Cost-optimal as a tool for helping in designing distribution transformer using particle swarm optimization

• Mohammad Amin Jarrahi ^[1] ; Emad Roshandel ^[2] ; Mehdi Allahbakhshi ^[1] ; Mohammad Ahmadi ^[1]
  1. [1] Shiraz University

     Shiraz University

     Irán

     
  2. [2] Isfahan University of Technology

     Isfahan University of Technology

     Irán

     
• Localización: Compel: International journal for computation and mathematics in electrical and electronic engineering, ISSN 0332-1649, Vol. 38, Nº 2, 2019, págs. 862-877
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Purpose – This paper aims to achieve an optimal design for distribution transformers considering cost and power losses. Particle swarm optimization (PSO) algorithm is used as an optimization tool for minimizing the objective functions of design procedure which are cost and electrical and iron losses.

    Design/methodology/approach – In this paper, distribution transformer losses are considered as operating costs. Also, transformer construction cost which depends on the amount of iron and copper in the structure is assumed as its initial cost. In addition, some other important constraints such as appropriate ranges of transformer efficiency, voltage regulation, temperature rise, no-load current, and winding fill factor are investigated in the design procedure. The PSO algorithm is applied to find optimum amount of needed copper and iron for a typical distribution transformer. Moreover, transformer impedance considered as a constraint to achieve an acceptable voltage regulation in the design process.

    Findings – It is shown that the proposed design procedure provides a simple and effective approach to estimate the flux and current densities for minimizing the active part cost and active power losses which means reduction in amount of transformer total owning cost (TOC).

    Originality/value – The methodology advances a proposal for reducing distribution transformers costs using PSO algorithm. The approach considers the aforementioned constraints and TOC to minimize the active part cost and maximize the efficiency. It is demonstrated that a designed transformer will not be optimum when the transformer losses over years are not considered in design procedure. Finally, the results prove the effectiveness of the proposed procedure in designing cost-effective distribution transformers from its initial cost until its whole life.