REVIEW STUDY ON DIRT REMOVAL USING MECHANICAL CLEANING TECHNIQUE: SEMIAUTOMATIC AND AUTOMATIC

Abstract

Objective: To carry out a literature review to identify the main studies found on mechanical cleaning techniques, divided into semi-automatic and automatic, which may or may not use water to remove dirt from photovoltaic modules. Theoretical Framework: The research is based on the analysis of scientific articles and publications on specialized websites in recent years, to identify publications on mechanical cleaning techniques according to inclusion and exclusion criteria. Method: The methodology adopted is an integrative review of the literature using published scientific articles and technical information taken from reliable websites as sources for research, including searches for titles relevant to the topic, as well as abstracts and articles relevant to the research. Results and Discussion: The results obtained revealed that the applications of the techniques in some studies showed efficiency of up to 25% in cleaning modules, using Arduino UNO microcontrollers combined with sensors and rotating brushes to remove dirt, developing a cleaning robot prototype, in addition to carrying out mechanical tests to verify the conservation status of the surface of the modules. Research Implications: The practical and theoretical implications of this research are discussed, providing insights into how the results can be applied or influence mechanical cleaning techniques. These implications may include the way dirt is removed from photovoltaic modules. Originality/Value: This study contributes to the literature with the main mechanical cleaning methods developed in recent years, combining the use or not of water to remove dirt in photovoltaic systems.

Alternate abstract:

Objetivo: Realizar uma revisão da literatura para identificar os principais estudos encontrados sobre técnicas de limpeza mecânica, divididas em semiautomática e automática, que podem ou não utilizar água para a remoção de sujidade em módulos fotovoltaicos. Referencial Teórico: A pesquisa fundamenta-se na análise de artigos científicos e publicações em sites especializados nos últimos anos, para identificar publicações de técnicas de limpezas mecânicas conforme critérios de inclusão e exclusão. Método: A metodologia adotada é uma revisão integrativa da literatura utilizados como fontes para pesquisas artigos científicos publicados e informações técnicas retiradas de sites confiáveis incluindo buscas de títulos pertinentes ao tema, bem como resumos e artigos pertinentes à pesquisa. Resultados e Discussão: Os resultados obtidos revelaram que as aplicações das técnicas em alguns estudos, apresentaram eficiência de até 25% na limpeza de módulos, utilização de microcontroladores Arduino UNO combinado com sensores e escovas rotativas para remoção de sujidade, desenvolvimento de protótipo de robô de limpeza, além da realização de ensaios mecânicos para averiguação do estado de conservação da superfície dos módulos. Implicações da Pesquisa: As implicações práticas e teóricas desta pesquisa são discutidas, fornecem além do insights sobre como os resultados podem ser aplicados ou influenciar práticas de técnicas técnicas de limpezas mecânicas. Essas implicações podem abranger a forma de remoção de sujidade em módulos fotovoltaicos. Originalidade/Valor: Este estudo contribui para a literatura com os principais métodos de limpezas mecânicas desenvolvidas nos últimos anos, combinando utilização ou não de água para remoção de sujidade em sistemas fotovoltaicos.

Alternate abstract:

Objetivo: Realizar una revisión de la literatura para identificar los principales estudios encontrados sobre técnicas de limpieza mecánica, divididas en semiautomáticas y automáticas, que pueden o no utilizar agua para eliminar la suciedad de los módulos fotovoltaicos. Marco Teórico: La investigación se basa en el análisis de artículos científicos y publicaciones en sitios web especializados de los últimos años, para identificar publicaciones sobre técnicas de limpieza mecánica según criterios de inclusión y exclusión. Método: La metodología adoptada es una revisión integradora de la literatura utilizando artículos científicos publicados e información técnica tomada de sitios web confiables como fuentes para la investigación, incluyendo búsquedas de títulos relevantes al tema, así como resúmenes y artículos relevantes a la investigación. Resultados y Discusión: Los resultados obtenidos revelaron que las aplicaciones de las técnicas en algunos estudios mostraron una eficiencia de hasta un 25% en la limpieza de módulos, utilizando microcontroladores Arduino UNO combinados con sensores y cepillos rotatorios para remover la suciedad, desarrollando un prototipo de robot de limpieza, además de realizar pruebas mecánicas para verificar el estado de conservación de la superficie de los módulos. Implicaciones de la investigación: Se discuten las implicaciones prácticas y teóricas de esta investigación, brindando perspectivas sobre cómo los resultados pueden aplicarse o influir en las técnicas de limpieza mecánica. Estas implicaciones pueden incluir la forma en que se elimina la suciedad de los módulos fotovoltaicos. Originalidad/Valor: Este estudio aporta a la literatura los principales métodos de limpieza mecánica desarrollados en los últimos años, combinando el uso o no de agua para eliminar la suciedad en sistemas fotovoltaicos.

