][]

Article

Intelligent Management of Alternative Energy Sources

through a System Based on Fuzzy Logic

Gestión inteligente de fuentes alternativas de energía

mediante un sistema basado en lógica difusa

José Alberto Ledesma ¹∗ Evert de los Rios ¹

Jair Villanueva ²

and Meglys Pérez ¹

1

Faculty of Engineering, Electronics and Telecommunications Engineering Program, Universidad Autónoma del Caribe, Barranquilla,

080020, Colombia; jose.ledesma@uac.edu.co; evert.delosrios23@uac.edu.co; meglys.perez@uac.edu.co

Faculty of Engineering, Mechanical Engineering Program, Universidad Autónoma del Caribe, Barranquilla, 080020, Colombia;

jair.villanueva94@uac.edu.co

2

∗

Correspondence: jose.ledesma@uac.edu.co

Citation: Ledesma, J.; De Los Ríos, E.; Villanueva, J.; Pérez, M. Intelligent Management of Alternative Energy Sources through a System

based on Fuzzy Logic. OnBoard Knowledge Journal 2025, 1, 1. https://doi.org/10.70554/OBJK2025.v01n01.04

Received: 21/01/2025, Accepted: 10/04/2025, Published: 04/06/2025

DOI: https://doi.org/10.70554/OBJK2025.v01n01.04

Abstract: This research explores the application of computational intelligence techniques, speciﬁcally fuzzy logic, in the

management of alternative energy sources such as wind, solar, and fuel cells. It proposes a hybrid energy generation

system and evaluates its performance by considering factors like availability, reliability, and responsiveness to energy

demand and environmental conditions. The study emphasizes the use of fuzzy logic-based control systems to optimize

energy distribution, enabling efﬁcient utilization of renewable energy sources. Furthermore, the management system is

tested under various scenarios to determine how it can adapt to different energy needs. The goal is to demonstrate that

intelligent management can lead to better energy efﬁciency in hybrid renewable energy systems. The computational

tool developed for this purpose simulates the system’s behavior and allows for detailed analysis, making it a valuable

resource for optimizing renewable energy management in diverse contexts, particularly in countries like Colombia.

Keywords: Alternative Energy; Energy Demand; Fuel Cells; Fuzzy Logic; Hybrid Generation; Intelligent Systems; Power

Management; Renewable Energy Systems; Solar Energy; Wind Energy

Resumen: Esta investigación explora la aplicación de técnicas de inteligencia computacional, especíﬁcamente la lógica

difusa, en la gestión de fuentes alternativas de energía como la eólica, solar y las celdas de combustible. Se propone

un sistema híbrido de generación de energía y se evalúa su rendimiento considerando factores como la disponibilidad,

conﬁabilidad y la respuesta a la demanda energética y las condiciones ambientales. El estudio destaca el uso de sistemas

de control basados en lógica difusa para optimizar la distribución de energía, lo que permite una utilización eﬁciente de

las fuentes de energía renovables. Además, el sistema de gestión se prueba bajo diversos escenarios para determinar

cómo puede adaptarse a diferentes necesidades energéticas. El objetivo es demostrar que la gestión inteligente puede

conducir a una mayor eﬁciencia energética en los sistemas híbridos de energía renovable. La herramienta computacional

desarrollada para este propósito simula el comportamiento del sistema y permite un análisis detallado, lo que la convierte

en un recurso valioso para optimizar la gestión de energías renovables en contextos diversos, especialmente en países

como Colombia.

OnBoard Knowledge Journal 2025, 1, 1.

https://revistasescuelanaval.com/obk/

Licensed by Escuela Naval de Cadetes "Almirante Padilla", COL.

This article is freely accessible and distributed under the terms and conditions

of Creative Commons Attribution (https://creativecommons.org/licenses/by/4.0/).

OnBoard Knowledge Journal 2025, 1, 1

2 of 18

Palabras clave: Celdas de Combustible; Demanda Energética; Energía Alternativa; Energía Eólica; Energía Solar;

Generación Híbrida; Gestión de Energía; Lógica Difusa; Sistemas de Energía Renovable; Sistemas Inteligentes

1. Introduction

As human needs grow increasingly complex, so do energy consumption patterns. Awareness of the

environmental damage caused by the human footprint has become a prominent topic of discussion in

contemporary society [13].

Regarding renewable energy generation, a common global challenge is the intermittent and disruptive

nature of these sources. For example, wind energy is only available when the wind is blowing, and photo-

voltaic energy is only available under adequate sunlight conditions [8]. Hydroelectric power, on the other

hand, is considered a reliable source for meeting large-scale, short-term power demands. Its signiﬁcance lies

in its ability to adjust water ﬂow through turbines to match changing loads, thereby reducing reliance on

coal-ﬁred and nuclear plants, which have slower response times [1]. However, hydroelectric power poses

signiﬁcant environmental challenges during infrastructure construction, making it less favorable as a “clean”

alternative compared to solar and wind sources. Research efforts have increasingly focused on mitigating

these weaknesses through sophisticated software solutions and smart grid technologies that enable real-time

power management based on end-user demand.

