Data Science and advanced data analytics are technologies that enable the development of low cost solutions with a high degree of customization. Based on that, this article presents a leakage detection system in a slurry pipeline using a combination of machine learning techniques. The techniques used to detect leakage were based on artificial intelligence, a machine learning model for the energy balance of the pipe combined with an anomaly detection technique approach. The system predicts energy at one point of the pipe based on another point, in order to infer if there is loss of energy (leakage) on it. Although the machine learning model used is a simple parametric linear regression and this technique is well-known in the artificial intelligence domain, its competitive differential is the use of an open source machine learning platform to implement them, which allows clients to have a customized model instead of using costly instrumentation with embedded systems. This work was fully implemented in a ore tailings pipeline of one of the biggest Brazilian iron ore companies. It has already detected several real leakage occurrences, greater than 70 m³/h with only three to five leak false alarms per month. Based on these results, this solution can be considered as an alternative solution for leak detection in short length pipelines, especially for the ones that transport iron ore tailings. Although it can detect leakage between a pipe sections, it cannot detect the exact point of the leak that motivates further development, such the use of wavelet package technique.
Resumo Para atingir a qualidade desejada no concentrado da flotação, numa usina de mineração de ferro, o operador tomava decisões com latência mínima de 2 horas, o equivalente ao tempo para recebimento de resultados da análise laboratorial. Um tempo morto tão elevado, tornava difícil colocar a planta em regime quando ocorria uma variação na alimentação da flotação ou na qualidade desejada. Dessa forma, o objetivo do presente trabalho foi construir um sensor virtual capaz de estimar o teor SiO2 e permitir a atuação do operador em um período mais curto. A alta variabilidade na alimentação, adicionada às dificuldades inerentes da predição envolvendo séries temporais tornaram o projeto singular. Foram aplicadas técnicas de mineração de dados e seleção de features em uma base de 03 meses abrangendo variáveis da Flotação e também da Deslamagem, e obtiveram-se entradas para elaboração de 03 modelos de Machine Learning: Random Forest, GradientBoostedTrees e MultiLayerPerceptron. O Soft Sensor baseado em redes neurais artificiais se mostrou estatisticamente mais eficiente na etapa de teste. Na versão online, o baixo erro médio absoluto obtido comprovou a robustez do modelo, entregou agilidade para a operação e certificou o poder dessa abordagem em processos industriais com alta latência de resultados laboratoriais.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.