Headnote

ABSTRACT

Objective: To carry out a literature review to identify the main studies found on mechanical cleaning techniques, divided into semi-automatic and automatic, which may or may not use water to remove dirt from photovoltaic modules.

Theoretical Framework: The research is based on the analysis of scientific articles and publications on specialized websites in recent years, to identify publications on mechanical cleaning techniques according to inclusion and exclusion criteria.

Method: The methodology adopted is an integrative review of the literature using published scientific articles and technical information taken from reliable websites as sources for research, including searches for titles relevant to the topic, as well as abstracts and articles relevant to the research.

Results and Discussion: The results obtained revealed that the applications of the techniques in some studies showed efficiency of up to 25% in cleaning modules, using Arduino UNO microcontrollers combined with sensors and rotating brushes to remove dirt, developing a cleaning robot prototype, in addition to carrying out mechanical tests to verify the conservation status of the surface of the modules.

Research Implications: The practical and theoretical implications of this research are discussed, providing insights into how the results can be applied or influence mechanical cleaning techniques. These implications may include the way dirt is removed from photovoltaic modules.

Originality/Value: This study contributes to the literature with the main mechanical cleaning methods developed in recent years, combining the use or not of water to remove dirt in photovoltaic systems.

Keywords: Automation, Rotating Brushes, Mechanical Cleaning, Cleaning Techniques.

RESUMO

Objetivo: Realizar uma revisão da literatura para identificar os principais estudos encontrados sobre técnicas de limpeza mecânica, divididas em semiautomática e automática, que podem ou não utilizar água para a remoção de sujidade em módulos fotovoltaicos.

Referencial Teórico: A pesquisa fundamenta-se na análise de artigos científicos e publicações em sites especializados nos últimos anos, para identificar publicações de técnicas de limpezas mecânicas conforme critérios de inclusão e exclusão.

Método: A metodologia adotada é uma revisão integrativa da literatura utilizados como fontes para pesquisas artigos científicos publicados e informações técnicas retiradas de sites confiáveis incluindo buscas de títulos pertinentes ao tema, bem como resumos e artigos pertinentes à pesquisa.

Resultados e Discussão: Os resultados obtidos revelaram que as aplicações das técnicas em alguns estudos, apresentaram eficiência de até 25% na limpeza de módulos, utilização de microcontroladores Arduino UNO combinado com sensores e escovas rotativas para remoção de sujidade, desenvolvimento de protótipo de robô de limpeza, além da realização de ensaios mecânicos para averiguação do estado de conservação da superfície dos módulos.

Implicações da Pesquisa: As implicações práticas e teóricas desta pesquisa são discutidas, fornecem além do insights sobre como os resultados podem ser aplicados ou influenciar práticas de técnicas técnicas de limpezas mecânicas. Essas implicações podem abranger a forma de remoção de sujidade em módulos fotovoltaicos.

Originalidade/Valor: Este estudo contribui para a literatura com os principais métodos de limpezas mecânicas desenvolvidas nos últimos anos, combinando utilização ou não de água para remoção de sujidade em sistemas fotovoltaicos.

Palavras-chave: Automação, Escovas Rotativas, Limpeza Mecânica, Técnicas de Limpeza.

RESUMEN

Objetivo: Realizar una revisión de la literatura para identificar los principales estudios encontrados sobre técnicas de limpieza mecánica, divididas en semiautomáticas y automáticas, que pueden o no utilizar agua para eliminar la suciedad de los módulos fotovoltaicos.

Marco Teórico: La investigación se basa en el análisis de artículos científicos y publicaciones en sitios web especializados de los últimos años, para identificar publicaciones sobre técnicas de limpieza mecánica según criterios de inclusión y exclusión.

