PASTEURIZER CONTROL SYSTEM IN A BREWERY FOR BEER PRODUCTION

In Partial Fulfillment

Of the Requirement for the Degree

Bachelor of Science in Chemical Engineering

Submitted by:

Abayan, Reginald R.

Namoc, Henry Darius B.

Tabanag, Hannah C.

Instructor:

Engr. Charlie Villarba

May 2024
I. Introduction

This section presents the objectives, explores the relevant literature, and discusses the

theoretical concepts pertaining to the dynamic system being controlled.

A. Objective

This project aims to develop and implement a customized pasteurizer control

system for brewery operations, optimizing the pasteurization process to enhance beer

production quality and consistency. This involves monitoring and controlling the

temperature for efficient heat treatment to ensure smooth and effective operation

while upholding quality and safety standards for the final product.

B. Background of the Study

Heat exchangers are essential components in various industrial processes,

including the pasteurization of liquids like beer. They facilitate the efficient transfer of

heat between two fluids while ensuring they remain physically separated to prevent

mixing. Plate heat exchangers (PHEs) are commonly utilized in the food and beverage

industry due to their ability to maintain hygiene standards and facilitate cleaning

processes. These exchangers consist of multiple plates arranged to allow for heat

exchange between the fluids flowing on either side of the plates.

Figure 1. Plate Heat Exchanger for Beer Pasteurizer

 The design and configuration of PHEs are essential in determining heat

transfer efficiency, overall process performance and precise temperature control

during pasteurization. Maintaining precise temperature control in this process is

crucial due to its direct impact on product safety. Traditional PID

(proportional-integral-derivative) controllers have been widely employed to regulate

pasteurization temperature by modulating the temperature of the heating medium or

adjusting power input. Even a minor deviation of just 1 ºC can diminish the

effectiveness of pasteurization by up to 25%. As a result, a safety margin of 3ºC is

typically incorporated to ensure optimal results. (Hasting, 1992, Ibarrola et al., 1998,

Negiz et al., 1998a, Khadir and Ringwood, 2003a, Berto and Silveira Junior, 2013).

Mathematical modeling and process simulation are vital tools used to design,

evaluate, and control continuous pasteurization processes. Their primary purpose is to

assess how the process's thermal effects impact the quality and safety attributes of the

product.(Grijspeerdt et al., 2004, Awuah et al., 2007, Kechichian et al., 2012,

Trystram, 2012). A dynamic model of a continuous pasteurization process can also be

useful for the design and tuning of controllers, the study of the effect of the fouling,

the scheduling of the production and cleaning. (Hasting, 1992, Manik et al., 2004,

Cheng and Friis, 2007, Machuda and Urresta, 2008).

II. Theory/Methods/Physics-based Modeling

This section lays the groundwork for grasping the principles, methodologies, and models

utilized in the project. It usually covers theoretical concepts, mathematical models, and

analytical methods utilized for designing and evaluating the control system.

A. Conceptual Framework, Methods or Control Strategies

The temperature control system for a PHEs like a pasteurizer, involves several

stages. Temperature sensors measure incoming and outgoing stream temperatures,

which are compared to a setpoint. Any variance between measured and desired

temperatures is processed by a control algorithm, typically a PID controller, which

adjusts the heat transfer rate to minimize deviations and maintain the setpoint.

Control actions from the algorithm, such as adjusting valves or heaters,

modulate heat exchange in the exchanger to control temperature. Monitoring and

control in pasteurizers rely on various instruments like temperature sensors, flow

meters, pressure transmitters, level sensors, control valves, PLCs, HMIs, and data

loggers.

Feedback control strategies, particularly PID controllers, are commonly

employed in pasteurizers to regulate temperature. PID controllers adjust manipulated

variables based on the difference between measured and setpoint temperatures,

maintaining desired temperatures by modifying heating or cooling mechanisms.

The control loop comprises temperature sensors, PID controllers, actuators

(like valves or pumps), and heating/cooling mechanisms. PID controllers employ

proportional, integral, and derivative control actions to regulate temperature, ensuring

stable control performance through parameter tuning.

