Abstraction: A microcontroller based viscousness measuring apparatus has been developed to mensurate kinematic viscousness of liquids utilizing redwood viscosimeter. This instrument system permits to maintain the temperature of the sample at any coveted value and recording of clip for 50 milliliter of sample aggregation to calculate viscousness of comestible oil samples and to direct informations to personal computing machine to enable the computing machine processing of such informations. A dedicated PIC16F877 based microcontroller board was employed for the hardware. A three bed nervous web is used to develop the viscousness at different temperature utilizing back extension algorithm and the trained nervous web is used to calculate the viscousness of comestible oils. The inside informations of its interface to mensurate kinematic viscousness and to mensurate, temperature and to command the temperature and evaluate consequences are explained in this paper.

Keywords: Microcontroller, kinematic viscousness, redwood viscosimeter, Back Propagation and Neural Network.


Viscosity measuring and control has great importance in nutrient industry and accurate cognition of viscousness is necessary for assorted industrial procedures. Viscosity is a direct measuring of a fluid ‘s quality. A alteration in viscousness can bespeak a cardinal alteration in the stuff under trial. Liquid viscousness, a basic physical belongings, straight influences unit operations such as pumping, filtration, filling, distillment, extraction, and vaporization, every bit good as heat and mass transportation. Viscosity is a really of import belongings of lubricating oil. Kinematic viscousness is defined as the ratio of absolute viscousness to mass denseness and has the unit of m2/s. Properties, like temperature and force per unit area, influence the Kinematic viscousness. Frying is one of the most normally used methods of nutrient readying in the place and in industry, and the drawn-out usage of oil for this purpose causes alterations in its physical and/or chemical belongingss ( Moretto and Fett, 1998 ) . It is, hence, indispensable to find viscousnesss of the comestible oils at assorted temperatures.

The application of unreal neural-networks in instrumentality peculiarly in measuring applications is the best increasing pick due to the efficiency in informations acquisition and processing capableness. Neural-networks will besides happen an extra use in instrumentality and measuring applications due to the accomplished many undertakings non executable utilizing conventional techniques such as the ability to larn and put to death monolithic analogue processing construction. This paper explains the measuring of viscousness of comestible oils utilizing microcontroller-based instrument and the feasibleness in calculation of viscousness of comestible oils at assorted temperatures utilizing nervous web.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

Design strategy

The viscosimeter consists of a Cu cup furnished with a arrow, which ensures a changeless caput and opening at the centre of the base of interior cylinder. The opening is closed with a ball, which is lifted to let the flow of oil during the experiment. The cylinder is surrounded by a H2O bath, which can keep temperature of the sample to be tested at required temperature by the solid-state relay and microcontroller. The microcontroller ( Block F ) with 8 K words of flash programmable and effaceable memory and 368 bytes of RAM. The PIC16F877 microcontroller is a low power, high public presentation RISC CPU 8-bit has two 8-bit and one 16-bit timer/counters, two gaining control, comparators and pulse breadth transition ( PWM ) faculties, a full semidetached house consecutive port, parallel ports, an on-chip oscillator, a programmable codification protection, 14 interrupt beginnings and a 10- spot 8 channel A/D convertor. It is designed for low power ingestion to mensurate the temperature of the sample a chromel alumel thermocouple in block C is used. An instrumentality amplifier kept in block E is designed to magnify ( derive 250 ) the voltage generated by the thermocouple. The parallel temperature in the signifier of electromotive force is given to Port A and it is digitized by the microcontroller. The solid-state power accountant in block G is built with optocoupler ( MOC3040 ) , Triac ( BT136 ) and other constituents are used to command the power of the electric warmer in block H. The pulsation breadth transition ( PWM ) is used to command the temperature. The microcontroller will read the informations for temperature of the sample and command the temperature of bath at the set temperature. The degree sensor in block B consists of IR emitter and sensor, which are used to mensurate 50 milliliter of sample aggregation in the glass jar. The end product from the IR sensor is given to a signal conditioning circuit in block D that provides an input ( Port C ) to the micro accountant. The micro accountant to calculate the viscousness of liquids uses the count value of the timer. Block L consists of keyboard and LCD show, which are interfaced with Port B and Port D of microcontroller. The keyboard is used to come in experimental parametric quantities and measuring in sequences into the system. The LCD show will expose the measured informations and the consequences

