Proposal for Thermal Management Systems for e-Bike Controllers

Adriano Figueiredo

, José Santos

, Tiago Silva

and Tiago Gândara

Department of Mechanical Engineering, University of Aveiro,

Campus Universitário de Santiago, 3810-193, Aveiro, Portugal

Keywords: e-Bike, Hybrid Cooling System, Thermal Management System (TMS), Active Cooling, Fuzzy Logic.

Abstract: The rise in urban e-bike adoption underscores the need for robust thermal management systems to enhance

the efficiency and reliability of critical components such as motor controllers. This study explores compact,

lightweight, cost-effective cooling systems to maintain optimal thermal conditions under diverse scenarios.

A comprehensive review of cooling technologies, including air, liquid, and hybrid systems, highlights the

advantages of phase change materials and heat pipe-based solutions, particularly in combination with forced

convection. Control systems leveraging fuzzy logic have emerged as the best solution due to their adaptability

and low computational requirements. This research establishes a foundation for integrating innovative thermal

management architectures into next-generation e-bikes, promoting sustainability and improved user safety.

1 INTRODUCTION

In recent years, society has become increasingly

aware of the rise in pollution and the greenhouse

effect. This awareness, combined with legislation on

emissions and the reduction of energy consumption,

has driven the search for more sustainable alternatives

(Wu et al., 2017). Until 2020, the transportation

sector was the second-largest contributor to

greenhouse gas emissions, primarily due to the

dominance of fossil fuels, as approximately 91% of

the energy consumed in this sector originates from

petroleum derivatives (Ritchie et al., 2024; IEA, n.d.).

Consequently, the transition to alternative energy

sources has become inevitable. This shift has been

encouraged by global climate policies, such as the

decarbonization targets established at the United

Nations Climate Change Conference (COP26),

aiming to accelerate the adoption of vehicles that

minimize these emissions (United Nations, n.d.).

This transportation shift prioritizes electric

vehicles, promoting cleaner, efficient mobility, while

e-bikes emerge as practical, sustainable solutions for

urban commuting with growing market value and

production (Wu et al., 2017; Mordor Intelligence,

https://orcid.org/0009-0003-8560-9359

https://orcid.org/0000-0003-0417-8167

https://orcid.org/0000-0003-0203-5488

https://orcid.org/0009-0004-2040-7155

2024). The advantages of this mode of transport

include flexibility, convenient charging (feasible in

any location) and the lack of a licensing requirement,

which facilitates its adoption. Moreover, e-bikes offer

greater ease of use than conventional bicycles. This

makes them particularly suited to densely populated

cities, where they contribute to alleviating traffic

congestion.

E-bikes are similar to conventional bicycles but

feature additional electrical components that provide

pedal-assist functionality. These components include

an electric motor, typically located in the wheel hub

or central region, which assists the rider's movement;

a rechargeable battery that supplies energy to the

motor; and a controller, which manages the power

delivered to the motor based on sensor data, such as

cadence and torque sensors, ensuring both efficiency

and safety. These elements are, therefore, essential

for the proper functioning of this mode of transport,

with various types of controllers available, offering

different levels of complexity and adaptability to

operational conditions.

Advancements in e-bike electronics follow a

miniaturization trend, with rising component density,

as per Moore's Law (Mollick, 2006). This increase in

Figueiredo, A., Santos, J., Silva, T. and Gândara, T.

Proposal for Thermal Management Systems for e-Bike Controllers.

DOI: 10.5220/0013433700003941

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 11th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2025), pages 605-615

ISBN: 978-989-758-745-0; ISSN: 2184-495X

605

component density results in higher heat generation

rates due to the Joule effect, leading to challenges

associated with overheating (Băjenescu, 2021).

Consequently, there is an imperative need to dissipate

this heat to ensure that components operate within

their proper temperature ranges and avoid

overheating, which could result in performance

degradation or equipment damage (Pecht et al.,

1992). According to the literature, approximately

55% of failures in electronic equipment are attributed

to overheating (Xiahou et al., 2018). Nowadays,

while controllers typically feature passive cooling

systems, such as fins, these are often insufficient to

dissipate all the generated heat. Therefore, it is

necessary to develop active cooling systems adapted

to existing controllers, ensuring proper operation and

preventing damage or performance loss.

For implementation in an e-bike controller, the

developed solution must meet essential requirements,

including compactness, low weight, resistance to

impacts and vibrations, low energy consumption,

cost-effectiveness, and ease of maintenance, as well

as protection against contaminants such as water and

dust. A market analysis of controllers revealed that

these components typically achieve efficiencies of

around 90% (Kelly Controls Inc, 2024), with the

power output of the most common e-bike motors

ranging from 250 to 1000 W (Li, 2023), resulting in a

maximum heat output of around 100 W. Furthermore,

the typical dimensions of these components range

from 100 𝑚𝑚 × 60 𝑚𝑚 × 30 𝑚𝑚 to 200 𝑚𝑚 ×

100 𝑚𝑚 × 50 𝑚𝑚.

As e-bikes continue to gain popularity, the

reliability of their components becomes increasingly

critical. Implementing cooling systems will extend

the lifespan of controllers and reduce costs associated

with maintenance and replacement, thereby

promoting broader and more sustainable adoption.

Additionally, temperature regulation of electrical

components prevents failures, safeguarding user

safety. Moreover, improving the efficiency of control

systems will result in more efficient energy

consumption (Ali et al., 2024), aligning with

environmental goals of reducing emissions and

energy usage.

2 LITERATURE REVIEW

E-bike controllers, like other electronic components,

are subjected to varying operating conditions,

including fluctuations in temperature and load,

primarily due to prolonged periods of continuous use.