 The PID controller enables precise temperature regulation in pasteurizers by

continuously adjusting heating or cooling systems based on measured temperatures,

maintaining temperature setpoints within acceptable limits and ensuring system

stability and optimal performance through effective tuning.

The principle of a heat exchange process using a heat exchanger is visualized

in Figure 1. A feed input at normal temperature that needs to be heated suddenly has

its temperature increased after a hot fluid which may be in the form of gas (steam) or

liquid (water) flows from parallel or opposite direction is allowed to flow into the heat

exchanger. The heat exchanger allowed the transfer of heat from the hot fluid to the

cooler fluid through convection and conduction and possibly radiation.

Figure 1. Basic Heat Exchange Process Principle

Image Source: https://www.rasmech.com/blog/plate-heat-exchanger/

This block flow diagram below illustrates the temperature control mechanism

for a heat exchanger (pasteurizer). Temperature sensors measure the temperatures of

the incoming and exiting streams at the beginning of the system. The measured

temperatures are then contrasted with the target temperature setpoint by the

temperature measurement and comparison block. This block computes the

temperature differential between the desired and measured temperatures.

 Figure 2. Block Flow Diagram of the Chosen Process

A temperature control algorithm, like a PID controller, receives the

temperature variance as input and determines the best course of action to adjust the

heat transfer rate. This algorithm's goal is to minimize the deviation and maintain the

intended temperature setpoint. The control output, which may be a heater or a valve,

receives a control action from the temperature control algorithm and adjusts the heat

exchanger's heat transfer accordingly. This control output controls the flow of heat

input to the system in order to control the temperature.

The heat exchanger receives the controlled output and transfers the required

amounts of heat between the input streams. It ensures that the target temperature is

reached by moving heat from one stream to the next. The output stream is the stream

that exits the heat exchanger at the desired temperature. This output stream is the

outcome of the temperature control system's successful regulation of the heat

exchange process.
B. Process Instrumentation

Temperature Sensors

- Throughout the pasteurization system, temperature sensors precisely measure

the temperature of the beer and the heating medium. These sensors ensure the

beer reaches and maintains the desired pasteurization temperature for the

appropriate amount of time.

Flow Meters

- Flow meters guarantee that both the beer and the heating medium flow at the

intended rates during pasteurization. Maintaining proper flow is crucial for

efficient heat transfer and consistent pasteurization throughout the beer.

Pressure Transmitters

- Pressure transmitters constantly monitor the internal pressure within the

pasteurizer. This feedback allows for maintaining optimal pressure levels

throughout the process, ensuring the safety and efficiency of the

pasteurization.

Level Sensors

- Capacitance or ultrasonic level sensors keep a watchful eye on the liquid

levels within the pasteurizer tanks. Proper liquid levels are essential for

effective pasteurization and preventing overflows that could lead to product

loss.

Control Valves

- Control valves play a critical role in regulating the flow of both the beer and

the heating medium. They adjust automatically based on signals from the
 control system to maintain the desired temperatures and flow rates for

effective pasteurization.

Programmable Logic Controller (PLC)

- The PLC acts as the central nervous system of the pasteurization process. It

receives data from all the sensors and transmits control signals to adjust valves

and other actuators. The PLC ensures all process parameters (temperature,

flow rate, pressure) remain within the setpoints for optimal beer

pasteurization.

Human-Machine Interface (HMI)

- The HMI serves as the operator's window into the pasteurization process. This

user-friendly interface displays real-time data on crucial parameters like

temperature, flow rates, and pressures. The HMI also allows operators to make

adjustments if needed during the pasteurization run.

Data Loggers

- Data loggers play a vital role in recording and storing essential parameters like

temperature, flow rates, and pressures throughout the entire pasteurization run.

