This document outlines the FEA simulation pipeline (steps) that Neara uses. It is recommended reading the terminology section at the bottom of this article. When FEA runs the simulation pipeline has two main steps:

There are two options for how to determine the forces in cables during the Conductor Solve step ruling span and network solve.

Ruling Span is simplified approach to determining horizontal tension in strain section. It replaces multiple spans with equivalent span for which horizontal tension is calculated. In ruling span simulation the endpoints of ruling span are fixed and no interactions between cables and structures is taken into account.

Once the horizontal tensions are solved for the equivalent span any global wind directions and actual cable weights are applied per span which can create an imbalance in the tensions in each span. These imbalances are not equalised but rather go into the point loads applied to the insulators.

In network solve all structures and cables comprising network of a simulated object are simulated simultaneously. This approach takes into account cable-structure interactions and deflections that affect resulting forces in the cables. In addition to Load Source solve options, one can also change options available in the Structure Model section to adjust the level of detail to which structures are modelled. Note there is a FEA setting Unstressed Lengths which controls how the unstressed lengths are calculated (click link for details).

The following steps are taken for each FEA simulation. note all of the steps are taken regardless of whether the simulation type is Ruling Span or Network Solve.

For a FEA simulation, steps 1-3 below are applicable to the Conductor Solves and step 4 for the Structure Solve. The Conductor Solves determine forces in cables (without point loads) and once the Conductor Solves have been completed the conductors are dropped from the structures and replaced with point loads. These point loads are decomposed into LTV components which have load factors applied and then each structure has an FEA solve for it.

Step 1: Stringing solve (in case when cable insulators are not clipped), used to determine unstressed length of cables at "Defined At" temperature without any permanent deformations in them.

Step 2: Creep and preload solves (if applicable). Unstressed lengths adjusted for permanent deformation from creep and preload (where the unstressed length used is the FEA solve that gives the maximum sag from the creep and preload solves).

Step 3: Simulation environment solve using the adjusted unstressed length without load factors

Step 4: Structure Solves with point loads applied

To help illustrate the different steps in the FEA pipeline consider the following example where the pole on the left is called P1, pole on the right P2 and the conductor is named cond.

Here the name of the creep environment is T20, the preload environments are Snow & Ice (with perpendicular wind directions) and Max wind (which has 4 wind directions). The simulated environment is Hot which has both creep and preload applied. The Load Source is set to Network Solve and the Boundary Constraint is set to Fixed Nodes.

The first FEA Solve will be the stringing solve which in this example will be named

NW (str): Fixed Nodes P1,P2,Cond:

This report will be shown in the Detailed (All) drop down. NW stands for Network Solve, (str) stands for stringing solve, Fixed Nodes is based on the Boundary Constraint (if it was set to Point Loads then the name in the report would change to Point Loads). P1,P2,Cond lists the model objects that were included in this stringing solve (if there are lots of model objects then instead of listing them the name will have X Poles, Y Conductor Sections:). During the stringing solve no load factors will be applied.

Note the stringing solve (str) report will be blank however this does not mean it failed to run. To view what happened during the stringing solve click record steps in the Advanced Options and then after running the FEA analysis select this report in Current Solves in the Advanced Options.

There is an exception where the stringing solve does not run and this is when all the spans in the strain section are length constrained (clipped insulators). In this case that length is the unstressed length (hence no need to solve for them).

The next set of FEA solves will be for the creep environment and the preload environments. These reports are shown in the Detailed (All) drop down.

For the creep environment the FEA solve will be named:

NW T20 as Creep: Fixed Nodes P1,P2,Cond:

The only new part of the naming here is as Creep indicating this is the simulation to calculate the tension in the creep environment. All the other naming follows the same pattern explained in step 1.

For the preload environments solves the names will be:

NW Max wind 0°: Fixed Nodes P1,P2,Cond:

NW Max wind 90°: Fixed Nodes P1,P2,Cond:

NW Max wind 180°: Fixed Nodes P1,P2,Cond:

NW Max wind 270°: Fixed Nodes P1,P2,Cond:

NW Snow & Ice A: Fixed Nodes P1,P2,Cond:

NW Snow & Ice B: Fixed Nodes P1,P2,Cond:

Note there is no as Load in the naming here (as in the case for the creep environment) since these runs can be used for other simulations (for example simulating the preload environment explicitly). The A & B refer to the perpendicular wind directions and 0°, 90°, ... refer to global wind directions.

