Otimização do controle nível do silo de cavacos em uma fábrica de celulose

• Autor(a) principal: Carrilho, Vicente Nunes
• Data de Publicação: 2010
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/5902
• topic: Otimização; Controle de nível do silo de cavacos; Cavacos; Fábrica de celulose; Celulose; Optimization; Chip bin level; Chips; Pulp mill; Pulp; CNPQ::CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL::TECNOLOGIA E UTILIZACAO DE PRODUTOS FLORESTAIS::TECNOLOGIA DE CELULOSE E PAPEL

This work shows the results using of advanced control algorithm and soft-sensor to increase the efficiency of the stage of pre-vaporization in the cooking plant process of extracting pulp, by optimizing the level of the Diamondback® chip bin. The extraction system chips presents several obstacles to effective control, such as: (1) points without stack of chips during the transition movement of the extraction system bin in the storage chips and consequently sudden decrease in the volume of chip bin, (2) compaction and uneven stack height causing high variability in the volume of chip bin, (3) without information in real time position of the threads in the operating display (4) high travel time between the extraction system and measurement points (volume of the silo and the balance volume). To eliminating the section without chips were available eight virtual switches limits, positioned by operator, on the operation screen, limiting the transition movement of the extraction system chips. The new control strategy the measurement is performed by a soft-sensor that predicts the volume of chips, using multiple linear regression. Using soft-sensor to measurement volume, so the retention time of the system was eliminate and decrease the variability of volume of chips bin. Using this new control strategy allied with soft-sensor proved to be adequate to solve the problem, the system became able to request of the Andritz company (70% + / - 10% of measurement section), The new control algorithm efficiency depends of the adjustments, tuning and learning with the system in operation.´