Such conditions can lead to equipment overheating,

compromising performance, and lifespan reduction,

with critical failures potentially occurring in extreme

cases (Băjenescu, 2021). The literature has explored

and proposed various cooling systems aimed at

improving the efficiency of electrical systems,

ensuring they adequately meet the thermal

requirements demanded during operation (Ali et al.,

2024; Lu et al., 2019; Y. Wu et al., 2024).

Despite the increasing use of e-bikes, the literature

does not provide significant information on cooling

systems for controllers, with the sole exception of

You-Ma Bang et al. (Bang et al., 2016). The available

literature focuses exclusively on cooling solutions for

battery modules in electric vehicles and electronic

components such as Central Processing Units

(CPUs). Therefore, this section will describe the most

used technologies for dissipating heat in electronic

systems, to provide a comprehensive overview of

existing systems. Subsequently, specific solutions

reported in the literature will be examined in detail,

highlighting the advantages and disadvantages of

each approach. Finally, literature addressing thermal

control algorithms will be analyzed, enabling the

adaptation of solutions to operational conditions.

2.1 Existing Technologies

According to the literature (Y. Wu et al., 2024; Lu et

al., 2019), cooling systems can be categorized into

three groups based on the need for external energy

consumption. Active systems require the input of

external energy to create artificial conditions that are

difficult to achieve naturally (Y. Wu et al., 2024).

Conversely, if the solution does not require external

energy and relies solely on environmental conditions

to operate, it is considered a passive technology. In

passive systems, the cooling medium moves through

forces such as capillary action or gravity, among

others, to transfer heat away from the system. Finally,

if the solution combines active and/or passive

technologies, it is classified as a hybrid system. As

shown in Figure 1, the primary passive systems

include phase change materials (PCM), heat pipes

(HPs), and fin-based systems, whereas the most

common active systems utilize air cooling, liquid

cooling, and thermoelectric cooling (TEC) modules.

2.1.1 Air Cooling

Air cooling systems, widely used in electronics, rely

on natural or forced convection for simple and cost-

effective heat management (Y. Wu et al., 2024).

Adding a fan increases air velocity and improves heat

removal efficiency. These systems benefit from their

VEHITS 2025 - 11th International Conference on Vehicle Technology and Intelligent Transport Systems

606

Figure 1: Classification of refrigeration systems, adapted

from (Y. Wu et al., 2024).

simplicity, compact size, lightweight design, low

cost, and elimination of electrical short-circuit risks

due to leaks (Y. Wu et al., 2024). Their effectiveness

depends on airflow velocity, surface geometry, and

airflow-surface interactions (Bergman et al., 2018),

with additional components like deflectors often used

to direct airflow.

2.1.2 Liquid Cooling

Liquid cooling uses a closed circuit where coolant

absorbs and dissipates heat, driven by a pump. Key

components include a heat exchanger (or radiator) for

heat dissipation and a reservoir for fluid storage.

Cooling is achieved either through direct contact—

where components are submerged in the fluid,

enhancing temperature uniformity but posing risks

like short circuits—or indirect contact, where heat is

transferred via heat sinks or cold plates to the coolant

flow (Y. Wu et al., 2024). These systems are

extensively used in industrial production, electronics,

and Battery Thermal Management Systems (BTMS)

due to their superior heat removal efficiency (Y. Wu

et al., 2024).

Their primary advantages lie in their high heat

removal capacity, attributed to the superior thermal

conductivity of liquids compared to air, enabling

greater reductions in the maximum temperatures of

electronic components (Y. Wu et al., 2024). On the

other hand, the main disadvantages of liquid cooling

systems are related to their complexity, which results

in significant space requirements, as well as the

potential for leaks. These leaks must be carefully

mitigated using high-quality seals, which increase the

solution’s overall cost (Lu et al., 2019; Y. Wu et al.,

2024).

2.1.3 Heat Pipes-Based Cooling

HPs are devices capable of efficiently transferring

heat with minimal temperature drops, operating based

on the phase change of the working fluid. These

systems exhibit very high effective thermal

conductivities, surpassing those of many known

metals (Y. Wu et al., 2024), which is advantageous

for heat exchange applications.

Traditional HPs comprise three fundamental

components: the enclosure, the working fluid, and the

wick structure. The enclosure isolates the working

fluid from the environment, ensuring leak prevention

and effective thermal conductivity. The working fluid

facilitates heat transfer, requiring properties such as

high thermal conductivity, high latent heat of

vaporization, low viscosity, and high surface tension

(Y. Wu et al., 2024). The wick structure generates

capillary pressure to drive fluid movement within the

system. These systems are further divided into three

main sections: the evaporator, the adiabatic section,

and the condenser. Heat is introduced at the

evaporator, transferring energy to the working fluid,

causing it to evaporate and increasing the local

pressure, which drives the fluid toward the condenser.

In the condenser, the latent heat is released, causing

the fluid to return to its liquid state. Finally, capillary

pressure and surface tension guide the fluid back to

the evaporator, restarting the cycle (Murshed et al.,

2017). This technology can operate in any orientation,

maintaining capillary pressure sufficient to overcome

gravity when the condenser is positioned below the

evaporator.

There are various configurations of Heat Pipes,

with the most common being Flat Heat Pipes (FHPs)

and Pulsating Heat Pipes (PHPs, also referred to as

Oscillating Heat Pipes - OHPs) (Y. Wu et al., 2024;

Lu et al., 2019). The operating principle of these types

of HPs is the same as that of conventional HPs, with

the primary difference being the geometry of the

solution. FHPs feature a flat surface in the evaporator

region to enhance contact with flat components, while

PHPs consist of a tube bent multiple times into a

serpentine shape. Unlike conventional HPs, PHPs

lack a capillary structure, and the fluid motion is

driven by pressure pulsations between the evaporator

and condenser regions.