Método: La metodología adoptada es una revisión integradora de la literatura utilizando artículos científicos publicados e información técnica tomada de sitios web confiables como fuentes para la investigación, incluyendo búsquedas de títulos relevantes al tema, así como resúmenes y artículos relevantes a la investigación.

Resultados y Discusión: Los resultados obtenidos revelaron que las aplicaciones de las técnicas en algunos estudios mostraron una eficiencia de hasta un 25% en la limpieza de módulos, utilizando microcontroladores Arduino UNO combinados con sensores y cepillos rotatorios para remover la suciedad, desarrollando un prototipo de robot de limpieza, además de realizar pruebas mecánicas para verificar el estado de conservación de la superficie de los módulos.

Implicaciones de la investigación: Se discuten las implicaciones prácticas y teóricas de esta investigación, brindando perspectivas sobre cómo los resultados pueden aplicarse o influir en las técnicas de limpieza mecánica. Estas implicaciones pueden incluir la forma en que se elimina la suciedad de los módulos fotovoltaicos.

Originalidad/Valor: Este estudio aporta a la literatura los principales métodos de limpieza mecánica desarrollados en los últimos años, combinando el uso o no de agua para eliminar la suciedad en sistemas fotovoltaicos.

Palabras clave: Automatización, Cepillos Rotativos, Limpieza Mecánica, Técnicas de Limpieza.

1 INTRODUCTION

The use of solar energy for the production of electrical energy is increasingly present in the global and national electrical matrix. According to the ( International Energy Agency , 2019), "the global installed photovoltaic capacity surpassed the 500 GW mark in 2018, while in Brazil it reached an installed capacity of 1,798 MW and was responsible for 3,461 GWh of electrical energy produced in the same year". According to Freitas Filho et al. (2020) We must also consider that, in Brazil, between 2017 and 2018, there was a 92.2% increase in the installed capacity of photovoltaic systems and a 316.1% increase in the production of electrical energy through solar plants (Empresa de Pesquisas Energéticas, 2019).

However, dirt on the surface of photovoltaic panels significantly affects their efficiency in generating energy, since any accumulation can cause shading on the cells, preventing them from generating electricity and preventing the absorption of all radiation incident on them (Cavalcante, et al. 2016). Therefore, regular cleaning is essential. of the panels to prevent the deposition of dirt.

1.1 OBJECTIVE

The purpose of this literature review is to identify the main results found, in recent years, on mechanical cleaning techniques, divided into semi-automatic (with the help of people) and automatic, both techniques may or may not use water to remove dirt from photovoltaic modules.

2 METHODOLOGY

This study presents a specific review of the literature with the purpose of evaluating mechanical cleaning techniques divided into semi-automatic and automatic.

According to ( Gandomzadeh , 2025), mechanical cleaning methods can also be classified into two subcategories: water-based cleaning and waterless cleaning for removing dirt from photovoltaic modules.

For the literary review, scientific articles and technical information taken from reliable websites were researched.

When including articles, the criteria considered in this review are represented in the flowchart in Figure 1.

For searches on specialized websites of manufacturers of mechanical cleaning systems for photovoltaic panels, the criterion used was that their product has application in photovoltaic plants.

From September 2024 to March 2025, research was carried out on the following portals: Science Direct and the portal of the Coordination for the Improvement of Higher Education Personnel (CAPES), as well as on specialized websites with the aim of identifying publications on mechanical cleaning techniques developed in recent times.

3 TECHNIQUES USED

According to Rodrigues and Diniz (2020), the effects resulting from dirt are correlated with factors that induce the loss of power in solar panels, requiring studies of techniques to remove this layer, in order to resume its normal operation.

According to Cavalcante et al. (2016), we can highlight the main mechanical cleaning technologies available on the market today, such as: semi-automatic cleaning, with a module washing technique using water and automated cleaning systems that can use water, brush, compressed air or a combination of techniques.

SILVA et al . (2024) indicated that the study of the automatic cleaning prototype obtained approximately 25 % efficiency, figure 2, in relation to measurements in which the panel is presented slightly dusty, or that is, dirt natural accumulated over time, exceeding expectations prior to project execution. That cleaning of the modules can be done daily or according to a schedule . They claim that a closed-loop system can be implemented, based on voltage drops, and also use activation systems such as Alexa or smart switches, in addition to the use of detergents to reduce the amount of dirt.