This data becomes invaluable for analyzing the process, troubleshooting any

issues, and ultimately optimizing the beer pasteurization process for consistent

quality.
C. Modeling Approach and Parameter Estimations

Figure 3. Block Flow Diagram of a Feedback Control Loop

Derivation of Working Equations

Assumptions:

First-Order Dynamic System

Steady-State Condition

Constant Heat Transfer Coefficients

Overall Mass Balance

𝑑(𝑉)
𝑑𝑡
= 𝐹𝑖𝑛 − 𝐹 𝑤ℎ𝑒𝑟𝑒: 𝐹𝑖𝑛 − 𝑚𝑎𝑠𝑠 𝑓𝑙𝑜𝑤 𝑟𝑎𝑡𝑒 𝑜𝑓 𝑏𝑒𝑒𝑟 𝑖𝑛

𝐹 − 𝑚𝑎𝑠𝑠 𝑓𝑙𝑜𝑤 𝑟𝑎𝑡𝑒 𝑜𝑓 𝑏𝑒𝑒𝑟 𝑜𝑢𝑡

Heat and Energy Balance:

𝑑(𝐸)
𝑑𝑡
= 𝐸𝑖𝑛 − 𝐸 + 𝑄

𝑑(𝐸) 𝑑(𝐻)
𝑑𝑡
= 𝑑𝑡
= 𝐻𝑖𝑛 − 𝐻 + 𝑄 ; 𝑤ℎ𝑒𝑟𝑒: 𝐻 = 𝑚𝐶△𝑇

𝑑(𝐸)
𝑑𝑡
= ρ 𝐹𝑖𝑛𝐶△𝑇𝑖𝑛 − ρ 𝐹𝐶△𝑇 + 𝑄 ; 𝑤ℎ𝑒𝑟𝑒: △𝑇 = 𝑇 − 𝑇𝑟𝑒𝑓

𝑑[𝑚𝐶(𝑇−𝑇𝑟𝑒𝑓)]
𝑑𝑡
= ρ𝐹𝑖𝑛𝐶(𝑇𝑖𝑛 − 𝑇𝑟𝑒𝑓) − ρ𝐹𝐶(𝑇 − 𝑇𝑟𝑒𝑓) + 𝑄

@ 𝑇𝑟𝑒𝑓 = 0 𝑎𝑛𝑑 𝐹𝑖𝑛 = 𝐹

𝑑(𝑇)
𝑚𝐶 𝑑𝑡
= ρ 𝐹𝐶(𝑇𝑖𝑛 − 𝑇) + 𝑄

@ 𝑚 = ρ𝑉

𝑑(𝑇)
ρ 𝑉𝐶 𝑑𝑡
= ρ 𝐹𝐶(𝑇𝑖𝑛 − 𝑇) + 𝑄

𝐷𝑖𝑣𝑖𝑑𝑖𝑛𝑔 𝑒𝑣𝑒𝑟𝑦𝑡ℎ𝑖𝑛𝑔 𝑏𝑦 𝐹, 𝐶 𝑎𝑛𝑑 ρ:

𝑉 𝑑(𝑇) 𝑄
𝐹 𝑑𝑡
= (𝑇𝑖𝑛 − 𝑇) + ρ𝐶𝐹

The Heat and Energy Balance Equation

𝑉 𝑑(𝑇) 𝑄 𝑉
𝐹 𝑑𝑡
= 𝑇𝑖𝑛 − 𝑇 + ρ𝐶𝐹
; 𝑤ℎ𝑒𝑟𝑒: 𝐹
= τ (𝑡𝑖𝑚𝑒 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡)

Dynamic Modeling of a Heat Exchanger

 Figure 4—

Assumption:

Flow rate is constant (flow in = flow out).

Tank is well mixed (Tᵢₙ = T).

Heat is provided by condensing steam

Volume is constant.

Given:

𝐿
𝐹𝑖𝑛 = 𝐹 = 100 𝑚𝑖𝑛
Controlled Variable (CV) = 𝑇

𝑇𝑖𝑛 = 20 °𝐶 Manipulated Variable (MV) = 𝑄

𝑇 = 𝑇𝑟𝑒𝑓 = 60 °𝐶 Disturbance Variable (DV) = 𝑇𝑖𝑛

𝑉 = 1000 𝐿 Set Point (𝑇𝑠𝑒𝑡) = 60 °𝐶

 𝑘𝐽
𝐶 = 2. 91 𝑘𝑔−𝐾

𝑘𝑔
ρ = 1. 004 𝐿

Step 1. Process Model:

