Surge Arrester: Minimizing Electric Field Stress in Resistors

The voltage distribution along the resistors in a surge arrester is uneven due to stray capacitances. It is possible to reduce the maximum electric field stress by a careful choice of position for the grading rings. In this example, OptiNet is used with ElecNet to find where the rings should be placed, and what their dimensions should be, in order to reduce the electric field stress in the resistors.


Results

Click image to see full size version
The electric field for this device is obtained using the 2d axisymmetric electrostatic solver in ElecNet. This figure displays the equipotential lines in the air and in the resistors.

Click image to see full size version
Variables: This figure shows the parameters that control the placing and the size of the grading rings. In OptiNet, the user specifies a minimum and a maximum value for those variables that can change, and OptiNet searches within this range to find the optimum design.

Click image to see full size version
Objective function: The maximum electric field in the resistors is defined as the objective function, while the goal is to minimize this quantity. In OptiNet, basic quantities, such as the maximum field value in a component, are easily available.

Click image to see full size version
Constraints: The constraints in this optimization process are the following:
  • the two rings must not overlap
  • the rings cannot get too close to the axis

Click image to see full size version
Graph of variables: For every iteration of the optimization process, OptiNet updates and displays the changes (in the form of graphs) for the goal, variables, objectives, and constraints -- these graphs are displayed on the Progress page. In this example, each of the six variables' graphs (in groups not exceeding five variables per window) is updated as OptiNet finds a new design.

Click image to see full size version
Results: OptiNet produces a report for each optimization run. In this report, the designs that satisfy the constraints are shown in the order that they are improved. The user can view each design individually. The report also shows the time that it took to arrive at the improved design. The values of all the variables and the optimization function are displayed in this report for every iteration. The values of each parameter can be examined to determine the sensitivity of the design to that particular parameter.

Click image to see full size version
Initial design: This figure shows the electric field in the three resistors for the initial design that the user supplied to OptiNet.

Click image to see full size version
Final design: In the optimum design that OptiNet generated, the maximum electric field in the three resistors is lower. It took OptiNet 622 seconds (about 10 minutes) on an AMD Athlon XP2800+ (2.08 GHz processor) to arrive at this design.