2.1.4 PCM-Based Cooling

PCMs are increasingly utilized in Thermal

Management Systems (TMS) due to their high energy

storage capacity, driven by their substantial latent

heat (Murshed et al., 2017). The optimal performance

is achieved when PCMs operate at their phase change

Proposal for Thermal Management Systems for e-Bike Controllers

607

temperature, maximizing heat absorption (Y. Wu et

al., 2024; Murshed et al., 2017). Additionally, they

are primarily classified based on their phase change

type, with solid-solid and solid-liquid PCMs being

the most common due to their stability and integration

ease (Y. Wu et al., 2024). PCMs can also be

categorized as organic or inorganic. Organic PCMs,

such as paraffin, offer high latent heat, good stability,

and no subcooling requirements (the need to cool

below the phase change temperature to solidify) but

have lower thermal conductivity and reduced

stability. In contrast, inorganic PCMs exhibit superior

latent heat and thermal conductivity but are less

reversible and prone to subcooling (Y. Wu et al.,

2024). The ideal PCM should combine high latent

heat, excellent thermal conductivity, and durability

over multiple phase change cycles.

These materials are notable for their high energy

storage capacity, zero energy consumption, low cost,

and compact size. They also mitigate temperature

peaks and promote thermal uniformity (Lu et al.,

2019). However, their main drawbacks are related to

subcooling phenomena, low thermal conductivity,

and limitations in cyclic operations. In cyclic

applications, the stored heat must be dissipated before

the next cycle begins, ensuring that the PCM can

absorb heat again. This prevents PCM from fully

changing phases and losing its properties. Therefore,

for cyclic operations or prolonged usage, an auxiliary

system, such as air cooling, is required to preserve the

PCM’s characteristics (Ling et al., 2015).

2.1.5 Thermoelectric Cooling

Thermoelectric modules utilize two different types of

semiconductors to convert thermal energy into

electrical energy (Seebeck effect) or electrical energy

into a temperature difference (Peltier effect), enabling

their use as coolers (TEC) or heat generators (TEG)

(Lu et al., 2019). Due to the application of electrical

current, a temperature difference is created between

the two sides of the module, resulting in one side

being colder (cold side) and the other side being

hotter (hot side). The main advantages of these

components include a lack of moving parts, no

refrigerants, and compactness (Murshed et al., 2017),

but they suffer from low energy conversion

efficiency, high material costs, and require cooling

for the hot side (Y. Wu et al., 2024; Lu et al., 2019).

Despite being in early research stages, particularly for

BTMS, thermoelectric modules show cooling

potential comparable to liquid systems (Y. Wu et al.,

2024).

2.1.6 Hybrid Cooling

After presenting all the cooling technologies, it is

evident that each has its advantages and

disadvantages. Additionally, it can be concluded that

no single solution, on its own, is fully effective in

addressing the problem of excessive heat.

Consequently, to propose a more robust solution,

these technologies are often combined, leveraging the

advantages of some to mitigate the limitations of

others (Ali et al., 2024). The literature provides

several examples of solutions that integrate different

technologies, some of which are described in the

following sections.

Ziye Ling et al. (Ling et al., 2015) proposed a

hybrid battery cooling system combining PCM

(RT44HC with Expanded Graphite - EG) and forced

convection to address PCM limitations under extreme

conditions, such as complete melting and loss of

latent heat. Comparing forced and natural convection,

the study showed that passive systems performed

adequately under low loads but failed at higher loads,

where insufficient heat dissipation led to PCM

melting and rising battery temperatures. In contrast,

the hybrid system efficiently cooled the PCM,

maintaining temperatures below 46°C (its phase

change temperature) between cycles, enabling full

recovery of latent heat capacity and consistent

performance. Additionally, the system's temperature

uniformity was satisfactory for both the passive and

hybrid solutions, with a maximum temperature

difference of less than 3°C. This uniformity is

attributed to the high thermal conductivity of the

PCM used, which was possible by the addition of EG.

In summary, this study demonstrates the distinct roles

played by PCM and forced convection within the

system. The PCM is responsible for controlling the

maximum temperature and ensuring temperature

uniformity, while forced convection enables the PCM

to recover its energy storage capacity between cycles,

preventing the degradation of its properties.

Although the described study successfully

combined PCM with forced convection, it does not

explore potentially more efficient or cost-effective

alternatives, such as the use of different types of

PCMs or optimizations in airflow design.

Incorporating longer or more variable testing,

including structural design modifications, could

provide additional insights into the practical

feasibility and efficiency of the solution. Finally, this

system offers a simple structure with high efficiency

and reliability, making it a suitable candidate for

application in an e-bike controller.

VEHITS 2025 - 11th International Conference on Vehicle Technology and Intelligent Transport Systems

608

On the other hand, J. Stafford et al. (Stafford et al.,

2012) developed a solution exploring the possibility

of replacing a large fan with an array of smaller fans,

offering greater flexibility. They tested two distinct

configurations: one with two axial fans and another

with three fans, each with a diameter of 24.6

millimeters, arranged in a line. These fans provide

cooling through the impact of air on finned heat sinks.

Additionally, two different scenarios were tested: one

where the air could exit the heat sink in all directions

("4-exit") and another where one direction was

blocked ("2-exit"), simulating a scenario where

thermally sensitive components are present in that

direction. The results showed improved heat transfer

with the "4-exit" setup due to air recirculation, while

"2-exit" reduced performance due to crossflow

effects. The three-fan array outperformed the large

fan by 35% at 9,000 rpm (central fan) and 21% at

7,000 rpm (peripheral fans), though performance

dropped at higher speeds. The two-fan array matched

a single fan up to 10,000 rpm but underperformed

beyond that.

Based on this study, it can be concluded that the

use of multiple fans can offer advantages in terms of

design flexibility, temperature uniformity, and

efficiency for applications with limited space.

Therefore, due to the spatial constraints in e-bikes, the

implementation of smaller fans, rather than a single

larger one, could be a viable solution to enhance

performance without increasing the occupied volume.