In Bilacenge, Southwest Sumba, Northeast Timor, Indonesia [8], the use of a smart microgrid integrating

both solar and wind energy sources has been investigated. A demonstration plant was designed around a

battery system and a PLC-based Energy Management System (EMS) capable of compensating for ﬂuctuations

in environmental conditions. The EMS processes data collected from various control units to maintain

balance in the power supply. Its data acquisition system comprises three main components: Smart Meter

Control (SMC), Smart Power Manager (SPM), and Battery Monitoring Unit (BMU), with the SMC providing

bidirectional communication between the supply-demand states and the EMS.

In Europe, the implementation of artiﬁcial intelligence (AI) within smart renewable energy systems

is also under active study. Research envisions intelligent agents powered by smart grid technologies and

AI algorithms that account not only for technical requirements but also social and economic considerations.

By combining data handling, data mining, and machine learning techniques, an integrated AI ecosystem is

expected to improve energy distribution and management through intelligent, adaptive decision making

[20].

In Colombia, the adoption of renewable energy is relatively recent. Hydroelectric power contributes

approximately 60% of national electricity generation, while fossil fuels represent around 30%. Despite these

contributions, signiﬁcant advancements are still needed to establish a strong renewable energy foundation

[4], particularly when considering the structural changes required for the implementation of smart software-

driven energy technologies [14].

This study proposes the design and implementation of a fuzzy logic controller for an alternative energy

plant composed of three generation sources: wind, photovoltaic, and fuel cells. By operating these sources

in parallel, the aim is to enhance supply reliability. Through fuzzy logic-based energy management, the

intention is to optimize system efﬁciency by prioritizing lower-cost renewable sources while activating the

fuel cell only when necessary to meet demand. This work aims to serve as a foundation for future applications

of intelligent software-based technologies within renewable energy systems in Colombia.

In Section 2 , the key contributions of this research are presented, highlighting the design and implemen-

tation of the fuzzy logic-based management system, the integration of a hybrid renewable energy plant, and

the development of a computational tool for evaluation. Section 3 discusses the state-of-the-art in renewable

energy systems, focusing on combinations of source types, control software, and smart power grids. Section

4 details the rationale for selecting wind energy as the primary focus, provides a descriptive model of the

proposed plant, and outlines the parameters used to assess the model’s efﬁciency. Section 5 presents an

overview of the software interface along with descriptions of its key features and functionalities. Section 6

OnBoard Knowledge Journal 2025, 1, 1

3 of 18

synthesizes the overall performance and behavior of the proposed model based on the experimental results

and discusses expectations for practical implementation and suggests potential strategies to further optimize

system performance.

2. Contributions

This study presents several novel contributions to the intelligent management of alternative energy

sources through a system based on fuzzy logic:

i.

Development of a Fuzzy Logic-Based Management System (GLD): A key contribution of this research

is the design and implementation of a Mamdani-type fuzzy inference engine that dynamically

manages three clean energy sources—wind, solar photovoltaic, and fuel cells. The system uses

environmental and demand variables to optimize energy distribution, enhancing the autonomy and

reliability of the hybrid power plant.

ii.

Integration of a Hybrid Renewable Energy Plant: The proposed model integrates wind turbines,

monocrystalline solar panels, and solid oxide fuel cells, each modeled according to real-world

parameters. This hybrid approach leverages the strengths of each energy source to ensure consistent

power supply under varying environmental conditions.

iii.

iv.

v.

Scenario-Based Management Strategies: The study deﬁnes and implements multiple management

scenarios: Proﬁtability, Reliability, Priority, and Percentage, to showcase the system’s ﬂexibility in

adapting to different operational goals. Each scenario allows customized decision-making for energy

distribution based on user-deﬁned criteria or system performance.

Creation of a Computational Tool with Graphical Interface: A user-friendly software tool was devel-

oped to simulate and evaluate the system. This tool allows users to deﬁne subsystem parameters,

select demand proﬁles based on real-time criteria, and observe system behavior through intuitive

graphical outputs.

Adaptation to Colombian Context: The research contextualizes the model for Colombia’s energy ma-

trix, which currently depends heavily on hydroelectric and fossil fuel sources. By demonstrating the

feasibility and advantages of intelligent renewable energy management, this study sets a foundation

for future implementation of smart grid technologies in the country.

vi.

Validation through Simulation: The proposed model is tested across different demand scenarios

using historical data and environmental conditions, validating its capability to balance energy supply

efﬁciently, minimize reliance on costly sources, and respond adaptively to ﬂuctuations.

3. Related Works

The state-of-the-art research in alternative energy management covers various dimensions, including

technological advancements, grid integration challenges, and optimization strategies.

In [19], Zhao et al. propose a novel hybrid system combining solar-assisted methanol reforming and fuel

cell power generation to address the hydrothermal management limitations of proton exchange membrane

fuel cells (PEMFC). By using methanol both as a coolant and as a source for hydrogen production via

solar-assisted reforming, the system achieves enhanced efﬁciency. Simulation results demonstrate controlled

temperature, improved power density, and signiﬁcant increases in energy and system efﬁciency compared to

conventional approaches.