Cavalcante et al ., (2016), describes that the technology uses an automatic system with the use of a programmable controller to clean panels using water and detergent through nozzles to spray the solar modules, with the need for occasional replenishment of the soap concentrate in a five-gallon reservoir. It can be adapted to any panel array configuration, as shown in Figure 3.

According to Mousavi and Farahani (2022), their dry cleaning system, which uses Cartesian robots with rotating microfiber brushes and air jets, as shown in figure 4, has the characteristics of having its own solar module with monitoring through sensors, allowing the batteries to be charged quickly, in addition to receiving notifications of abnormal behavior in real time for remote correction.

Olorunfemi et al . (2022) proposed a low-cost application using the Arduino UNO and a TCS3200 color sensor in conjunction with rotating brushes without the use of water. The author highlights that the system may be ineffective in relation to old dirt and the need for periodic replacement of the cleaning brushes. That the color sensor, figure 5. requires a maximum distance of 3 cm to recognize dirt. He reaffirms the need for not needing to use water or perform manual cleaning. He states that the cleaning frequency does not influence the total cost of operating the system. And that the proposed model can operate for many years without the need for maintenance.

Sugiartha et al. (2020) describe a semi-automatic cleaning system where a person manually controls, using buttons, a robot that sprays water while cleaners move back and forth to perform cleaning. The authors observed that the greater the repetition, the better the cleaning efficiency and that in tests with repetitions of 10, 20 and 30 times (figure 6) they obtained results that provided 57.0%, 79.1% and 86.7%, respectively, of the performance of the initial clean surface condition. For future work, implement a microcontroller to make it an autonomous system.

According to the company Next Automation (2024), its solar module cleaning system is adapted through a hydraulic system driven by the tractor itself, in addition to accessories such as a cleaning brush, water storage tank and hydraulic control unit. This tractor cleaning method can be adapted for dry and wet cleaning purposes depending on the customer's needs. It can be used on any type of soil, however it requires space for the vehicle to pass between the solar system sets, as shown in figure 7.

Qdah et al. (2019 ) created a cleaning system applicable to homes and businesses that can be easily used for any photovoltaic array, powered by the solar modules themselves, using a DC motor, pump, water jets, and cleaning rod, with horizontal movement and a speed of 0.13 m/s. The design was developed in aluminum, as it is light and resistant to corrosion, in addition to having lower maintenance costs. In the tests carried out, it presented an efficiency of 13.78% for the cleaning module with a reduction of 34.68% in the input power of 805 W/m2, as shown in Figure 8, and an efficiency of 9% for the dusty surface panel at 42% in the input power of 460 W/m2, as shown in Figure 9.

Deb and Brahmbhatt (2018) developed a waterless cleaning system using an Arduino UNO, rotating brushes, and an ultrasonic sensor to detect obstacles. The prototype cost around $450 to produce. Figure 10 shows an energy efficiency of around 9.05%. According to the authors, the maintenance cost is insignificant compared to the promising benefits it offers.

Parrot et al. (2018) observed that the cleaning system using silicone rubber brushes showed a 3% better performance over a two-week period when compared to daily manual cleaning, which always occurred in the morning from 6:30 to 7:00 am. After evaluating, figure 11, the cleaning with 1000 cycles using the electroluminescence technique, there was no damage such as microcracks or broken cells in the modules, which indicates that the quality of the photovoltaic system was preserved.

Cavalcante et al. (2016) created a prototype of a cleaning system that uses rainwater, using Arduino UNO to control a water pump, Brucutu-type sprayers, servomotors and filters, and its cost, figure 12, was 1,184.68 reais. They evaluated the technical and economic viability of the prototype by comparing the result with the Heliotex self-cleaning system (US$ 2,258.00), which presented a cost six times lower, in addition to cleaning less aggressive to the environment, due to the reuse of water.

Wuxi company Wanlv Intelligent Technology CO., LTD (2024) the solar module cleaning robot works with the use of water, is operated by a person through a wireless remote control controller, in addition to using rotating brushes to remove dirt. It weighs approximately 34 kg and can be used in systems with a working inclination of up to 20°, as shown in figure 13.