Note all the unstressed lengths shown in this set of reports are the unstressed lengths without creep or preload taken into account however they are reporting the unstressed length at the wire temperature for each environment. Hence it could appear like the unstressed lengths are different between this set of reports but the difference will be coming from different wire temperatures in the environment (if αΔt is used to adjust the unstressed lengths they will all match up: where α is coefficient of linear expansion and Δt is the temperature difference between environments).

The next step is a FEA solve for the environment in the simulation. This will use the unstressed length that has been adjusted for permanent deformations from creep and preload. This reports are shown in the Detailed (All) drop down.

The name of this report will be:

NW Hot+Lx6+C: Fixed Nodes P1,P2,Cond:

Here the +Lx6 indicates that 6 preload simulations have been applied prior to this simulation and the +C indicates that the creep simulation has also been applied prior to this simulation (where the simulation that creates the max sag is selected).

The unstressed length in this report will include the permanent deformation due to creep and preload and will be at the wire temperature specified in the environment Hot.

The results in these reports will not include load factors however they will include strength reduction factors for components. Also these results will not include any point forces since the cables are still included in the FEA solves (the cables will be replaced by point forces when running FEA solves for each structure).

The final step moves into the structure simulations. Here the conductors are removed and instead point loads are applied to the insulator end points. These point loads are resolved into LTV components and then factored based on any load factors set for that environment. Note if only the conductor was selected and no structures then there will be no LF FEA solves in the Detailed (All) section.

So in this case reports will be:

LF Hot+Lx6+C: Point Loads P1:

LF Hot+Lx6+C: Point Loads P2:

Note that each structure is analysed separately and will trigger its own FEA Solve.

The structure deflections and utilisations will be different in the LF reports than the NW (or RS) reports since the LF reports have applied the load factors (which typically will mean the structures will experience higher loads and larger deflections).

In the bottom right corner the number of FEA solves outstanding along with failed solves are displayed. Clicking on this give additional options such as pausing, changing the number of processers the FEA engine is using and clearing failed solves warning messages.

(1) Allows FEA calculations to be paused. Note there is no way to clear the scheduled runs from here.

(2) Number of Processers allocated to FEA engine

(3) This will clear any of the failed solves shown in (5). Note that clearing the runs will not clear these warnings

(4) Number of FEA Solves left. Note this number can go up as a FEA Simulation is performed the FEA engine will discover how many FEA Solves are required (eg additional FEA solves might be required beyond the boundary of the strain sections depending on the FEA settings).

(5) Number of failed FEA Solves

The following terminology will be used throughout this article.

FEA Pipeline (also referred to as pipeline) refers to the overall steps, comprised of FEA solves, involved in running a single FEA simulation

FEA Simulation (also referred to as simulation) refers to simulating a single object in a single environment using FEA with specific FEA options. Hence each row in the Simulations table is referred to as a simulation (hence to run all the rows here 8 simulations would be performed).

Note that each simulation runs independently but some key results are cached such as unstressed lengths at stringing conditions to avoid duplicating calculations. For example the unstressed lengths in stringing conditions calculation is common across all environments and hence the FEA engine will first check if a FEA Solve has been performed that contains the unstressed length, if it is found then it retrieves this result otherwise it will initiate an FEA Solve to obtain the unstressed lengths.

FEA Solve (also referred to as solve) is a single invocation of the FEA solver which will create a report in the Detailed Results (All) and can be visually viewed in Current Solves in the Advanced Options (if record steps has been selected in the Advanced Options first). Note an FEA simulation will often require multiple FEA Solves.

Unstressed Length is the length of the conductor if it has zero tension (effectively lying on the ground in a straight line). Neara reports on unstressed length for each simulation where such length is at the temperature of the environment and contains permanent deformations. Unstressed lengths from step 1 of simulation pipeline, however, are stringing unstressed lengths at “Defined At” temperature of conductor and without any permanent deformations. Note that the stringing unstressed length is the same for all simulated environments and hence when simulating multiple environments this calculation is only performed once and then subsequent simulations will use these results.

[1] Thai, H. T., & Kim, S. E. (2011). Nonlinear static and dynamic analysis of cable structures. Finite Elements in Analysis and Design, 47(3), 237-246. https://doi.org/10.1016/j.finel.2010.10.005