Paisarn Naphon et al. (Naphon et al., 2012)

proposed a cooling system for a CPU based on a

Vapor Chamber. This system operates similarly to

Flat Heat Pipes and is composed of two copper plates

(upper and lower), a wick structure that facilitates the

flow of the liquid phase (condensate) back to the

evaporator region (located on the lower plate) through

capillary action, and sinter columns, which ensure

proper fluid distribution within the chamber and

provide internal structural support. These

components ensure efficient fluid circulation,

preventing dry-out (where fluid circulation ceases,

resulting in the absence of fluid in the evaporator and

a consequent temperature increase within the

chamber) and enabling effective heat transfer, hotspot

mitigation, and temperature uniformity. Additionally,

a heat sink was attached to the condenser region,

increasing the heat removal surface area and

enhancing heat dissipation. An axial fan was also

added to further improve energy dissipation, as

illustrated in Figure 2. Compared to a conventional

copper plate system, the Vapor Chamber achieved

6.89% lower CPU temperatures and 10.53% lower

energy consumption for 90% of CPU’s maximum

load, leveraging latent heat for superior performance.

Optimal results were achieved with smaller heat input

areas and sufficient fluid levels, ensuring stronger

capillary action and efficient heat distribution.

Figure 2: Diagram of the solution (Naphon et al., 2012).

This solution can be employed for cooling the

controller of an e-bike due to its structural simplicity,

compact dimensions (60 mm × 60 mm × 3.8 mm),

absence of moving parts, and thermal efficiency.

However, the study could have explored alternative

working fluids to determine which properties yield

the most optimal performance.

Xinxi Li et al. (X. Li et al., 2019) proposed a

cooling system for a lithium battery module utilizing

a thermoelectric cell, with its hot side attached to a

heat sink. Additionally, an axial fan was incorporated

to induce forced convection and enhance heat

dissipation from the heat sink. The results

demonstrated that the proposed system outperformed

both natural and forced convection systems (using

only the fan and heat sink without the thermoelectric

module), achieving lower battery cell temperatures

and reduced rates of temperature increase. For a

battery discharge rate of 3C, the proposed solution

achieved a maximum temperature of 65.43 °C,

compared to 70.525 °C for the system without the

thermoelectric module (using only a heat sink and

forced convection) and 78.30 °C with natural

convection alone. Additionally, the proposed solution

demonstrated improved temperature uniformity.

The tests conducted demonstrate that

thermoelectric modules are effective in reducing the

temperature of electrical components. However, they

should only be employed in extreme situations,

serving as an additional safety measure rather than

operating continuously. Continuous operation would

result in high energy consumption due to the low

efficiency of these devices.

Qingchao Wang et al. (Wang et al., 2015)

proposed a solution for a BTMS consisting of PCM

and an OHP, with paraffin chosen as the PCM due to

its low cost, good thermal storage capacity, and

melting point (≈ 41°C). In the presented solution, the

Proposal for Thermal Management Systems for e-Bike Controllers

609

evaporator of the OHP is located within the PCM,

which is positioned in the space between the battery

cells to remove heat from them. Therefore, the PCM

delays battery overheating by absorbing heat during

phase change, while the OHP, using acetone as the

working fluid (boiling point of approximately 56°C),

transfers heat from the PCM to a water-cooled

condenser. Results showed temperature delay

improvements of 68.36%, 81.33%, 57.92%, and

37.01% at power levels of 20, 25, 30, and 35 W,

respectively. However, as the power increases, this

effect diminishes because PCM's energy storage

capacity is depleted. This situation could be mitigated

if the PCM had a higher thermal conductivity, or if

the OHP started operating at a lower temperature than

the PCM's phase change temperature, thereby

efficiently removing heat from it. This does not occur

because the OHP’s operating temperature coincides

with the boiling temperature of its working fluid,

which is higher than the PCM’s phase change

temperature.

In conclusion, compared to solutions with natural

convection, the proposed solution exhibits lower

maximum temperatures, which is the intended

outcome. However, it is not feasible to implement it

in an electric bike controller due to the need for a

vacuum pump, which is essential to ensure the proper

operation of the OHP. On the other hand, from the

work presented, it can be concluded that this system

was not well designed. Therefore, when used with

PCM, the HPs should have internal fluids with boiling

temperatures lower than the phase change

temperature of the PCM, ensuring that the PCM does

not completely undergo a phase change.

Wayan Nata Septiadi et al. (Septiadi et al., 2022)

proposed a BTMS for e-bike using PCM, traditional

Heat Pipes and forced ventilation. To enhance PCM

conductivity and stability, 20% in mass of expanded

graphite was added to paraffin, increasing liquid

phase stability (as the liquid fraction was retained in

the EG pores due to capillary forces) and thermal

conductivity by 107 times but reducing latent heat

and storage capacity. As shown in Figure 3, the PCM

surrounds the batteries, absorbing their heat and

transferring it to the evaporator, with fins on the

condenser improving heat dissipation. Tests showed

temperature reductions of 1.84°C, 2.69°C, and

6.62°C compared to a battery without a cooling

system, for discharge rates of 0.5, 1, and 1.5C,

respectively, with excellent temperature uniformity.

Forced convection further reduced cell temperatures

and improved uniformity as fan speed increased.

Figure 3: Exploded view of BTMS (Septiadi et al., 2022).

From the analysis of the work described, it can be

concluded that the presented system is effective in

dissipating heat from a battery and could also be

adapted for use in an e-bike controller due to its

compact design. Once again, it was confirmed that

PCM is a viable solution for lowering battery

temperatures while promoting good thermal

uniformity. However, the battery temperature

consistently remained below the phase change

temperature of the PCM composite (≈41.6°C),

suggesting that the PCM may not have undergone a

phase change and instead functioned as a thermal

conductor rather than an energy storage medium. This

hypothesis is supported by the observation that

increased ventilation led to a reduction in battery

temperature—an outcome unlikely if the PCM were

partially melted, as this process is nearly isothermal.

