FAQ（常见问题解答）¶

Which Quiver cards are supported in derived series logic?¶

The following Quiver cards are supported in derived series logic:

Object time series property
Filter time series
Derivative
DSP filter
Integral
Sample
Segment statistics
Time series formula
Combine time series
Cumulative aggregate
Periodic aggregate
Rolling aggregate
Relative time series
Shift time series
Shift date/time
Numeric parameter
String parameter
Date/time parameter
Date/time range parameter
Duration unit parameter
Boolean parameter

Why did I receive an error for multiple root objects?¶

If you receive an error for multiple root objects in the Quiver creation dialog, your analysis likely started from a sensor object type and used two different individual objects for comparison. For example, you may have pulled and compared the sensor objects Inlet pressure and Outlet pressure from the sensor object type instead of the Machine 1 root object type. To create a derived series, you will need to start from the root object instance.

Review the logic requirements for setting up a templated derived series to learn why derived series must be generated from the perspective of a root object.

Can I apply the same derived series template to different objects?¶

Yes. The logic and calculations are templated so results can be calculated from the perspective of any bound object with the same object type. Review the section on storing derived series in the Ontology, and notice that the same codex template is applied to different Machine objects or sensor objects.

What is the difference between the codex template RID and the derived series RID in the URL?¶

The derived series RID is for the Quiver resource that holds the resource name, description, and other metadata. The codex template RID listed on the Derived series management page is a reference to the template logic used when the derived series data is calculated at runtime. Review the section on storing derived series in the Ontology for more details.

How can I reference a specific version of derived series logic in a TSP?¶

To reference a specific version of logic, your TSP value should look like the following:

{"templateRid":"ri.codex-emu.main.template.8da5f759-4b...","templateVersion":"0.0.x"}

If my input series changes, does the derived series update accordingly?¶

Since derived series are calculated at read time, the derived series will show any updates to the input series that are available. For example, any change to the input series that are visible in Quiver will also be reflected in the derived series.

Can I use my derived series as an input to another derived series?¶

Yes, derived series can be nested and used to calculate another derived series. You can use the derived series in the same way you would use a raw time series. Make sure to follow the derived series creation requirements so that the new derived series functions as expected.

When I update the logic of a derived series, will applications that consume that derived series also update accordingly?¶

Derived series offer the advantage of immediate logic updates without consuming pipeline resources. Applications that render the derived series (usually a Quiver analysis or a Quiver dashboard embedded in an application) will calculate the derived series using the updated templated logic.

Can a TSP contain both derived series and non-derived time series?¶

Yes! A TSP can have a mix of values which are either a series ID for (non-derived) time series or Codex template RIDs for derived series. This is typically used with a sensor object type. If your TSP is backed by multiple syncs and contains derived series, your Codex template RID for the derived series should not be in the qualified series ID format.

中文翻译¶

常见问题解答¶

派生序列逻辑支持哪些 Quiver 卡片？¶

派生序列逻辑支持以下 Quiver 卡片：

对象时间序列属性（Object time series property）
过滤时间序列（Filter time series）
导数（Derivative）
DSP 滤波器（DSP filter）
积分（Integral）
采样（Sample）
分段统计（Segment statistics）
时间序列公式（Time series formula）
合并时间序列（Combine time series）
累计聚合（Cumulative aggregate）
周期聚合（Periodic aggregate）
滚动聚合（Rolling aggregate）
相对时间序列（Relative time series）
偏移时间序列（Shift time series）
偏移日期/时间（Shift date/time）
数值参数（Numeric parameter）
字符串参数（String parameter）
日期/时间参数（Date/time parameter）
日期/时间范围参数（Date/time range parameter）
时长单位参数（Duration unit parameter）
布尔参数（Boolean parameter）

为什么我会收到多个根对象的错误？¶

如果在 Quiver 创建对话框中收到多个根对象的错误，您的分析很可能是从传感器对象类型（sensor object type）开始的，并且使用了两个不同的单个对象进行比较。例如，您可能从传感器对象类型中提取并比较了传感器对象 Inlet pressure 和 Outlet pressure，而不是从 Machine 1 根对象类型（root object type）开始。要创建派生序列，您需要从根对象实例（root object instance）开始。

请查阅设置模板化派生序列的逻辑要求，了解为什么派生序列必须从根对象的角度生成。

我可以将相同的派生序列模板应用于不同的对象吗？¶

可以。逻辑和计算是模板化的，因此可以从任何具有相同对象类型的绑定对象（bound object）的角度计算结果。请查阅在 Ontology 中存储派生序列部分，注意相同的 codex 模板被应用于不同的 Machine 对象或传感器对象。

URL 中的 codex 模板 RID 和派生序列 RID 有什么区别？¶

派生序列 RID 是用于保存资源名称、描述和其他元数据的 Quiver 资源的标识符。派生序列管理页面上列出的 codex 模板 RID 是对在运行时计算派生序列数据时使用的模板逻辑的引用。更多详情请查阅在 Ontology 中存储派生序列部分。

如何在 TSP 中引用特定版本的派生序列逻辑？¶

要引用特定版本的逻辑，您的 TSP 值应如下所示：

{"templateRid":"ri.codex-emu.main.template.8da5f759-4b...","templateVersion":"0.0.x"}

如果我的输入序列发生变化，派生序列会相应更新吗？¶

由于派生序列是在读取时计算的，因此派生序列会显示输入序列中所有可用的更新。例如，在 Quiver 中可见的输入序列的任何更改也会反映在派生序列中。

我可以将派生序列用作另一个派生序列的输入吗？¶

可以。派生序列可以嵌套使用，用于计算另一个派生序列。您可以像使用原始时间序列（raw time series）一样使用派生序列。请确保遵循派生序列创建要求，以便新的派生序列按预期运行。

当我更新派生序列的逻辑时，使用该派生序列的应用程序会相应更新吗？¶

派生序列的优势在于可以立即更新逻辑，而无需消耗流水线资源。渲染派生序列的应用程序（通常是嵌入在应用程序中的 Quiver 分析或 Quiver 仪表板）将使用更新后的模板逻辑来计算派生序列。

TSP 可以同时包含派生序列和非派生时间序列吗？¶

可以！TSP 可以混合包含两种值：非派生时间序列的序列 ID（series ID）或派生序列的 Codex 模板 RID。这通常与传感器对象类型一起使用。如果您的 TSP 由多个同步支持并包含派生序列，则派生序列的 Codex 模板 RID 不应采用限定序列 ID（qualified series ID）格式。