillation column can be achieved intelligently. The RTD PT100 would be used for measuring the temperature of the homogenous solution and at required set-point the lighter solvent will be evaporated leaving behind heavier one. The whole system would also be monitored and recorded using HMI.

Benefits of the Project

Following are the benefit of our system

1. Addresses the local sugar industry problem, and make it automatic.
2. The system would give high separation and energy efficiency.
3. The separation would be achieved at liquid temperature significantly less than the boiling point of the mixture.
4. Flexibility as alcohol fermentation can be accomplished batch wise or continuously.
5. Potential to be used in local industry

Technical Details of Final Deliverable

A Programmable Logic Controller is a solid control system that continuously monitors the status of devices connected as inputs. It is a control system with functions to control logic, sequencing, timing, arithmetic data manipulation and counting capabilities. It can be viewed as an industrial computer that has a central processor unit, memory, input output interface and a programming device. The central processing unit provides the intelligence of the controller. Input output interface is a communication link between field devices and the controllers; field devices are wired to the I/O interfaces. Through these interfaces the processor can sense and measure physical quantities regarding a machine or process, such as, proximity, position, motion, level, temperature, pressure etc.

The feed stream would fetched continuously in the distillation column.  Initially, the column would be empty which needs to be filled and maintained at required level, so level transmitter would provides the status of the column to the PLC which controls the feed pump. After reaching at required level PLC would takes the values of RTD in terms of decimal values and provides the fuzzified value to command which would compares the current temperature with the set-point and according to the equations based on the Sugeno fuzzy model output would be sent to the control circuitry of the AC voltage controller which variably controls the RMS voltages of the heater. The heater would be run variably until the temperature of the column would reached the set point. At  set point temperature and pressure the alcohol would vaporized completely from the solution and would be collected in the container via condenser and would leave behind only one solvent that is water. The water would then have drained out through the solenoid valve and would be collected in the container.  The operation would be monitored and controlled manually as well through HMI touchscreen.

Final Deliverable of the Project HW/SW integrated systemCore Industry FoodOther Industries Energy , Manufacturing , Health Core Technology NeuroTechOther Technologies Artificial Intelligence(AI), OthersSustainable Development Goals Industry, Innovation and Infrastructure, Responsible Consumption and ProductionRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	77000
PLC s71200	Equipment	1	30000	30000
Pressure Transducer	Equipment	1	5000	5000
RTD	Equipment	1	2500	2500
Heater	Equipment	1	3000	3000
SCR voltage regulator	Equipment	1	10000	10000
Agitator	Equipment	1	18000	18000
Thesis Printing	Miscellaneous	1	1000	1000
Shippment cost	Miscellaneous	1	5000	5000
wires	Miscellaneous	1	1000	1000
others	Miscellaneous	1	1500	1500