The integration of renewable energy sources into smart grids presents considerable challenges, as

discussed by Khalid in [10]. Smart grid technologies are essential in facilitating this transition by automating

and digitizing grid operations, thereby enhancing ﬂexibility and reliability. The review highlights the

importance of addressing user acceptance, operational ﬂexibility, and regulatory frameworks to fully realize

the potential of smart grids in accommodating variable renewable energy sources and transforming electricity

markets.

Yan et al. [17] focus on the operational costs caused by wind power ﬂuctuations, employing a data-driven

model to accurately estimate power system operating costs under various wind scenarios. By clustering

different ﬂuctuation patterns and leveraging deep neural networks, the study offers insights into optimizing

OnBoard Knowledge Journal 2025, 1, 1

4 of 18

system operations and mitigating additional costs, demonstrating high accuracy and effectiveness in cost

analysis.

Zhang et al. [19] address the optimization of wind power consumption in district heating systems by

proposing a mathematical model and scheduling strategy that maximizes wind power utilization while

maintaining system reliability. Taking into account the dynamic characteristics of cogeneration components,

including thermal inertia and load demand, the study provides practical solutions for efﬁcient integration of

wind power into heating systems.

Renewable energy supply chain management is explored by Sarkar and Seo [15], who model the

conversion of waste into renewable energy to sustainably meet increasing demand. Their energy supply

chain model encompasses waste collection and energy generation, considering scenarios with both unequal

and equal power distribution among supply chain participants. Using game theory and coordination policies,

the research demonstrates how cooperation maximizes proﬁts and underscores the importance of ﬂexibility

and automation in production systems.

Taken together, these studies reveal a thorough exploration of methods to optimize alternative energy

management, addressing technological, operational, and economic challenges to facilitate efﬁcient integration

into existing energy systems.

4. Methodology

4.1. Selection of Alternative Energy Sources

In selecting alternative energy sources for evaluation in the fuzzy logic-based management system,

considerations were made regarding technological development, cost, and future prospects within the renew-

able energy sector. As previously mentioned, wind energy is widely adopted in China due to its simplicity,

efﬁciency, and rapid return on investment [18], making it highly competitive compared to combustion-based

generation methods. Wind energy was chosen in this study because of its technological maturity and, in most

cases, cost-effective implementation.

Although solar photovoltaic energy involves a relatively high initial investment, ongoing scientiﬁc and

economic efforts continue to improve its viability and prominence [6]. Predicting the performance of such

installations is feasible, thanks to existing studies on solar irradiation patterns.

Lastly, hydrogen-derived energy used in fuel cells, while not entirely new, has gained considerable

momentum recently, particularly in the automotive sector. Its appeal lies in its relative simplicity and

controllable power generation, which depends on hydrogen ﬂow within the system. This energy source,

combined with the autonomy of photovoltaic systems and the reliability and continuity of fuel cell systems,

offers promising beneﬁts [5;7].

This research employs generic models of a vertical-axis wind turbine, commonly used in wind farms;

monocrystalline silicon solar panels, prevalent in medium- to large-scale photovoltaic installations; and

solid oxide fuel cells (SOFC), selected for distributed generation. The core contribution of this work lies in

the integration of these alternative energy sources, harnessing their individual strengths synergistically to

mitigate challenges that arise when operating in isolation. Consequently, the system achieves a balance by

combining the robustness and high power capacity of the wind system, the storage capacity and autonomy

of the photovoltaic system, and the reliability and continuous operation of the fuel cell system.

4.2. Alternative Energy Plant Modeling

Figure 1 illustrates the block diagram of the alternative energy plant, which includes its subsystems of

wind, photovoltaic, and fuel cell, for which energy management is desired.

The arrows illustrate the types of communication between the systems, showing that the fuzzy logic-

based management system (GLD) receives information from the wind generation system (SGE), photovoltaic

system (SGF), and fuel cell system (SGC). Simultaneously, it sends back information regarding the percent-

age of their required power, indicating bidirectional communication. The management system maintains

unidirectional communication only with the load, from which it requests data on electrical demand. After

OnBoard Knowledge Journal 2025, 1, 1

5 of 18

Figure 1. Block diagram of the proposed system.

Source: The authors.

completing the management process, the energy supplied by each system is aggregated through the summing

block, with the goal of meeting the electrical demand of the load (DEC).

4.2.1. Wind Generation System (SGE)

The Wind Generation System (SGE) primarily consists of the wind turbine model [20], which includes

parameters such as the number of turbines, blade length, high and low cut-in speeds, power coefﬁcient,

generator efﬁciency, and multiplier efﬁciency. Together, these factors determine the system’s nominal power.

However, the available energy depends on the wind speed incident on the turbine blades, which will be

modeled based on different wind proﬁles, as detailed in Table 1.

Not all the energy is captured by the generator due to losses during the energy conversion process,

associated with both mechanical and electrical components. Therefore, it is crucial to consider the mechanical

multiplier efﬁciency (η_mec) and the electrical generator efﬁciency (η_elec).