  • The opening is opened and the timer in the microcontroller is started and the timer is stopped when the IR sensor senses the degree of sample.
  • Measurement

    Edible oils like sunflower oil are purchased in local commercialism. The rule of operation of redwood viscosimeter is based on measuring of clip required to run out changeless measure of liquid through the narrow capillary tubing. The cylinder of Redwood viscosimeter is filled up to a fixed tallness with sample of sunflower oil. The fit temperature of the sample is given through the keyboard to maintain the temperature of H2O bath at the set temperature. When raising the ball valve, the timer in the micro accountant is started and when the degree in the jar reaches 50 milliliter grade the IR sensor senses the degree and the timer in the microcontroller is stopped. From the count value in the timer clip taken for the aggregation of 50cc is measured and the Kinematic viscousness is computed by the microcontroller utilizing the look? =A*t- ( B/t ) in centistokes ( 1 ) Where A=0.26 and B=172 ( when T & gt ; 34 ) are invariables of the viscosimeter which depends upon the diameter and tallness of cylinder, diameter of opening and the length of opening. ? =Kinematic viscousness of liquid and the t=time required to go through 50cc of liquid. The Experiment is repeated to mensurate Kinematic viscousness of sunflower oil for different temperature runing from 30 & A ; deg ; c to 90 & A ; deg ; degree Celsiuss.

    Nervous Network in Measurement Applications Learning and Testing

    Nervous Network is used to find the kinematic viscousness of comestible oils at assorted temperatures utilizing back extension acquisition. A three-layered nervous web ( ANN ) holding seven nerve cells in concealed bed and one nerve cell in the input bed and one nerve cell in the end product bed is used. The figure of concealed nerve cells is determined through empirical observation. The nerve cells in the hidden and end product beds have sigmoid maps [ 3 ] . The weights between end product and the concealed beds are updated utilizing the pseudo electric resistance control algorithm [ 4 ] . It was found that for utilizing this regulation, convergence is comparatively faster than the original generalised delta regulation. A big? would rush up the convergence ab initio but oscillation tends to change as the mistake increasingly becomes little and therefore it has to be reduced. However alpha and beta are at 0.8 and -0.15 severally [ 5 ] . A Neural Network is trained to temperature as input vector and Kinematic viscousness as end product vector by utilizing the back extension algorithm method. The input and end product vectors which are obtained from the experiments is used for larning. The aim of preparation is to set the weights so that application of a set of inputs produces the coveted set of end products. Before the preparation procedure, the weights are initialized to little random Numberss. Under Supervised acquisition, both inputs and end product informations are given as informations for the preparation. In this procedure, the weights are modified and the system is trained so as to acquire the coveted end product for a given input The preparation form for the input vectors is temperature of the sample and the end product vector is kinematic viscousness. The sigmoid map is implemented for both input and end product to develop the nervous web. Having trained, the nervous web is used to treat the kinematics viscousness densenesss of sunflower oil at assorted temperatures.

    Software is developed in ‘C ‘ linguistic communication to initialise LCD show, to get down ADC, to read informations from ADC, to mensurate temperature of the sample, to command the temperature of bath by giving bids to solid province accountant, to read informations from degree detector, to mensurate the clip of the aggregations of 50cc of liquid with the aid of the degree sensor, to give informations for invariables A and B and to calculate the viscousness and to expose the consequences in the LCD.

    Consequences and Discussion

    The developed system has been used for mensurating kinematic viscousness of fresh ( without heating ) sunflower oil and another sample, which is used in frying status ( 190oc ) . The viscousness is measured at different temperature for both fresh and used samples. Comparing its responses to consequences obtained by rotational type viscosimeter checks the truth and duplicability of the developed instrument. Table 1 shows the kinematic viscousness values of sunflower oil for temperatures runing from 30 & A ; deg ; C to 90 & A ; deg ; C. From the Table it is observed that measured kinematic viscousness of samples lessening with addition in temperature. This is due to a higher thermic motion among the molecules, cut downing intermolecular forces, doing flow among them easier and cut downing viscousness

