Semisupervised refrigeration plant cooling disaggregation by means of deep neural network ensemble

Cirera, Josep; Carino, Jesus A.; Zurita, Daniel; Ortega, Juan Antonio

doi:10.1109/isie.2019.8781335

Cited by 1 publication

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The NILM approach for cooling loads is detailed and validated, with a mathematical simulation performed in Matlab and in a real refrigeration system, in our previous work [25]. However, as part of the methodology presented in this article, a brief summary of the disaggregation structure is presented.…”

Section: Load Disaggregationmentioning

confidence: 99%

“…In the present article, the proposed methodology to manage the cooling loads of a refrigeration system considers the estimation of the cooling loads, with a novel NILM strategy presented in our previous work [25], and takes advantage of such estimation to reduce the compressors electrical expenditure under non-scheduled processes. Such increase of the compressors operation efficiency is achieved modulating the load by means of the switching of the evaporators allocated in the different spaces to refrigerate.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Improving the Energy Efficiency of Industrial Refrigeration Systems by Means of Data-Driven Load Management

et al. 2020

Self Cite

View full text Add to dashboard Cite

A common denominator in the vast majority of processes in the food industry is refrigeration. Such systems guarantee the quality and the requisites of the final product at the expense of high amounts of energy. In this regard, the new Industry 4.0 framework provides the required data to develop new data-based methodologies to reduce such energy expenditure concern. Focusing in this issue, this paper proposes a data-driven methodology which improves the efficiency of the refrigeration systems acting on the load side. The solution approaches the problem with a novel load management methodology that considers the estimation of the individual load consumption and the necessary robustness to be applicable in highly variable industrial environments. Thus, the refrigeration system efficiency can be enhanced while maintaining the product in the desired conditions. The experimental results of the methodology demonstrate the ability to reduce the electrical consumption of the compressors by 17% as well as a 77% reduction in the operation time of two compressors working in parallel, a fact that enlarges the machines life. Furthermore, these promising savings are obtained without compromising the temperature requirements of each load.

show abstract

Section: Load Disaggregationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%