Wuxi company Wanlv Intelligent Technology CO., LTD (2024), the rotating brushes, made of nylon, with double head WLS-4 are driven by a DC motor powered by a lithium battery, in addition to having a telescopic handle. Its cleaning system is through water, as shown in figure 14 and figure 15.

4 FINAL CONSIDERATIONS

Considering that the vast majority of photovoltaic systems are installed in desert and arid regions, the use of water to remove dirt is not a good alternative due to its scarcity.

However, a table comparing the main data obtained during the literature review can be presented, as shown in Table 1.

References

REFERENCES

Cavalcante, M. M, Marcelino, J. E. C., Delgado, D. B. M., Viana, E. C. (2016). Protótipo de um sistema automatizado para higienização de painéis solares planos agrupado a um sistema de reaproveitamento de água. Anais do II SINGEP - Simpósio Internacional de Gestão de Projetos, Inovação e Sustentabilidade, (n.p.), s.p.

Deb, D., Brahmbhatt, N. L. (2018). Review of yield increase of solar panels through soiling prevention, and a proposed water-free automated cleaning solution, Renewable and Sustainable Energy Reviews, (82), 3306-3313.

Empresa de Pesquisa Energética (2019). Balanço energético nacional, s.l., (n.p.), s.p.

Freitas Filho, M. F., Araújo, D. N., Carvalho, P.. C. M., Sasaki, J. M. (2020). Análise da composição físico-química da sujidade de plantas fotovoltaicas: estudo de caso para Fortaleza. Revista Tecnologica, (41), 1-17.

Gandomzadeh, M., Yaghoubi, A. A., Hoorsun, A., Parsay, A., Gholami, A., Zandi, M., Gavagsaz-Ghoachani, Kazem, R. H. A. (2025). Dust mitigation methods and multi-criteria decision-making cleaning strategies for photovoltaic systems: Advances, challenges, and future directions. Energy Strategy Reviews, (57), 101629.

International Energy Agency. (2019). Snapshot of global markets, s.l., (n.p.), s.p.

Mousavi, S., Farahani, G. (2022). Introducing a new method of automatic cleaning of the PV array surface using a suction robot, Mechatronics, (85), 102845.

Next Automation (2024). Solar Panel Cleaning Tractors, s.l., (n.p.), s.p.

Olorunfemi, B. O., Nwulu, N. I., Ogbolumani, O. A.(2022). Solar panel surface dirt detection and removal based on arduino color recognition, MethodsX, (n.p.), 101967.

Parrott, B., Zanini, P. C., Shehri, A., Kotsovos, K., Gereige, I. (2018). Automated, robotic drycleaning of solar panels in Thuwal, Saudi Arabia using a silicone rubber brush, Solar Energy, (171), 526-533.

Qdah, K. S., Abdulqadir, S. A., Harbi, N. Y. A., Soqyyah, A. Z., Isa, K. J., Alharbi, M. Y., Binsaad, N. M. (2019). Design and performance of PV dust cleaning system in medina region, Journal of Power and Energy Engineering, (7), 1-14.

Rodrigues, J. A. P., Diniz, S. A. C. (2020). Estudo de revisão sobre as técnicas aplicadas na remoção de sujidades depositadas sobre a superfície de módulos fotovoltaicos, bem como os seus efeitos, Anais Congresso Brasileiro de Energia Solar - CBENS, (n.p.), s.p.

Silva, R. C., Oliveira, V. C., Vasconcelos, M. S. A., Barreto, G. O. (2024). Processo Automático para Limpeza de Sistema Fotovoltaico. Revista Projectus, (8), 67-85.

Sugiartha, N., Ardana, I. G. N., Sugina, I. M., Widiantara, I. B. G., Suparta, I. N., Adi, I. K. (2020). Preliminary design and test of a water spray solar panel cleaning system, Journal of Physics: Conference, (1450), 012108.

Wuxi Wanlv Intelligent Technology CO. (2024). Escova rotativa de cabeça dupla WLS-4, s.l., (n.p.), s.p.

Wuxi Wanlv Intelligent Technology CO. (2024). Automatic Cleaning Robot WLS-73S, s.l., (n.p.), s.p.