In

the liquid flow process industry, the

flow of the liquid change in irregular manner due to the inefficient processes.

As the Flow rate in a process industry depends upon a number of parameter so the

process will not be give the expected output as it is caused by the improper setting of parameters.

The improper parameter settings could threaten the processes. In this paper, we

utilize the Flower Pollination Algorithm methods and ANOVA to obtain the optimum

conditions of a flow process and to gain

the percentage of contributions of each parameter. A verification test was

carried out to inspect among the ANOVA & FPA ,FPA produce the optimum

result than ANOVA .120 sets of data is used for constructing the objective

function by using ANOVA while 18 sets of data are used for the verification

purpose.

Keywords

– Liquid flow process ,Optimization , ANOVA, FPA .

Introduction

.

In

most of the industrial applications, there is a need to calculate the inputs to

a process that will drive its outputs in a desired way and thus achieve some

optimum (desired) goal.In such applications,a mathematical input–output model

for the process is usually derived.The model could be based on the physical

phenomena or available historical input–output data. Once the model is

developed, mathematical techniques can be applied to determine the inputs to

the process that will satisfy a certain given criteria.combustion engines 21–24, two-stage combustor burning ethylene

(doped with ammonia) in air 25, catalytic

distillation 26 and desulphurization of

hot metal and steel 27 those are the

industrial process where the modelling and optimization research have been

conducted.The developed optimization algorithm is tested on a novel flow

thermal sensor whose inputs are the flow velocity and fluid temperature and

output is the voltage measurement.29

present thermal flow sensor has a high sensitivity at low flow rates because of

the non-linear transfer function of the sensor which makes the device

especially suitable for very low flow rates measurements.From the experimental

set up proveides 5 different variables where four inputs (sensor output, pipe diameter,

liquid conductivity ,liquid viscosity ) & single output thats flow rate .An

objective function is constructed with help of the four parameters which makes

this non linear.

2

LITERATURE REVIEW

Liquid

flow optimization is the one of the process where the optimized flow in a

process plant can be achieved from a set of value of the process parameters.

Once the model is developed, mathematical techniques can be

applied to determine the inputs to the process that will satisfy a certain

given criteria.combustion engines 21–24,

two stage combustor burning ethylene (doped with ammonia)in air 25,

catalytic distillation 26 and desulphurization of hot metal and steel 27

those are the industrial process where the modelling and

optimization research have been conducted.An inverse model can be learned by an

artificial neural network to calculates the inputs to a process based on the

forward (input to output) neural net model in 28.An

advantage of the method is that it keeps the forward ANN which is obtained from

the computationally expensive training and can be re-used for other purposes

such as prediction and adaptive control.The developed optimization algorithm is

tested on a novel flow thermal sensor whose inputs are the flow velocity and fluid

temperature and output is the voltage measurement.29

present thermal flow sensor has a high sensitivity at low flow

rates because of the non-linear transfer function of the sensor which makes the

device especially suitable for very low flow rates measurements.The sensitivity

of the measured velocity is approximately 0.3% at low flow velocities and it

increases with velocity to reach 3% at high velocities. The development

of a Fuzzy Temperature compensation scheme (FTCS) for hot wire mass airflow

(MAF) sensor is used to compensate the measurement error occurred by using

Sugeno type FIS for temperature of 60C-100C.It verify the performance of the

proposed hot wire MAF sensor temperature compensation scheme.The effectiveness

of the proposed fuzzy compensation scheme is verified with the estimation error

within only ±1% over full scale value 45.

Real-world

optimization problems are very complex and challenging to solve, and many

applications have to deal with these problems. To solve such problems, approximate

optimization algorithms have to be used, though there is no guarantee that the

optimal solution can be obtained 1. Over the last few decades optimization

algorithms have been applied in extensive numbers of difficult problems.

Several nature-inspired algorithms have been developed over the last few years

by the scientific community 2 4 5. The reproduction of flower is achieved

via the pollination process. Flower pollination can be described as the

distribution processes of pollen through a wide range of pollinators such as

insects, birds, bats and some other animals 7.

The purpose of this

study was to find the optimum conditions of the process since they were

unknown.The application of FPA &

ANOVA method is expected to help reduce the amount of time for which the liquid

flow process produce the optimum output.