For modeling the incident wind on the system, wind standards provided by the meteorological service

of Ernesto Cortissoz International Airport in Barranquilla, Colombia, were used. These standards are

summarized in Table 1.

Table 1. Wind Speed Classiﬁcation

TYPE

Breeze

Light

SPEED KM / H

1 – 5

6 – 11

SPEED KNOTS

0.5 – 2.7

3.2 – 5.9

Gentle

Moderate

Fresh

Strong

Gale

12 – 19

20 – 28

29 – 38

39 – 49

50 – 61

62 – 74

6.5 – 10.3

10.8 – 15.1

15.7 – 20.5

21.1 – 26.5

27.0 – 33.0

33.5 – 40.0

Storm

Source: The authors.

Wind system model structure can be seen in Figure 2. This shows as well the inputs and outputs of the

wind system.

OnBoard Knowledge Journal 2025, 1, 1

6 of 18

Figure 2. Block diagram of the proposed system.

Source: The authors.

4.2.2. Photovoltaic Generation System (SGF)

The Photovoltaic Generation System (SGF) comprises solar panels and a battery bank, as detailed in

studies such as [ ] and [11]. Key parameters, including the number of panels, panel power, battery capacity,

3

voltage, and losses in converters and self-discharge, play an integral role in determining the system’s nominal

power. However, the actual energy output depends on incident solar irradiation and the stored energy within

the battery bank. The modeling process accounts for overall system efﬁciency, incorporating losses from

components such as the battery bank, power converter, and self-discharge. Autonomy time is inﬂuenced by

battery capacity, nominal voltage, and the power generated by the panels, as discussed in [16].

Calculations of available energy require ensuring that energy production sufﬁciently compensates for

system losses so that usable energy exceeds these losses. Nominal energy is computed considering factors

such as the number of panels, energy provided per panel, and the peak solar hours at the installation site.

Solar irradiation, which measures the power received per unit area from sunlight, plays a crucial role, as

explained in [2]. Modeling approaches include varying weather conditions: sunny, cloudy, and rainy, using

solar irradiation data to reﬂect changes in irradiation levels.

Figure 3 illustrates the photovoltaic system model structure, outlining its inputs and outputs, thereby

providing a visual representation of the system’s conﬁguration and operational dynamics, as discussed in the

referenced studies.

4.2.3. Fuel Cell Generation System (SGC)

The Fuel Cell Generation System (SGC) is based on the Solid Oxide Fuel Cell (SOFC) model, commonly

used in distributed energy systems, as described in [9]. This model considers parameters such as the number

of fuel cells and their intrinsic characteristics, including utilization factor, activation losses, ohmic losses, and

concentration losses. These factors, combined with hydrogen ﬂow, determine the system’s nominal power,

which remains available as long as the system operates.

OnBoard Knowledge Journal 2025, 1, 1

7 of 18

Figure 3. Block diagram of the photovoltaic system.

Source: The authors.

In the initial modeling phase of the SOFC, accurately accounting for the chemical reactions within the

cell is critical. Not all hydrogen entering the cell reacts with oxygen; therefore, a utilization factor coefﬁcient

is introduced. This coefﬁcient increases with the amount of hydrogen undergoing the reaction and decreases

inversely with the total hydrogen input to the cell.

Calculating the power output of the fuel cell system requires knowledge of the cell’s DC voltage. Given

the inherent losses within the fuel cell, these must be incorporated into calculations in accordance with

Kirchhoff’s laws. The research treated the available power of the SOFC as a controlled variable dependent

on hydrogen ﬂow, which is assumed to be constant. Figure 4 illustrates the model structure of the fuel cell

system, providing an overview of its inputs and outputs.

4.2.4. Electric Demand Models (DEC)

The modeling of Electric Demand (DEC) is based on accumulated knowledge of electricity consumption

patterns in the traditionally recognized residential, commercial, and industrial sectors. This demand model is

developed using historical data obtained from [12], which characterize these demand types.

In the residential sector, demand peaks occur during family-oriented hours, speciﬁcally between 7:00

PM and 9:00 PM.

The analysis also revealed that the commercial sector experiences its peak consumption during typical

working hours, approximately from 8:00 AM to 8:00 PM, reaching up to 10,000 kW around 10:00 AM.

The industrial sector shows a longer duration of high energy consumption due to the continuous and

energy-intensive nature of industrial processes, with demand peaking at about 4,000 kW around 1:00 PM.

OnBoard Knowledge Journal 2025, 1, 1

8 of 18

Figure 4. Block diagram of the fuel cell system.

Source: The authors.

4.2.5. Fuzzy Logic-Based Management System (GLD)

For the design of the Fuzzy Logic-Based Management System (GLD), the general structure of fuzzy logic-

based systems (Figure 5) was used as a reference model. This ﬁgure illustrates the fundamental components

of such systems. In this research, a Mamdani-type model was employed.

The system includes an input block containing the most relevant information about the system to be

controlled. In this case, the inputs consist of meteorological variables such as wind speed and solar irradiation,

the available hydrogen quantity from the fuel cell system, and the electrical demand curve (DEC) that the

system aims to satisfy.

