Getting rid of spikes in sensor data

When performing analyses, it's important to pre-process data in order to get good results. One such example is getting rid of spikes in sensor data. Four different ways are shown here to get rid of said spikes and get the results you are looking for.

Option A: Create a filter based on a value-based search

Perform a value-based search to look for values smaller than the spikes.
Keep results to exclude the spikes.
When looking at the statistics table, the values are now displayed without considering the spikes.

Advantages: Quick & Easy

Disadvantage: Less flexible for next steps

Option B: Clamping the data by using a formula

Create a formula to over clamp the spikes to a lower value (threshold for maximum of spikes).
```
if(A>250,250,A)  
```
Variable mapping:
A = Original tag with spikes
When looking at the statistics table, the values are now displayed with an upper limit for the spikes.

Advantages: Simple concept, Flexible

Disadvantage: Small loss of accuracy

Option C: Creating a time-shifted tag by using another formula

Create another formula to use the last good point as a reference point.
```
if(A>250,B,A) 
```
Variable mapping:
A = Original tag with spikes
B = Original tag with spikes, shifted by 1 minute
When looking at the statistics table, the values are now displayed precisely.

Advantage: Accurate

Disadvantage: May require fine-tuning

Option D: Creating an interpolated tag by using another formula

Create a formula to linearly interpolate between the start and the end value of an interval when the if-condition is not true.
```
if(A<250,A,sqrt(-1)) 
```
Variable mapping:
A = Original tag with spikes
When looking at the statistics table, the values are now displayed precisely like in option C.