You are on page 1of 12

Maths

Modelling temperature change in a cup of tea

Candidate code: hyc709

1
Table of Content

S.No Content Page Number


1 Introduction to the Topic 3
3 Statement of Task 3
3 Data 4
4 Mathematical Investigation 5
5 Conclusion 11
6 Limitations and scope of further research 12

2
INTRODUCTION: -

In my daily schedule of day, I always drink a cup of tea. I love to sip my tea and

gradually enjoy it while I forget any other issues in life. Perhaps the only problem

faced by a tea lover would be the temperature dropping. Thus, I chose this topic for

my internal assessment as I thought it would be fun to use Maths to help me solve

my daily problems. And when I have a sip of tea in everyday of my life I think that in

winters and summers there is a difference between the surrounding temperature

and the temperature of tea which affects tea’s temperature over time. The

surrounding temperature decides the rate of temperature fall and the temperature

until it reaches and remains same. I explored this topic to know how I can investigate

the change of temperature for a cup of tea and finally I thought of using different

types of modeling to achieve a relevant conclusion.

Statement of task:

Statement of plan: I will measure the temperature change for a cup of tea for an

interval of 5 minutes from the moment it is served until the moment when

temperature gets constant and no change occurs. Then, I will use different types

of modeling by using linear modeling, polynomial graphs and logistic equation to

find different equations for the graph and use it to compare the result of rate of

temperature cool down and to see which model provides most accurate result.

Simple apparatus will be used for this investigation. For this experiment, I will a

cup of tea, a stopwatch and a thermometer.

3
Data:-
Time (Min) Temperature (Celsius℃)
0 80.2
5 72.0
10 64.5
15 59.1
20 54.5
25 51.0
30 47.5
35 45.0
40 42.5
45 40.5
50 38.5
55 36.5
60 35.4
65 33
70 31
75 29
80 27
85 25

Table 1: data obtained by recording the temperature of tea over time.

The initial temperature of tea was 80.2 degree ℃ and I measured the temperature in 5 min
intervals. Now, I will model graphs like scatter diagrams and use linear equation and also the
best fit line to show the variations of tea with temperature and time.

4
Mathematical Investigation

I will start my investigation by plotting the scatter graph of the data. It will give us the brief idea

about the data.

Plotting scatter graph for the data

I used Microsoft excel to plot a scatter graph as shown below with time plotted

as x-axis and temperature plotted on y-axis.

Scatter graph
90
80
70
Temperature

60
50
40
30
20
10
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
time (mins)

Graph 1: scatter graph of the data

The correlation is negative. So the power would also be negative.

Now I will use three different models to find an equation for this scatter diagram and then
find the most suitable one.

5
Plotting linear equation graph

A best fit line will be made and the equation and regression will be calculated through
Microsoft excel only.

90
80
y = -- 0.570x + 69.36
70
Temperature

60 R² = 0.930
50
40
30
20
10
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
time (mins)

Graph 2: linear equation for the graph


Equation 2for the best fit line: = −0.570 + 69.36
Regression ( ) = 0.930

Finding Correlation Coefficient

I found the Pearson’s correlation coefficient using my GDC. It will tell us about the strength of
the correlation between time and temperature.

6
Calculated Correlation coefficient = −0.9648

Observation:

After drawing a best fit line for the scatter diagram, the linear equation and the regression of

line were found. Also the Pearson’s Correlation coefficient was found which suggests that the

correlation is strong negative as the value is negative and very close to -1. Still, the line of best

fit does not fit the data exactly and hence we reject this linear model.

In order to find if there is a more accurate equation for the scatter diagram, a polynomial graph

will be plotted now.

Plotting polynomial equation graph

90
80 2
70 y = 0.006x - 1.0794x + 76.148
R² = 0.9856
Temperature

60
50
40
30
20
10
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
Time (mins)

Graph 3: polynomial equation for the graph

7
2
Equation for the best fit line: = 0.006 − 1.079 + 76.14
2
Regression ( ) = 0.985

Observation:

After drawing a trend line that follows a polynomial equation, the regression of line was found

which suggests that this equation is more suitable to use for this data than a linear equation

because the value is even closer to 1. The polynomial equation is more relevant for the data

than linear equation as the regression for this graph is more than the regression obtained by a

linear equation.

In order to find if there is an even more suitable equation for the scatter diagram, a graph given

by a logarithmic equation will be plotted now.

Plotting logarithmic equation graph

100
90
80
70
Temperature

y = -13.32ln(x) + 90.154
60
R² = 0.9318
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90
Time (mins)

Graph 4: logarithmic equation for the graph

8
Equation for the best fit line: = −13.3 ln( ) + 90.15
2
Regression ( ) = 0.931

Observation:

After drawing a trend line that follows a logarithmic equation, the regression of line was found

to be 0.931 which is just 0.001 more than the regression found by a linear graph. This also gives

a strong positive correlation but not the most relevant for the data obtained till now. The

polynomial equation is still more relevant for the data than this logarithmic equation as the

regression for this graph is less than the regression obtained by a polynomial equation

In order to find if there is an even more suitable equation for the scatter diagram, a graph given

by a logistic equation will be plotted now.

9
Plotting logistic equation graph

90
80
70 -0.013x
y = 72.893e
Temperature

60
50 R² = 0.9873
40
30
20
10
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
time (mins)

Graph 5: logistic equation for the graph


Equation2 for the best fit line: = 72.89 —0.01
Regression ( ) = 0.987

Observation:

After drawing a trend line that follows a logistic equation, the regression of line was found to be

the really close to 1 which suggests a strong relationship between the variables. The regression

for logistic equation is the maximum out of all regression that were found.

10
Conclusion

At first I decided to use only 2 types of equation, linear and polynomial, to investigate which

one is more suitable to model the decrease of temperature of a cup of tea. Then, I decided to

expand my investigation by using another type of equation to give the relationship between

temperature and time, which was logistic equation. The obtained regression for all model

show that the logistic equation is the most relevant and suitable model for the data because

the regression is the maximum out of all and was 0.987 which is very close to 1. This suggests

a strong relation between the variables which is given by use of logistic equation. Therefore,

the equation 3 fits best to the data and can be considered a reliable model for representing

change of temperature of tea. This equation can be used effectively to find the temperature

of a cup of tea after any required time.

Equation for the Regressio


Model type graph n

Linear = −0.570 + 69.36 0.930

Polynomial = 0.006 2 − 1.079 + 76.14 0.985

Logarithmic = −13.3 ln( ) + 90.15 0.931

Logistic
—0.01

= 72.89 0.987

Table 1: Concluding result shown by each model of graph.

11
Limitations and scope of further research

A limitation to this investigation would be that the surrounding temperature depends on a

lots of factors like weather, climate etc. and this investigation is carried out in summers. These

physical factors will affect the rate of cool down and will affect the results obtained by the 3

models. Also, more types of models could have been used to check whether another form of

equation gives even more reliable relationship.

12

You might also like