    In the present survey ANN system is used to gauge the kinematic viscousness of fresh and used sunflower oil at assorted temperatures. The information in the Table 1 is used as the preparation form for used and used sunflower oil. After preparation of the nervous web the weight of the nervous web are saved. They can be reloaded at any clip to prove experiment informations to find the viscousness at unknown temperatures. The Table 2 gives the kinematic viscousness of sunflower oil by the Measurement set up every bit good as computed by the designed nervous web. From the tabular array 2 it is observed that the designed nervous web can calculate kinematic viscousness of sunflower oil at known temperature. The Table 3 gives the kinematic viscousness of fresh and used sunflower oil at assorted unknown temperature computed utilizing nervous web, which are non used for preparation. The mistake in the calculation of kinematics viscousness of sunflower oil utilizing nervous web is found to be less than 1 % . The mistake can be minimized by increasing the figure of preparation rhythms and by altering the figure of nerve cells in the concealed bed. It is observed that even in the absence of measured kinematic viscousness of oil it is possible to obtain dependable matching estimations by trained ANN.

    Statistical analysis:

    Statistical analysis was made utilizing package SPSS version 12. Table3.shows the information analysis studied for the deliberate viscousness ( dependent ) with rise in temperature ( independent ) . It was found that the power map to be the best curve tantrum with minimal mistake if the R sq value approaches 1.It was noted that R sq value ranges from 0.976 to 0.998. Therefore one can reason that our arrested development theoretical account consequence is significantly better anticipation of truth.


    The PIC16F877 micro accountant based instrument system is developed to mensurate kinematic viscousness comestible oils at different temperatures runing from 30 & A ; deg ; c to 90 & A ; deg ; c.. The fluctuation of viscousness at different temperatures will bespeak the non – debasement of the oils samples. The temperature measuring and control system is tested and the mistake in measuring of temperature is found to be within 1 % . The mistake occurred in kinematic viscousness finding by the nervous web construct every bit good as our new hardware system was found less than 1 % . The mistake in the kinematic viscousness finding in the nervous web with mention to the experimental denseness measuring is found to be less than 2 % . The measuring system was configured to run over the temperature scope of 25 & A ; deg ; C to 100 & A ; deg ; C. The system is extremely dependable, easiness of managing, less expensive and portable.


    1. J.C.O.Santos, A.G. de Souza, M.M.Silva Conceriao, M.C.D. , prasad. A thermoalalytic and Kinematic survey of sunflower oil. Brazilian Journal of chemical technology No.3, 2004, pp. 265-273.
    2. M.A Rao, Rheology of liquid nutrients, Journal Tenure surveies, 1997, pp. 135 – 168.
    3. M.A.Eiteman, and J.W Goodrum, Rheology of the triglycerides tricaprion, tricaprylin and tricaprin and of Diesel fuel. Minutess ASAE 1993, pp. 503 – 507.
    4. C.Klotutar, D. Rudan – Tasic and K. Romih, Densities and thermic enlargement coefficients of some vegetable oils. Acta chim. Slov 1997, pp. 45 – 55.
    5. M. Bernard Oliver, M. John, Electronic Measurement and Instrumentation -Cage ( Mc graw hill ) , 1992.
    6. J.C.O Santos, I.M.G. Department of State Santos, A.G. Desow, S. Prasad, and A.V Department of State Santos, Thermal stableness and Kinematic survey on thermic decomposition of commercial comestible oils by thermo gravitation. Journal of Food Science,2002, pp. 1393 – 1398.
    7. PIC 16F877 Microcontroller datasheet, Microchip engineering. pp
    8. M, S. Khalid Omatu, The Neural Network Controller for a Temperature Control System, IEEE control systems, vol. 12, 1992, pp.58-64.
    9. C.A.O. Nascimento, R. Giudici, and R. Guardani, Neural Network Based Approach for optimisation of Industrial Chemical Processes, Comp. Chem. Eng. , 24, 2000, pp. 2303
    10. Kamarthi, S.V. and Pittner, S, Accelerating nervous web developing utilizing weight extrapolation, Neural Network, 12, 1999, 1285 – 1299,

    I'm Niki!

    Would you like to get a custom essay? How about receiving a customized one?

    Check it out