These GLD inputs are converted into fuzzy values using input membership functions, which can be of

various types. For this study, triangular and trapezoidal membership functions were selected due to their

simplicity and practical implementation.

The inference block uses a set of “if-then” rules that deﬁne the conditions the GLD must meet. The

output block employs output membership functions, also triangular and trapezoidal, that convert the fuzzy

decisions into speciﬁc numerical values using the center of gravity method.

In this research, the outputs correspond to the autonomy percentages (demand satisfaction) for each

subsystem (SGE, SGF, and SGC) relative to the electrical demand curve. Once the autonomy values for each

system are determined, simple management rules are applied to coordinate system operation under different

scenarios, as described later.

4.3. Inputs for the Fuzzy Logic-Based Management System (GLD)

In the design of the GLD, the operational variables of the hybrid alternative generation plant were used

as inputs. These variables include wind speed, solar irradiation, and the available quantity of hydrogen, all

of which directly affect the amount of electrical power each system can generate. Together with the electrical

OnBoard Knowledge Journal 2025, 1, 1

9 of 18

Figure 5. Block model of the proposed GLD.

Source: The authors.

demand, these variables are fed into the GLD and converted into fuzzy values using membership functions,

as illustrated in Figures 6 and 7.

The membership function for the input from the Wind Generation System (SGE) is deﬁned based on the

parameters described in Table 1, which outlines wind types and their corresponding speeds. With appropriate

adjustments to the measurement system, the same methodology can be applied to the Photovoltaic Generation

System (SGF), which is modeled considering the daily solar radiation curve, and to the Fuel Cell Generation

System (SGC), where the input is determined by the hydrogen storage level, expressed as a percentage of the

fuel content in the tank.

Figure 6. GLD input membership function (WIND SPEED).

Source: The authors.

The membership function for the Electrical Demand Curve (DEC) input is normalized with respect to

the maximum and minimum values of the demand curve. This normalization ensures that the fuzzy values

derived from electrical demand are bounded and can be applied to any type of demand curve.

4.3.1. Inference Engine of GLD

The inference engine of the Fuzzy Logic-Based Management System (GLD) operates using a set of

“if-then” rules designed to optimize the operation of the proposed alternative energy generation plant. The

system aims to:

OnBoard Knowledge Journal 2025, 1, 1

10 of 18

Figure 7. GLD entry membership function (DEMAND).

Source: The authors.

•

Fully satisfy the Electrical Demand Curve (DEC), thereby increasing the autonomy and reliability of the

proposed system.

Determine the required power output for each subsystem of the alternative energy plant based on the

selected management scenario, providing high ﬂexibility and efﬁciency.

Permit the use of energy from the electrical grid only when the hybrid alternative energy system is

unable to meet the DEC.

To achieve these objectives, a rule base was developed based on accumulated operational experience

from each generation subsystem. Each subsystem has an associated fuzzy inference engine that calculates the

autonomy or demand satisfaction level speciﬁc to that subsystem. This inference engine estimates, based on

fuzzy variables related to each subsystem (wind speed, solar irradiation, and hydrogen level) and the inherent

capacity of each subsystem (deﬁned by the model parameters), the percentage of demand satisfaction for

each subsystem.

This enables the GLD to determine, through straightforward rules and according to the speciﬁed

management criteria, the operational percentage for each subsystem. For the Wind Generation System (SGE),

16 rules were implemented, using the membership functions depicted in Figures 6 as inputs and Figure 9 as

the output. Figure 8 illustrates the response surface, showing the relationship between demand and wind

inputs and the SGE’s demand satisfaction percentage output.

Figure 8. Work area generated by the GLD for the satisfaction of the SGE.

Source: The authors.

OnBoard Knowledge Journal 2025, 1, 1

11 of 18

Observing Figure 8, the system’s behavior can be described by the rules associated with the Wind

Generation System (SGE). These rules are as follows:

•

If WIND is CALM and DEMAND is HIGH, then SATISFACTION is NONE.

If WIND is CALM and DEMAND is MEDIUM, then SATISFACTION is LOW.

If WIND is CALM and DEMAND is LOW, then SATISFACTION is MEDIUM.

If WIND is GENTLE and DEMAND is HIGH, then SATISFACTION is LOW.

If WIND is GENTLE and DEMAND is MEDIUM, then SATISFACTION is MEDIUM.

If WIND is GENTLE and DEMAND is LOW, then SATISFACTION is HIGH.

If WIND is LIGHT and DEMAND is HIGH, then SATISFACTION is MEDIUM.

If WIND is LIGHT and DEMAND is MEDIUM, then SATISFACTION is HIGH.

If WIND is LIGHT and DEMAND is LOW, then SATISFACTION is VERY HIGH.

If WIND is MODERATE and DEMAND is HIGH, then SATISFACTION is HIGH.

If WIND is MODERATE and DEMAND is MEDIUM, then SATISFACTION is VERY HIGH.

If WIND is MODERATE and DEMAND is LOW, then SATISFACTION is TOTAL.

