Learning_in_progress

Thank you John, this helped tremendously 😄

Hello again 🙂 I am trying to estimate how much of a batch results in finished product. To accomplish this my plan was to: 1. Find the maximum value of two signals for each step of the stepwise signal. 2. For each step, take the biggest value and subtract from it the stepwise signal value. 3. Display the differences (lost product) along with the start and end time of the batches (stepwise signal). I started off by finding where the stepwise signal changed by using this formula: $b.valueSearch(5min, isNotBetween(-1000,1000)) Then I took the inverse of these points to get capsules for each step (not shown in image). Then I tried to get the maximum values of one of the signals for each step-capsule: $p1.aggregate(maxValue(), $s.removeLongerThan(5d), maxKey()).toStep() Unfortunately, this did not work as intended because some peaks cross from one step to another, which creates a fake max value for that step. However, this might not cause any problems as long as the real peak from the other signal which belongs to that step is greater than the fake peak. Another issue was that the new stepwise signal created did not match the old stepwise signal. Hopefully there is a better method for doing this.

Thank you for your help Chris! ^^

How do I make a capsule from a condition? I am very new to Seeq, and it is very much possible that I am doing this in a more complicated way than necessary, so I will try to explain the goal and problem I have. I have a signal which I have filtered to only show me the interesting regions. In these regions, the signal has a start value of about 38 and slowly decreases until around 34 around which there is a small bump and then the signal decreases again until it reaches a another bump between 25 and 23. There is some variations for the different regions. My goal is to find the time between the lowest point of the first bump and the first occurrence after where the signal has a value of 25.2. Depending on the second bump, there might be several such points after the first bump, or only one. What I have done: I took the derivative of the signal and smoothed it using the agileFilter to be able pick out the first bump (This also gave me the second bump, but I don't care about that). Then I did a valueSearch to get capsules for when the smooth derivative is greater than 0.001 for at least 1 minute, which gave me a capsule for each peak. Then I only focused on the first bump/peak by using intersect to get capsules for when the filtered signal (blue) is greater than 30 and overlapping with the smooth derivative peaks (capsules). Then I tried to get the minimum value of the signal that lies within the duration of the newest capsule, using the minValue function. However, this did not work, because the newest capsule is actually a condition. Does anyone know how can I make the condition into a capsule, or get around this issue? Any help would be greatly appreciated 🙂 PS. I am trying to do this using the Formula tool.