𝑉 𝑑(𝑇) 𝑄
𝐹 𝑑𝑡
= 𝑇𝑖𝑛 − 𝑇 + ρ𝐶𝐹
(eq 1)

Step 2: Developing the Transfer Function for the Model:

𝑉 𝑑(𝑇) 𝑄
𝐹 𝑑𝑡
= 𝑇𝑖𝑛 − 𝑇 + ρ𝐶𝐹
(eq 1)

@ 𝑆. 𝑆:

𝑑(𝑇)
𝑑𝑡
= 0, 𝑇(0) = Ŧ, 𝑇𝑖𝑛(0) = Ŧ𝑖𝑛, 𝑄(0) = Ǭ

𝑄
0 = 𝑇𝑖𝑛 − 𝑇 + ρ𝐶𝐹

● Deviation Variables:

𝑇' = 𝑇 − Ŧ where:

𝑇', 𝑇𝑖𝑛', 𝑎𝑛𝑑 𝑄' 𝑎𝑟𝑒 𝑡ℎ𝑒 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠, 𝑎𝑛𝑑

𝑇𝑖𝑛' = 𝑇𝑖𝑛 − Ŧ𝑖𝑛 Ŧ, Ŧ𝑖𝑛, 𝑎𝑛𝑑 Ǭ 𝑎𝑟𝑒 𝑡ℎ𝑒 𝑠𝑡𝑒𝑎𝑑𝑦 𝑠𝑡𝑎𝑡𝑒 𝑣𝑎𝑙𝑢𝑒𝑠.

𝑄' = 𝑄 − Ǭ

𝑑(𝑇)
𝑇' = 𝑇 − 60 ; ρ 𝑉𝐶 𝑑𝑡
= ρ 𝐹𝐶(𝑇𝑖𝑛 − 𝑇) + 𝑄

𝑇𝑖𝑛' = 𝑇𝑖𝑛 − 20 ; 0 = ρ 𝐹𝐶(Ŧ𝑖𝑛 − 𝑇𝑟𝑒𝑓) − ρ 𝐹𝐶(Ŧ − 𝑇𝑟𝑒𝑓) + 𝑄 , @

𝑇'𝑟𝑒𝑓 = 0
𝑄' = 𝑄 − 194. 78 ;

𝐿 𝑘𝑔 1 𝑚𝑖𝑛 𝑘𝐽
Ǭ = (100 𝑚𝑖𝑛
)(1. 004 𝐿
)( 60 𝑠𝑒𝑐 )(2. 91 𝑘𝑔−𝐾
)(333. 15°𝐶 − 293. 15°𝐶)

𝑘𝐽
Ǭ = 194. 78 𝑠𝑒𝑐
= 194. 78 𝑘𝑊

● Eq 1 becomes:

𝑉 𝑑(𝑇') 𝑄'
𝐹 𝑑𝑡
= 𝑇𝑖𝑛' − 𝑇' + ρ𝐶𝐹

● Taking the Laplace Transform, eq 1 becomes:

1 𝐾 𝑉 1
𝑇'(𝑠) = ( τ𝑠+1 )(𝑇 '(𝑠)) + ( τ𝑠+1 )(𝑄'(𝑠)) 𝑤ℎ𝑒𝑟𝑒: τ = 𝐹
𝑎𝑛𝑑 𝐾 = 𝐹𝐶
𝑖𝑛

● Plugging in the Constant Values:

1 𝐾
𝑇'(𝑠) = ( τ𝑠+1 )(𝑇 '(𝑠)) + ( τ𝑠+1 )(𝑄'(𝑠))
𝑖𝑛

1000 𝐿
τ= 𝐿 = 10 𝑚𝑖𝑛 = 600 𝑠𝑒𝑐
100 𝑚𝑖𝑛

1 1 5000 𝐾 𝐾
𝐾= 𝐹𝐶
= 𝐿 𝑘𝑔 𝑘𝐽 1 𝑚𝑖𝑛 = 24347 𝑘𝐽 = 0. 20536 𝑘𝐽
(100 𝑚𝑖𝑛
)(1.004 𝐿
)(2.91 𝑘𝑔−𝐾
)( 60 𝑠𝑒𝑐
) 𝑠𝑒𝑐 𝑠𝑒𝑐