If WIND is STRONG and DEMAND is NOT HIGH, then SATISFACTION is TOTAL.

If WIND is STRONG and DEMAND is HIGH, then SATISFACTION is HIGH.

If WIND is VERY STRONG and DEMAND is NOT HIGH, then SATISFACTION is TOTAL.

If WIND is VERY STRONG and DEMAND is HIGH, then SATISFACTION is VERY HIGH.

Similarly, to determine the satisfaction level of demand for the Photovoltaic Generation System (SGF),

11 rules were implemented, taking into account the membership functions and comparing demand and solar

irradiation inputs with the SGF satisfaction percentage output. Likewise, to assess the demand satisfaction

for the Fuel Cell Generation System (SGC), 9 rules were established.

4.3.2. GLD Outputs

In the design of the Fuzzy Logic-based Management System (GLD), the outputs were taken as the

autonomy or percentage of satisfaction of the DEC for each subsystem. Figure 9 displays the membership

functions used to determine the output or response of the GLD. This is similarly employed to obtain autonomy

in each subsystem.

Figure 9. Membership function of the outputs to the GLD.

Source: The authors.

As can be seen, the GLD output provides a percentage from 0 to 100% of the demand satisfaction that

each subsystem can handle on its own.

4.4. GLD Management Scenarios

For this research, the management scenarios were deﬁned as follows. These scenarios are designed to

demonstrate the ﬂexibility of the GLD in achieving different management objectives. The scenarios are:

•

PROFITABILITY: In this scenario, the objective is to meet the demand using the most cost-effective

subsystem. Currently, wind generation systems are more economical than photovoltaic systems, which

in turn are more cost-effective than fuel cell systems, considering both efﬁciency and operational costs.

OnBoard Knowledge Journal 2025, 1, 1

12 of 18

•

RELIABILITY: Here, the management system prioritizes the subsystem generating the highest power

output for a given Electrical Demand Curve (DEC).

PRIORITY: In this mode, the operator speciﬁes the order of energy supply from each subsystem according

to their own criteria.

PERCENTAGE: This selection requires the operator to set maximum operation percentages for each

subsystem. It should be noted that the plant as a whole can deliver between 0% and 100% of the power

required by the DEC.

5. Implementation

5.1. Computational graphical tool

To implement the proposal, a computational tool with a graphical user interface was developed. This

tool incorporates models for the subsystems SGE, SGF, and SGC, including all adjustable parameters and

fuzzy input variables necessary for managing the hybrid alternative energy generation system. The tool’s

main features are:

i.

Selection of Demand Curve: The user begins by selecting the demand curve on which the system will

operate. This curve is generated through a fuzzy inference system that takes inputs such as sector,

month, day, and time to produce the corresponding demand proﬁle for those parameters.

Subsystem Parameters: Users can input the operational parameters for each subsystem (SGE, SGF,

and SGC), which are then used to calculate the nominal power of each. Users can also select which

subsystems are enabled for management. With three systems available, seven possible subsystem

combinations exist, but at least one subsystem must be enabled alongside the grid for the management

system to function.

ii.

iii.

iv.

Execution and Analysis: The tool can be executed for a speciﬁc point on the demand curve or, if

desired, can analyze the entire curve multiple times, enabling the generation of various scenarios for

the same demand proﬁle.

Graphical User Interface: The interface provides controls for selecting energy sources, monitoring

system management according to speciﬁed criteria, and evaluating results. It allows for evaluating

individual cases or performing batch analyses for a given daily demand curve. Results controls enable

selection of response curves representing electrical power for each system and the demand. Graphs

display the percentage of demand satisfaction and the operation percentage determined by the GLD

for each subsystem and the grid, based on the chosen management criteria.

The tool (Figure 10) offers a comprehensive platform for experimenting with and analyzing the perfor-

mance of the GLD under various scenarios and input conditions.

5.2. Obtaining electricity demand curves

The implemented system for this research establishes and weights the demand curve on which it

operates. This curve incorporates essential temporal variables, including sector, season, day of the week, and

time of day. Different combinations of these inputs produce various demand curves used for management

purposes. These curves are derived from historical demand data corresponding to the speciﬁed sectors.

Additionally, the system allows users to input a predeﬁned demand curve for further analysis. The

manager can work with any established curve and determine the appropriate management strategy for each

scenario. The resulting electric demand curves conform to the format described earlier in this chapter.

Figure 11 shows the control panel used to enter parameters for generating the demand curve. Each

parameter features a dropdown menu containing descriptive options, including:

•

Sector: Residential, commercial, and industrial.

Season: Summer, spring, autumn, and winter.

Day of the week: Beginning of the week, midweek, and weekend.

Hour of the day: Early morning, morning, afternoon, and evening.

OnBoard Knowledge Journal 2025, 1, 1

13 of 18

Figure 10. Interface implemented for the evaluation of the GLD.

Source: The authors.

Users can select values for each parameter, and the system generates the corresponding electric demand

curve based on historical usage patterns.

Figure 11. Electrical demand control panel.

Source: The authors.

