Sistema de iluminação de granjas controladas por lógica fuzzy
Carregando...
Data
2021-12-17
Autores
Carvalho Neto, Joubert Alves de
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A iluminação em granjas para a produção do frango de corte não somente influencia o comportamento dos frangos, como também seu sistema fisiológico e sua engorda. Além do bem-estar animal, outro fator que deve ser considerado na iluminação de uma granja é o consumo de energia elétrica, cujo impacto pode ser significativo na receita do criador. Entretanto, o processo de controle de iluminação em granjas é pouco explorado se comparado aos processos de ventilação e refrigeração, calefação, abastecimento de água e ração. Por meio desta pesquisa, buscou-se desenvolver um sistema de controle automático de iluminação para granjas e analisar a possibilidade de integrá-lo com luz natural pelo processo de inferência fuzzy. Para construí-lo, foi proposta uma modelagem teórica dos materiais necessários à sua construção, assim como suas respectivas simulações computacionais para previsão e análise de seu comportamento. Os materiais foram responsáveis pelas etapas de entrada, processamento e saída de sinais. O sinal de entrada, em porcentagem, trata-se do erro da intensidade de luz em relação ao setpoint que incide no interior da granja, e o material responsável por sua coleta foi o módulo GY-302. Após coleta, o sinal foi processado por um sistema de inferência fuzzy com o auxílio do software MATLAB. Fazendo uso da inferência fuzzy avaliou-se ciclicamente o grau de pertinência do sinal de entrada em relação ao valor de setpoint para, após o processo de “defuzzificação”, gerar valores de correções em porcentagem, que foram incrementos e decrementos do valor do tempo de disparo para ativar o Triac. Eles foram tratados como sinais de saída e, consequentemente, direcionados ao Dimmer Shield, que foi responsável pelo controle de intensidade luminosa da lâmpada LED. Uma vez montado o protótipo, foram feitos testes para investigar seu comportamento em diferentes situações. O primeiro teste consistiu em analisar se o protótipo alcançava os valores de setpoint da intensidade luminosa sem a interferência de luz externa. As lâmpadas LED foram iniciadas com 0 e 100% de sua potência elétrica, e durante esses testes o protótipo alcançou os valores de setpoint. No segundo teste, o protótipo foi exposto à fonte de luz externa ao sistema, gerada por uma lâmpada incandescente para simular a incidência de luz natural. A intensidade da lâmpada incandescente foi controlada por um dimmer analógico (manual) e verificada por um luxímetro comercial. Nesses testes, o protótipo foi forçado a iniciar com o valor de setpoint e a tensão elétrica nesse instante foi aferida. Em um determinado instante, a luz externa era inserida no sistema e, em outro, a luz externa era retirada. O protótipo indicou graficamente o aumento da intensidade luminosa quando a luz externa era inserida, corrigindo-a até alcançar novamente o setpoint. Durante o tempo em que o setpoint foi alcançado, a tensão das lâmpadas LED foi aferida novamente, a qual mostrou uma diminuição de seus valores, ou seja, a diminuição da potência consumida por elas. O protótipo também se mostrou efetivo no controle de mudanças de setpoint, assim como no controle das horas necessárias para a escuridão na granja.