1 1 5000
𝑇'(𝑠) = ( 600𝑠+1 )(𝑇 '(𝑠)) + ( 600𝑠+1 )( 24347 )(𝑄'(𝑠))
𝑖𝑛

1 1 5000 𝐾
𝑇'(𝑠) = ( 600𝑠+1 )(𝑇 '(𝑠)) + ( 600𝑠+1 )( 24347 𝑘𝑊
)(𝑄'(𝑠))
𝑖𝑛

1 0.20536
𝑇'(𝑠) = ( 600𝑠+1 )(𝑇 '(𝑠)) + ( 600𝑠+1 )(𝑄'(𝑠))
𝑖𝑛
 Figure 5

Changes:

𝑆𝑡𝑒𝑝 𝑐ℎ𝑎𝑛𝑔𝑒 𝑜𝑓 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑡𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒, 𝑇𝑖 = 12 °𝐶

𝑆𝑡𝑒𝑝 𝑖𝑛𝑐𝑟𝑒𝑎𝑠𝑒 𝑜𝑓 𝑠𝑡𝑒𝑎𝑚 ℎ𝑒𝑎𝑡, 𝑄 = 60 °𝐶

What are the effects of the changes in the disturbance and the heat supplied?

SCILAB XCOS TUNING

A graphical modeling and simulation tool for dynamic systems is Scilab Xcos. Users

can connect different blocks that represent different system components to create

block diagrams and simulate the behavior of those components.

 Figure 6. Scilab XCOS flow of a pasteurizer

Shown in Figure 6. is a closed-loop Proportional-Integral Control system, The

system's multiplexer (MUX) receives the output and enables it to be plotted on a

graph to observe the system’s response to the step input. The system's output

undergoes stepwise changes as the step input is introduced with the disturbance, and

the feedback loop adjusts to these changes until the desired state is maintained. The

graph generated by the scope illustrates the behavior of the system overtime, showing

its response to the step input and displaying the resulting output.

Figure 7. Scilab XCOS Simulation Results

The graph shown in Fig 7. is a comparison between the resulting output with the step

input over time. At first, the system is unstable due to the introduction of

disturbances, this is depicted by the large oscillations. The output is then measured by

a sensor and the process variable is then compared to the set variable (setpoint), the

error which is the difference between the set variable (SV) and process variable (PV)

is then determined and will be regulated by the Proportional-Integral Controller, The

controller will then assign a control action into the controlled variable (CV) which is

the heater or valve, making adjustments, whether to increase or decrease the amount

of heat transfer into the system. This process will continue until the desired state is

achieved (T = 60°C), as observed in Figure 7., the output’s response stabilized over

time.

SIMULINK

Another powerful graphical modeling and simulation tool for dynamic systems is

Simulink. Like Scilab Xcos, it enables users to construct block diagrams by

connecting functional blocks representing various system components. Integrated

with MATLAB, Simulink provides an interactive environment for the design,

simulation, and analysis of complex control systems.

Figure 8. SIMULINK flow diagram of a pasteurizer

Shown in Figure 8. is a closed-loop Proportional-Integral Control system that is

similar to the simulation that was performed using Scilab XCOS. There were no

changes in the operational parameters. This is to confirm the accuracy of the

simulation results that was done on Scilab XCOS.

Figure 9. SIMULINK Simulation Results

The graph shown in Figure 9. is similar to the simulation results from the Scilab

XCOS.
References:

[1]Carola G.C.C. Gutierrez, Diniz, G. N., & Jorge A.W. Gut. (2014). Dynamic simulation of

a plate pasteurizer unit: Mathematical modeling and experimental validation. Journal of Food

Engineering, 131, 124–134. https://doi.org/10.1016/j.jfoodeng.2014.01.029

[2]Carvalho, G., Leite, A. C. C., Leal, R., & Pereira, R. N. (2023). The role of emergent

processing technologies in beer production. Beverages, 9(1), 7.

https://doi.org/10.3390/beverages9010007