5.3. Rated power of the alternative energy plant

Using the models described in the previous chapter, the nominal power of each subsystem is calculated

based on its operating parameters. The actual operating power of each subsystem depends on the values of

its respective operating variables. For the wind subsystem, this variable is wind speed; for the photovoltaic

subsystem, it is solar irradiation; and for the fuel cell subsystem, it is the hydrogen level. The combined

nominal power of these three subsystems constitutes the nominal power of the alternative energy plant.

OnBoard Knowledge Journal 2025, 1, 1

14 of 18

Figures 12, 13, and 14 show the control panels for each subsystem, as detailed in Section 4.2 of this

chapter. These panels display the various parameters that determine the nominal power of each subsystem.

Figure 12. Wind system control panel.

Source: The authors.

Figure 13. Photovoltaic system control panel.

Source: The authors.

OnBoard Knowledge Journal 2025, 1, 1

15 of 18

Figure 14. Fuel cell system control panel.

Source: The authors.

Figure 15 illustrates the hybrid system as a whole. The nominal power of each subsystem is displayed

based on the parameters shown in the previous ﬁgures. Additionally, this panel allows for manual adjustment

of the fuzzy inputs for each system, using the descriptors for the GLD inputs as described in Section 4.3.

Figure 15. Hybrid alternative generation plant selection panel.

Source: The authors.

5.4. Management of the energy generated by the plant

The management of the energy generated by each subsystem of the plant is handled by the GLD. Using

the fuzzy logic inference engine described in Section 4.3, the GLD determines the demand satisfaction level

for each subsystem based on the input random variables. It then speciﬁes the percentage of electrical power

required from each subsystem according to the selected management scenario.

6. Conclusions

After implementing a hybrid electrical power generation system model based on three clean alternative

sources and developing a management system capable of optimizing the energy produced under various

conditions, the analysis of results leads to the following conclusions:

OnBoard Knowledge Journal 2025, 1, 1

16 of 18

The hybrid approach, integrating wind, photovoltaic, and fuel cell technologies, has proven to be an

effective strategy for enhancing both the reliability and autonomy of the power system. By diversifying

energy sources, the system achieves greater adaptability to ﬂuctuating environmental conditions, mitigating

the intermittent nature of renewable resources.

The fuzzy logic-based management system demonstrates signiﬁcant capability in optimizing plant oper-

ation by dynamically adjusting the contribution of each energy source according to real-time environmental

inputs and electrical demand. The developed computational tool, featuring an intuitive graphical interface,

offers a versatile platform for parameter tuning and scenario testing, facilitating comprehensive evaluation

and validation of the management strategy. This tool effectively maximizes the utilization of available energy,

providing a viable and sustainable solution for hybrid renewable energy systems.

Among the subsystems analyzed, the Solid Oxide Fuel Cell (SGC) stands out as the most reliable, due

to its power output being directly controllable via hydrogen ﬂow. Unlike wind and photovoltaic sources,

whose outputs naturally ﬂuctuate with uncontrollable environmental factors, the fuel cell system maintains

power output close to its nominal capacity, ensuring system stability. However, this advantage is offset by

higher operational costs associated with its raw materials, which affects overall proﬁtability compared to the

other subsystems.

Future investigations present a compelling opportunity to enhance the robustness of these ﬁndings by

employing more detailed measurement methods. While the current study offers valuable insights into the

reliability, proﬁtability, and autonomy of a hybrid renewable energy system, further research could explore

speciﬁc aspects more deeply to achieve a comprehensive understanding.

One direction for future work involves reﬁning the data collection process. Utilizing advanced sensing

technologies and precise instrumentation can provide more granular data on the performance variables of

each energy subsystem, such as wind speed, solar irradiance, and hydrogen ﬂow. This increased level of

detail will not only improve the accuracy of the conclusions but also offer a more nuanced perspective on the

complex interactions within the hybrid system.

Furthermore, expanding experimentation across diverse environmental conditions and geographical

locations would generate a broader dataset. This geographical variability could include regions with distinct

weather patterns, enabling a comprehensive analysis of the system’s performance under various climatic

scenarios. Such an approach would be invaluable for tailoring the system’s design and management strategies

to different environmental contexts, ultimately enhancing its adaptability and reliability.

Author Contributions: José Ledesma: Writing – original draft, Supervision, Project administration, Funding acquisition.

Ever De Los Ríos: Conceptualization, Methodology, Data curation. Jair Villanueva: Software, Visualization, Validation,

Formal analysis. Meglys Pérez: Investigation, Resources, Writing – review & editing.

All authors have read and agreed to the published version of the manuscript. Please refer to the CRediT taxonomy for the

deﬁnitions of the terms. Authorship is limited to those who have made substantial contributions to the reported work.

Funding: This research received no external funding.

Institutional Review Board Statement: Not applicable, since the present study does not involvehuman personnel or

animals.

Informed Consent Statement: This study is limited to the use of technological resources, so nohuman personnel or

animals are involved.

Conﬂicts of Interest: Under the authorship of this research, it is declared that there is no conﬂict of interest with the

present research.