Consequently, if this scenario is accurate, it implies

that for the tested discharge rates, the number of heat

sinks or the fan speed could be reduced. This would

lower costs and energy consumption while ensuring

PCM undergoes a phase change, fully utilizing its

properties.

Finally, Y. Salami Ranjbaran et al. (Ranjbaran et

al., 2023) proposed a hybrid battery cooling system

combining PCM and air cooling to enhance efficiency

with minimal energy use. The design includes PCM-

filled containers between battery cells and air

channels for cooling. Simulations tested nine

scenarios by varying passageways (number of

vertical sections from 2–4) and air pressure (1.1–1.3

atm) to evaluate key parameters. Results showed that

more passageways or higher inlet pressure reduced

battery temperature, which remained below 314K

(≈41°C) in all tests. Additionally, the results revealed

that PCM's performance depends on its melted

fraction. If this fraction is very low, heat conduction

is promoted, which is not the most effective heat

transfer mode for PCM due to its low thermal

conductivity. The optimal PCM performance

occurred with a melted fraction of 15–25% to 60–

70%, as excessive melting caused insulation effects.

VEHITS 2025 - 11th International Conference on Vehicle Technology and Intelligent Transport Systems

610

Temperature differences within the battery were

minimal, with a maximum of 1.57°C for four

passageways.

In conclusion, this solution demonstrates high

versatility, making it easily adaptable for electric bike

controllers. Furthermore, it can be tailored for

systems with higher thermal demands by utilizing

PCMs with greater thermal conductivities, thereby

facilitating faster heat transfer to dissipation sites.

Increasing ventilation power or the quantity of PCM

could also be viable strategies, although these may

introduce challenges related to higher energy

consumption and increased system weight.

Additionally, to enhance heat dissipation, fins can be

incorporated either within the cooling channel or on

other air-exposed surfaces, thereby increasing fluid

contact area and improving heat dissipation

efficiency. In summary, this solution shows

significant potential for electric bike controllers due

to its flexibility.

Additionally, cooling solutions for components of

e-bikes and electric scooters, such as battery modules,

have been explored in the literature. Chandra Sahwala

et al. (Sahwal et al., 2022) developed an algorithm to

control the temperature of various elements in the

powertrain of e-bikes and electric scooters using two

coolant loops and a refrigeration loop. However, this

system is highly complex and spatially demanding, as

it includes pumps for fluid circulation, compressors,

radiators, and condensers, in addition to pipes for

coolant flow. Therefore, such configurations are

excessively intricate and impractical for traditional e-

bikes and electric scooters due to spatial constraints.

Similarly, Jaydeep M. Bhatt et al. (Bhatt et al., 2021)

developed a BTMS based on evaporative cooling to

assess its impact during the charging process of lead-

acid batteries in e-bikes and electric scooters. This

approach has proven to be highly effective in

reducing battery temperature and ensuring a uniform

thermal distribution among the cells throughout

charging. However, as in the previous case, the

system exhibits high complexity and requires

considerable space due to the need for additional

components, such as an air blower and liquid

reservoirs. Consequently, such solutions are

unfeasible for conventional e-bikes and electric

scooters due to space limitations and can only be used

as a plug-in to enhance the charging process.

On the other hand, You-Ma Bang et al. (Bang et

al., 2016) optimized the geometry of a heatsink and

demonstrated the inability of passive systems to

effectively control the temperature of controllers.

Thus, it is crucial to find a balance between spatial

constraints and thermal efficiency, as proposed by

Wayan Nata Septiadi et al. (Septiadi et al., 2022).

2.2 Control Systems

An active system should also include a monitoring

and control system to ensure that the cooling adapts

to operating conditions. In this way, the active system

can increase or decrease its cooling capacity by

adjusting parameters such as fan speed, which helps

reduce energy consumption compared to continuous

operation scenarios.

2.2.1 Existing Control Systems

The literature identifies various algorithms for

controlling the system and ensuring appropriate

operating conditions (Ali et al., 2024). PID

(Proportional-Integral-Derivative) controllers are

popular for their simplicity and effectiveness but rely

on predefined parameters, limiting optimal

performance (Ali et al., 2024). On the other hand,

Fuzzy Logic Controllers (FLC) use linguistic

variables and "If-Then" rules, based on the designer's

experience and intuition, to handle nonlinear systems

and uncertainties, excelling in scenarios lacking

precise mathematical models (Ali et al., 2024).

Advanced approaches include AI-based algorithms

like Deep Learning (DL) and Artificial Neural

Networks (ANN), which improve efficiency and

overall performance but require extensive training

data (Ali et al., 2024). Model Predictive Control

(MPC) dynamically adapts to real-time data, offering

superior performance under dynamic conditions,

being a forward-looking methodology, i.e. oriented

towards the future (Ali et al., 2024). Finally, there are

hybrid systems that combine multiple control

algorithms to increase the efficiency, robustness, and

adaptability of controllers.

2.2.2 Developed Controllers

Shupeng Zhang et al. (Zhang et al., 2023) developed

a battery heating system using an electrothermal film

controller was chosen for its simplicity, low

computational requirements and the lack of a

requirement for prior data compared to Dynamic

Programming (DP), which, although optimal, is

unsuitable for real-time systems. Since the energy

used for heating is drawn from the battery, it is crucial

to establish a balance between energy consumption

and heating to optimize the vehicle's range. The block

diagram of the proposed controller is presented in

Figure 4, showing that the controller incorporates

Proposal for Thermal Management Systems for e-Bike Controllers

611

three inputs: State of Energy (SoE), battery

temperature, and motor load to adjust heating power

dynamically. To evaluate the proposed solution under

dynamic driving cycles, tests were conducted

comparing the proposed FLC-based solution, a

continuous heating system operating at maximum

power, a system without heating, and one using DP

for different initial battery temperatures and SoE

levels. The results revealed that the FLC achieved a

range improvement of 3.6 to 5.3% compared to the

continuous maximum-power heating system and 5.8

to 150.4% compared to the system without heating.

