# Using MTK when I \*import\* ModelingToolkit?

**URL:** <https://discourse.julialang.org/t/using-mtk-when-i-import-modelingtoolkit/133681>\
**Category:** Modelling & Simulations\
**Created:** [November 5, 2025, 2:19pm UTC](https://discourse.julialang.org/t/using-mtk-when-i-import-modelingtoolkit/133681 "2025-11-05T14:19:53Z")\
**Posts on this page:** 3\
**Page:** 3

<div class="post-metadata">

**Author:** ![BLI](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/bli/32/37206_2.png) [@BLI](https://discourse.julialang.org/u/BLI)\
**Post date:** [November 12, 2025, 11:30am UTC](https://discourse.julialang.org/t/using-mtk-when-i-import-modelingtoolkit/133681/41 "2025-11-12T11:30:57Z")

</div>

> [@baggepinnen](#):
>
> Use `complete(model).z` for the output. It’s unfortunate that this discrepancy exists, but I think @cryptic.ax has some explanation for why it’s hard to get rid of.

Yeah, that also works. And it saves me from creating the “fictitious” downstream system.

Of course, if I produce true “components”, then the straightforward thing is to use “analysis point” names also for the output in `linearize`.

---

<div class="post-metadata">

**Author:** ![baggepinnen](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/baggepinnen/32/693_2.png) [@baggepinnen](https://discourse.julialang.org/u/baggepinnen)\
**Post date:** [November 12, 2025, 1:19pm UTC](https://discourse.julialang.org/t/using-mtk-when-i-import-modelingtoolkit/133681/42 "2025-11-12T13:19:07Z")

</div>

> [@BLI](#):
>
> But with “analysis points”, I must instead think in terms of linearization from one connection point to another connection point.

I still think of this as linearizing between an input and an output. If there is nothing connected at either end, just use the variable rather than an analysis point, you should never have to create any dummy systems just for linearization to work.

> [@BLI](#):
>
> In my simple example above, at the outset, that is the only component there is, and there is no component upstream or downstream in the information flow.
> 
> So I need to create fictitious upstream and downstream “components”

no, if there is nothing connected, just use variables rather than analysis points.

---

<div class="post-metadata">

**Author:** ![baggepinnen](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/baggepinnen/32/693_2.png) [@baggepinnen](https://discourse.julialang.org/u/baggepinnen)\
**Post date:** [November 13, 2025, 9:18am UTC](https://discourse.julialang.org/t/using-mtk-when-i-import-modelingtoolkit/133681/43 "2025-11-13T09:18:20Z")

</div>

I just realized that I have in fact documented analysis points in a bit more detail here

> **[Analysis Points | Dyad](https://help.juliahub.com/dyad/dev/manual/advanced/analysis_points.html)**
>
> Learn about analysis points in Dyad, their purpose, and how to use them for advanced system analysis.

This shows their use from Dyad, but they of course behave the same in MTK, it’s only the syntax for creating them that is slightly different. In MTK, you can create them by adding the name straight to a connection itself `connect(out, :name, in)`.

[Previous page](https://discourse.julialang.org/t/using-mtk-when-i-import-modelingtoolkit/133681.md?page=2)