References

1. Blakers, A., Stocks, M., Lu, B., and Cheng, C. (2021). A review of pumped hydro energy storage. Preprint / no journal

info available.

2. Enríquez, H. (2004). Diseño de sistemas eléctricos. Editorial Limusa, México D.F. Capítulo 5.

3. Escudero, J. (2008). Manual de energía eólica. Mundi-Prensa, Madrid, España. Capítulo 5.

OnBoard Knowledge Journal 2025, 1, 1

17 of 18

4. Garcia-Guarin, J., Rivera, S., and Rodriguez, H. R. (2019). Smart grid review: Reality in colombia and expectations. J.

Phys.: Conf. Ser., 1257:012011.

5. Hall, D. J. and Colclaser, R. G. Transient modeling and simulation of tubular solid oxide fuel cells. IEEE Transactions on

Energy Conversion.

6. Hanif, M. S., Qasim, I., Malik, M. I., Nasir, M. F., Ahmad, O., and Rashid, A. (2024). Development of low-cost and

high-efﬁciency solar modules based on perovskite solar cells for large-scale applications. Heliyon, 10(4):e25703.

7. Hashem, M. (2009). Modeling and control of fuel cells: distributed generation applications. IEEE Press, Canada. Chapters 2

and 3.

8. Hilal, H., Riza, Prastawa, A., and Matsumura, T. (2018). Turning renewable energy be a dispatchable electric generation

through energy management system technology: Sumba smart micro grid case. In 2018 Conference on Power Engineering

and Renewable Energy (ICPERE), pages 1–4, Solo, Indonesia.

9. Huld, T. and Suri, M. (2007). Pvgis solar irradiation data. http://re.jrc.ec.europa.eu/pvgis/apps/radmonth.php.

European Communities.

10. Khalid, M. (2024). Smart grids and renewable energy systems: Perspectives and grid integration challenges. Energy

Strategy Reviews, 51:101299.

11. Méndez, J. and Cuervo, R. (2007). Energía solar fotovoltaica. FC Editorial, Madrid, España. Capítulos 2 y 5.

12. Oñate, D. (2006). Diseño de una instalación solar fotovoltaica. ISOFOTON.

13. Patlins, A., Caiko, J., Kunicina, N., Zhiravetska, A., and Riashchenko, V. (2020). Climate education: Challenges

of climate change and energy policies. In 2020 IEEE 61th International Scientiﬁc Conference on Power and Electrical

Engineering of Riga Technical University (RTUCON), pages 1–7, Riga, Latvia.

14. Rey, J., Vergara, P., Osma, G., and Ordóñez, G. (2013). Analysis of the inclusion of smart grids technology in the

colombian electric power system. In 2013 International Symposium on Power Quality, SICEL.

15. Sarkar, M. and Seo, Y. W. (2021). Renewable energy supply chain management with ﬂexibility and automation in a

production system. Journal of Cleaner Production, 324:129149.

16. Yahyaoui, I. (2018). Advances in renewable energies and power technologies: volume 1: solar and wind energies. Book.

17. Yan, J., Liu, S., Yan, Y., Zhang, H., Liang, C., Wang, B., and Liu, Y. (2024). A data-driven model for power system

operating costs based on different types of wind power ﬂuctuations. Journal of Environmental Management, 351:119878.

18. Zhang, Y., Ningbo, W., Ding, K., Zhou, Q., Gao, P., and Zhang, Z. (2019). The key technology of the coordinated

control system of wind power and energy-intensive load. In 4th International Conference on Intelligent Green Building and

Smart Grid (IGBSG), Hubei, China.

19. Zhao, K., Kong, H., Tan, S., Yang, X.-G., Zheng, H., Yang, T., and Wang, H. (2023). Analysis of a hybrid system

combining solar-assisted methanol reforming and fuel cell power generation. Energy Conversion and Management,

297:117664.

20. S¸erban, A. C. and Lytras, M. D. (2020). Artiﬁcial intelligence for smart renewable energy sector in europe—smart

energy infrastructures for next generation smart cities. IEEE Access, 8:77364–77377.

Authors’ Biography

José Alberto Ledesma Electronic and Telecommunications Engineer from Universidad

Autónoma del Caribe, with a Master’s degree in Electronic Engineering from Universidad del

Norte. He is a Minciencias associate and currently works as a full-time professor at Universidad

Autónoma del Caribe.

Evert De Los Ríos Researcher and faculty member in the Electronics and Telecommunications

Engineering program at Universidad Autónoma del Caribe.

OnBoard Knowledge Journal 2025, 1, 1

18 of 18

Jair Villanueva Ph.D. in Biomedicine, researcher and faculty member in the Mechatronics

Engineering program at Universidad Autónoma del Caribe.

Meglys Pérez Researcher and faculty member in the Electronics and Telecommunications

Engineering program at Universidad Autónoma del Caribe.

Disclaimer/Editor’s Note: Statements, opinions, and data contained in all publications are solely those of the individual

authors and contributors and not of the OnBoard Knowledge Journal and/or the editor(s), disclaiming any responsibility

for any injury to persons or property resulting from any ideas, methods, instructions, or products referred to in the

content.