Furthermore, the FLC exhibited a range performance

very close to that achieved with DP, demonstrating

that FLCs can deliver near-optimal solutions while

requiring significantly lower computational power.

Figure 4: Block diagram of used FLC (Zhang et al., 2023).

Yanqi Diao et al. (Diao et al., 2024) designed a

hybrid Fuzzy Logic-PID controller for hypersonic

vehicle nose cooling systems, addressing PID

limitations in adaptability to varying conditions, due

to predefined parameters. Therefore, the FLC

dynamically adjusts PID terms, enabling scenario-

specific adaptation. Compared to a standalone PID

baseline, the hybrid controller achieved similar

performance under static conditions but

outperformed in dynamic scenarios, with faster

stabilization and reduced temperature oscillations.

This highlights the hybrid controller’s robustness and

improved performance through adaptive PID

parameter tuning, showcasing the advantages of

combining control methods for complex systems.

M. A. Hannan et al. (Hannan et al., 2019)

developed an enhanced FLC with a population-based

optimization technique, particularly Particle Swarm

Optimization (PSO), to regulate lithium battery

temperatures, addressing performance degradation

outside the optimal range. PSO was used to optimize

FLC membership functions, which define how inputs

and outputs are converted between real values and

fuzzy variables, as can be seen in Figure 5. The

proposed algorithm outperformed both FLC and PID

controllers in simulations, achieving better

stabilization times (PID: 88 min, FLC: 40 min, PSO-

FLC: 32 min) and lower overshoot. Additionally, it

was observed that the proposed algorithm presented

similar results to the FLC, albeit superior, indicating

that the optimization algorithm contributed to

improved performance over the conventional

controller. Therefore, it is possible to conclude that

the precise definition of membership functions is

critical. However, the results obtained were limited to

simulations and were not experimentally validated.

Additionally, the author could have considered other

metrics, such as the execution time of the algorithms.

Lastly, due to its iterative nature, PSO requires a

model or prior data to evaluate the various potential

solutions within the domain and identify the optimal

one. Therefore, in the absence of such data or models,

the application of PSO becomes unfeasible.

Figure 5: Proposed PSO based FLC algorithm

implementation (Hannan et al., 2019).

3 CRITICAL ANALYSIS

The proposed solution, as outlined in Section 1,

prioritizes compactness and lightweight design,

excluding liquid-based cooling systems due to their

space demands and high maintenance. Air cooling

systems stand out for their simplicity, low cost, space

efficiency, and versatility. They can be combined

with other systems, such as PCM, to enhance

temperature uniformity. Replacing a larger fan with

several smaller ones can increase design efficiency

and flexibility, while also promoting better

temperature uniformity. All these characteristics

make air cooling systems suitable for this application.

On the other hand, PCMs are promising due to their

ability to absorb large amounts of heat, regulating the

temperature of the controller to the phase-change

temperature. Therefore, it is crucial to select a PCM

with a phase-change temperature below the

maximum allowable limit of the component.

However, these materials need to be housed in

enclosures to prevent leakage of the liquid phase and

potential short circuits, complicating their cooling.

Thus, despite the high potential of PCM, precautions

must be taken to ensure it remains within the optimal

liquid fraction range, enhancing the system's

VEHITS 2025 - 11th International Conference on Vehicle Technology and Intelligent Transport Systems

612

performance and preventing a complete phase

change, which could lead to a failure of the cooling

system. Similarly, HPs can also be suitable for the

solution. Traditional ones, due to their circular shape,

can be used in conjunction with other technologies,

such as PCM, to improve heat removal efficiency.

However, on their own, they have low efficiency due

to the small contact area. The use of OHPs is not

viable for this application due to the need for a

vacuum pump, which is too large for this use. Finally,

Flat HPs can be used on flat surfaces due to their

geometry. These systems are effective in conducting

heat from the hot surface to the area where the heat

will be dissipated, with a lower thermal resistance

than that of a copper plate. Thermoelectric modules

enable rapid and controlled cooling, being ideal for

high heat dissipation requirements. However, the

efficiency of these devices is limited and decreases as

the temperature difference between the hot and cold

sides increases, which makes these systems less

effective for prolonged use. Therefore, the use of

thermoelectric modules may be more suitable as a

backup solution, complementing other cooling

systems, particularly in extreme heat situations when

other technologies are unable to remove the required

amount of heat. Finally, fins can be used in any

solution, increasing the contact area with the heat

dissipation medium without increasing the volume of

the system, leading to higher energy dissipation.

Based on the analysis of the literature, PCM

emerges as an ideal solution, ensuring thermal

uniformity and peak absorption. In turn, the cooling

of this material can be carried out through channels,

as proposed by Ranjbaran (Ranjbaran et al., 2023),

with thermoelectric modules also available to assist

cooling in extreme scenarios. This solution can be

enhanced by optimizing the channel flow, through

channel geometry, or by adding fins, enabling higher

heat dissipation. On the other hand, cooling could be

achieved using traditional HPs, with the evaporator

placed inside the PCM. In this case, the heat absorbed

by the PCM would be conducted through the HP and

dissipated in the condenser. Both scenarios would use

fans to promote forced convection, ensuring greater

heat removal capacity. Thus, both described solutions

are compact and can be controlled to reduce energy

consumption, and increase battery autonomy.