Lighting on farms for broiler production not only influences the behavior of broilers, but also their physiological system and their fattening. In addition to animal welfare, another factor that must be considered when lighting a farm is the consumption of electricity, which can have a significant impact on the creator's income. However, the lighting control process in farms is little explored compared to the processes of ventilation and cooling, heating, water supply and feed. Through this research, we sought to develop an automatic lighting control system for farms and analyze the possibility of integrating it with natural light through the fuzzy inference process. To build it, a theoretical modeling of the materials needed for its construction was proposed, as well as their respective computer simulations for prediction and analysis of their behavior. The materials were responsible for the input, processing and output stages of signals. The input signal, in percentage, is the error of light intensity in relation to the setpoint that affects the interior of the farm, and the material responsible for its collection was the GY-302 module. After collection, the signal was processed by a fuzzy inference system with the help of MATLAB software. Using fuzzy inference, the degree of relevance of the input signal in relation to the setpoint value was cyclically evaluated to, after the "defuzzification" process, generate correction values in percentage, which were increments and decrements of the value of the shot to activate the Triac. They were treated as output signals and, consequently, directed to the Dimmer Shield, which was responsible for controlling the luminous intensity of the LED lamp. Once the prototype was assembled, tests were carried out to investigate its behavior in different situations. The first test consisted of analyzing whether the prototype reached the light intensity setpoint values without the interference of external light. The LED lamps were started with 0 and 100% of their electrical power, and during these tests the prototype reached the setpoint values. In the second test, the prototype was exposed to a light source external to the system, generated by an incandescent lamp to simulate the incidence of natural light. The intensity of the incandescent lamp was controlled by an analog dimmer (manual) and verified by a commercial lux meter. In these tests, the prototype was forced to start with the setpoint value and the electrical voltage at that moment was measured. At a given moment, the external light was inserted into the system and, at another, the external light was removed. The prototype graphically indicated the increase in light intensity when the external light was inserted, correcting it until reaching the setpoint again. During the time that the setpoint was reached, the voltage of the LED lamps was measured again, which showed a decrease in its values, that is, a decrease in the power consumed by them. The prototype also proved effective in controlling setpoint changes, as well as controlling the hours required for darkness on the farm.
Lighting on farms for broiler production not only influences the behavior of broilers, but also their physiological system and their fattening. In addition to animal welfare, another factor that must be considered when lighting a farm is the consumption of electricity, which can have a significant impact on the creator's income. However, the lighting control process in farms is little explored compared to the processes of ventilation and cooling, heating, water supply and feed. Through this research, we sought to develop an automatic lighting control system for farms and analyze the possibility of integrating it with natural light through the fuzzy inference process. To build it, a theoretical modeling of the materials needed for its construction was proposed, as well as their respective computer simulations for prediction and analysis of their behavior. The materials were responsible for the input, processing and output stages of signals. The input signal, in percentage, is the error of light intensity in relation to the setpoint that affects the interior of the farm, and the material responsible for its collection was the GY-302 module. After collection, the signal was processed by a fuzzy inference system with the help of MATLAB software. Using fuzzy inference, the degree of relevance of the input signal in relation to the setpoint value was cyclically evaluated to, after the "defuzzification" process, generate correction values in percentage, which were increments and decrements of the value of the shot to activate the Triac. They were treated as output signals and, consequently, directed to the Dimmer Shield, which was responsible for controlling the luminous intensity of the LED lamp. Once the prototype was assembled, tests were carried out to investigate its behavior in different situations. The first test consisted of analyzing whether the prototype reached the light intensity setpoint values without the interference of external light. The LED lamps were started with 0 and 100% of their electrical power, and during these tests the prototype reached the setpoint values. In the second test, the prototype was exposed to a light source external to the system, generated by an incandescent lamp to simulate the incidence of natural light. The intensity of the incandescent lamp was controlled by an analog dimmer (manual) and verified by a commercial lux meter. In these tests, the prototype was forced to start with the setpoint value and the electrical voltage at that moment was measured. At a given moment, the external light was inserted into the system and, at another, the external light was removed. The prototype graphically indicated the increase in light intensity when the external light was inserted, correcting it until reaching the setpoint again. During the time that the setpoint was reached, the voltage of the LED lamps was measured again, which showed a decrease in its values, that is, a decrease in the power consumed by them. The prototype also proved effective in controlling setpoint changes, as well as controlling the hours required for darkness on the farm.
Descrição
Palavras-chave
Frango de corte, Bem-estar animal, Energia elétrica, Luz, Inferência Fuzzy, Broiler chicken, Animal welfare, Electricity, Light, Fuzzy inference