The standard industrial process to produce medium voltage electric cables based on EPDM is characterized by a crosslinking obtained through peroxides with either nitrogen or steam (pressurized water vapor). Suboptimal material crosslinking is usually due to wrong vulcanization conditions, sometimes a consequence of a temperature drop of the vulcanizing agent along the production line. A Genetic Algorithm (GA) is here proposed to either predict the final crosslinking degree or select the optimal vulcanization conditions to maximize final insulator mechanical properties. In the first case, the algorithm is applied to a real steam water plant, where the final curing degree of the cable is experimentally obtained with Differential Scanning Calorimetry (DSC). The unexpected under-vulcanization of the item is a consequence of a steam temperature drop along the line, accurately predicted by the numerical approach proposed through the minimization of the difference between numerically predicted and experimentally determined crosslinking degree. In the second case, the GA is applied to a production line that uses nitrogen as vulcanization agent, with the aim of finding the optimal temperature and curing velocity to adopt in order to maximize final mechanical properties, such as tensile and tear strength. Both practical applications discussed show how production conditions could be automatically calculated according to the cable parameters, by increasing quality reliability and reducing scraps.
Numerical assessment of rubber insulated electric cables plants efficiency using nitrogen and steam water as curing agents
MILANI, GABRIELE;
2016-01-01
Abstract
The standard industrial process to produce medium voltage electric cables based on EPDM is characterized by a crosslinking obtained through peroxides with either nitrogen or steam (pressurized water vapor). Suboptimal material crosslinking is usually due to wrong vulcanization conditions, sometimes a consequence of a temperature drop of the vulcanizing agent along the production line. A Genetic Algorithm (GA) is here proposed to either predict the final crosslinking degree or select the optimal vulcanization conditions to maximize final insulator mechanical properties. In the first case, the algorithm is applied to a real steam water plant, where the final curing degree of the cable is experimentally obtained with Differential Scanning Calorimetry (DSC). The unexpected under-vulcanization of the item is a consequence of a steam temperature drop along the line, accurately predicted by the numerical approach proposed through the minimization of the difference between numerically predicted and experimentally determined crosslinking degree. In the second case, the GA is applied to a production line that uses nitrogen as vulcanization agent, with the aim of finding the optimal temperature and curing velocity to adopt in order to maximize final mechanical properties, such as tensile and tear strength. Both practical applications discussed show how production conditions could be automatically calculated according to the cable parameters, by increasing quality reliability and reducing scraps.File | Dimensione | Formato | |
---|---|---|---|
2016_Simulation_and_Modelling_Milani_Milani_Chapter.pdf
Accesso riservato
Descrizione: 2016_Simulation_and_Modelling_Mil_Mil
:
Publisher’s version
Dimensione
533.18 kB
Formato
Adobe PDF
|
533.18 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.