Based on preliminary calculations and

considering that the e-bike controller generates 100W

of heat and the battery can only operate continuously

under these conditions for 2 hours, approximately 720

kJ of thermal energy are generated (Eq. 1). To store

this amount of energy and considering a PCM with a

latent heat of fusion (C) of 195 kJ kg

⁄

(pure paraffin

(Wang et al., 2015)), approximately 3.7 kg of paraffin

is required (Eq. 2), equivalent to a volume of

4.5 × 10



m³, assuming a liquid density of 910

kg m



⁄

(Wang et al., 2015). However, as noted, the

molten fraction of the PCM must remain below 70%

to ensure proper operation. Taking this limitation into

account, the required amount of paraffin increases to

5.27 kg (Eq. 3), representing a volume of 5.97 ×



m³ (PCM’s density = 0.7 × 910 + 0.3 ×

822 = 883.6 kg m



⁄

This value represents the

maximum quantity of PCM needed in the absence of

an active cooling system. As discussed, a cooling

system will be incorporated, which will allow for a

reduction in the required PCM volume.

𝐸 = 𝑃 × 𝑡 = 100 W × 2 × 3600 = 720 kJ (1)

720 = 𝑚 × 𝐶 ⟺ 𝑚 = 720 195

⁄

=3.7 kg (2)

𝑚=

 

⁄

.

= 5.27 kg (3)

The control system to be adopted must have a

processing demand that is not excessively high, so it

can be applied in real-time on microcontrollers such

as an ESP32 microcontroller, which have limited

computational and energy resources. Additionally, as

this project is in an early phase without detailed

thermal data or models, control system selection must

reflect these limitations. Iterative, AI-based, or

predictive algorithms are excluded, making PID

controllers and FLC promising options. However, it

would still be difficult to select the ideal PID

parameters to maximize controller performance.

Therefore, in the early stage, the FLC seems to be the

best solution, due to its ease of implementation,

computational efficiency, simplicity, and the fact that

it is based on experience and intuition. Developing

precise rules and membership functions will be key to

optimizing performance, requiring empirical testing

and validation to ensure adaptability to varying

conditions without overloading computational

resources.

After the solution is developed and the control

system is implemented, the tests conducted will

enable data collection that can serve as a foundation

for the development of more complex models. This

data could be used to refine the FLC rules, explore

hybrid controllers such as FLC-PID, or even

implement artificial intelligence to develop more

advanced predictive models. While these algorithms

may increase computational demands, this can

potentially be mitigated by using remote servers to

perform the necessary calculations.

Proposal for Thermal Management Systems for e-Bike Controllers

613

4 CONCLUSIONS

By evaluating various cooling technologies, PCMs

and heat pipe-based systems stand out for thermal

efficiency, compactness, and temperature peak

management. The combination of PCMs with

auxiliary cooling methods, such as forced convection,

further enhances system performance while

maintaining compactness and energy efficiency. In

the literature review, there was found a lack of

information regarding active cooling solutions for

controllers. Thus, the development of this novel

solution will enrich the literature and foster

advancements in this field, particularly in the design

of compact systems suitable for space-constrained

applications, providing a viable alternative to liquid-

based cooling solutions, which are impractical in such

scenarios.

Control strategies, particularly those leveraging

fuzzy logic, were highlighted for their adaptability

and low computational overhead, making them

suitable for real-time application on resource-

constrained microcontrollers. These approaches also

lay the groundwork for future optimization through

advanced algorithms and data-driven models.

By adopting the proposed hybrid cooling system

combining phase change materials with auxiliary

methods such as forced convection, e-bike controllers

can achieve improved thermal management,

significantly extending their operational lifespan and

reliability. Implementing fuzzy logic-based control

strategies ensures that these systems are energy-

efficient and adaptable to dynamic operating

conditions, further enhancing battery life and safety.

This integration not only aligns with global

environmental and sustainability goals by reducing

energy waste and promoting e-bike adoption but also

supports the development of resilient urban mobility

solutions. Consequently, these advancements can

catalyze the shift towards greener, more efficient

transportation networks, delivering tangible benefits

to users, and society.

ACKNOWLEDGEMENTS

This work was developed in the scope of the project

AM2R – “Agenda Mobilizadora para a Inovação

Empresarial do Setor das Duas Rodas” [644866475-

00000012 – Project n. 15], financed by PRR –

Recovery and Resilience Plan under the Next

Generation EU from the European Union, and by the

projects UIDB/00481/2020 and UIDP/00481/2020 -

Fundação para a Ciência e a Tecnologia,

https://doi.org/10.54499/UIDB/00481/2020.

REFERENCES

Wu, W., Yang, X., Zhang, G., Chen, K., & Wang, S. (2017).

Experimental investigation on the thermal performance

of heat pipe-assisted phase change material based

battery thermal management system. In Energy

Conversion and Management, 138, 486–492.

10.1016/j.enconman.2017.02.022

Ritchie, H., Rosado, P., & Roser, M. (2024). In Breakdown

of carbon dioxide, methane and nitrous oxide emissions

by sector. Our World in Data. Retrieved from

https://ourworldindata.org/emissions-by-sector

A. (n.d.). Transport. Retrieved from

https://www.iea.org/energy-system/transport

United Nations. (n.d.). COP26: Together for our planet |

United Nations. Retrieved from

https://www.un.org/en/climatechange/cop26

Mordor Intelligence (2023). "Europe E-bike Market Size &

Share Analysis – Industry Research Report - Growth

Trends. Retrieved from

https://www.mordorintelligence.com/industry-

reports/europe-e-bike-market

Mollick, E. (2006). Establishing Moore’s law. In IEEE

Annals of the History of Computing, 28(3), 62–75.

10.1109/mahc.2006.45

Băjenescu, T. (2021). Miniaturisation of electronic

components and the problem of device overheating. In

Electrotehnica Electronica Automatica, 69(2), 53–58.

10.46904/eea.21.69.2.1108006

Pecht, M., Lall, P., & Hakim, E. B. (1992). The influence

of temperature on integrated circuit failure

mechanisms. In Quality and Reliability Engineering

International, 8(3), 167–176. 10.1002/qre.4680080304

Xiahou, G., Zhang, J., Ma, R., & Liu, Y. (2018). Novel heat

pipe radiator for vertical CPU cooling and its

experimental study. In International Journal of Heat

and Mass Transfer, 130, 912–922.

10.1016/j.ijheatmasstransfer.2018.11.002

Kelly Controls Inc. (2024). KL DC/DC Converter (36V-

180V input) (13.5V output) (30A-60A) - Kelly

Controls. Retrieved from

https://kellycontroller.com/shop/kl-dcdc/

Li, J. (2023). E-bike Controller: the newest 'essential'

component. Retrieved from

https://www.firstcomponents.com/e-bike-controller

Ali, Z. M., Jurado, F., Gandoman, F. H., & Ćalasan, M.

(2024). Advancements in battery thermal management

for electric vehicles: Types, technologies, and control

strategies including deep learning methods. In Ain

Shams Engineering Journal, 15(9), 102908.

10.1016/j.asej.2024.102908

Lu, M., Zhang, X., Ji, J., Xu, X., & Zhang, Y. (2019).

Research progress on power battery cooling technology

for electric vehicles. In Journal of Energy Storage, 27,

101155. 10.1016/j.est.2019.101155

VEHITS 2025 - 11th International Conference on Vehicle Technology and Intelligent Transport Systems

614

Wu, Y., Yang, B., Zhang, X., & Ying, S. (2024). Research

progress in battery thermal management system under

vessel working conditions. In Journal of Energy

Storage, 96, 112761. 10.1016/j.est.2024.112761

Bergman, T. L., Lavigne, A. S., Incropera, F. P., & DeWitt,

D. P. (2018). In Fundamentals of heat and mass

transfer (8th ed.). Wiley & Sons.

Murshed, S. S., & De Castro, C. N. (2017). A critical review

of traditional and emerging techniques and fluids for

electronics cooling. In Renewable and Sustainable

Energy Reviews, 78, 821–833.

10.1016/j.rser.2017.04.112

Ling, Z., Wang, F., Fang, X., Gao, X., & Zhang, Z. (2015).

A hybrid thermal management system for lithium ion

batteries combining phase change materials with

forced-air cooling. In Applied Energy, 148, 403–409.

10.1016/j.apenergy.2015.03.080

Stafford, J., Fortune, F., & Newport, D. (2012). Thermal

performance characteristics of integrated cooling

solutions consisting of multiple miniature fans. In

Journal of Physics Conference Series, 395, 012029.

10.1088/1742-6596/395/1/012029

Naphon, P., Wongwises, S., & Wiriyasart, S. (2012). On the

thermal cooling of central processing unit of the PCs

with vapor chamber. In International Communications

in Heat and Mass Transfer, 39(8), 1165–1168.

10.1016/j.icheatmasstransfer.2012.07.013

Li, X., Zhong, Z., Luo, J., Wang, Z., Yuan, W., Zhang, G.,

Yang, C., Yang, C. (2019). Experimental investigation

on a thermoelectric cooler for thermal management of

a Lithium-Ion battery module. In International Journal

of Photoenergy, 2019, 1–10. 10.1155/2019/3725364

Wang, Q., Rao, Z., Huo, Y., & Wang, S. (2015). Thermal

performance of phase change material/oscillating heat

pipe-based battery thermal management system. In

International Journal of Thermal Sciences, 102, 9–16.

10.1016/j.ijthermalsci.2015.11.005

Septiadi, W. N., Alim, M., & Adi, M. N. P. (2022). The

application of battery thermal management system

based on heat pipes and phase change materials in the

electric bike. In Journal of Energy Storage, 56, 106014.

10.1016/j.est.2022.106014

Ranjbaran, Y. S., Shojaeefard, M., & Molaeimanesh, G.

(2023). Thermal performance enhancement of a passive

battery thermal management system based on phase

change material using cold air passageways for lithium

batteries. In Journal of Energy Storage, 68, 107744.

10.1016/j.est.2023.107744

Zhang, S., Li, T., & Chen, L. (2023). Fuzzy logic control of

external heating system for electric vehicle batteries at

low temperature. In World Electric Vehicle Journal,

14(4), 99. 10.3390/wevj14040099

Diao, Y., Liu, X., Bian, Y., Zheng, J., Zhou, W., & Zhang,

P. (2024). Application of fuzzy PID control algorithm

in hypersonic vehicle transpiration cooling control. In

International Journal of Thermal Sciences, 208,

109457. 10.1016/j.ijthermalsci.2024.109457

Hannan, M. A., Young, Y. S., Hoque, M. M., Ker, P. J., &

Uddin, M. N. (2019). Lithium ion battery thermal

management system using optimized Fuzzy controller.

In IEEE Industry Applications Society Annual Meeting.

10.1109/ias.2019.8912339

Bang, Y., Patil, M. S., Seo, J., & Lee, M. (2016).

Experimental and numerical study on the thermal

performances of battery cell and ECU for an E-Bike. In

Lecture notes in electrical engineering, 195–204.

10.1007/978-3-319-50904-4_19

Sahwal, C. P., Dinh, T. Q., & Sengupta, S. (2022).

Controller development of thermal management system

for electric bikes. In Energy Reports, 8, 437–446.

10.1016/j.egyr.2022.10.135

Bhatt, J. M., Ramana, P. V., & Mehta, J. R. (2021).

Experimental investigation on the impact of

evaporative cooling based battery thermal management

system on charging process of valve regulated lead acid

batteries in E-bike. In Journal of Physics Conference

Series, 2070(1), 012087. 10.1088/1742-

6596/2070/1/012087

Proposal for Thermal Management Systems for e-Bike Controllers

615