equinor TimeSeriesAnalysis issues

equinor / TimeSeriesAnalysis

Library that combines control engineering, dynamic simulation and machine learning on time-series. Developed to describe industrial processes and -automation. Lightweight, robust and fast for use in advanced analytics. Built on .NET to run anywhere.

https://equinor.github.io/TimeSeriesAnalysis

Apache License 2.0

11 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

[Snyk] Security upgrade Microsoft.Extensions.Configuration.Json from 6.0.0 to 7.0.0

#136 steinelg closed 1 hour ago
0
[Snyk] Security upgrade Microsoft.Extensions.Configuration.Json from 7.0.0 to 8.0.1

#135 steinelg closed 1 hour ago
0
GainSchedSimulator is a candiate to be removed

#134 steinelg opened 1 week ago
0
better python integration (pythonnet?)

#133 steinelg opened 1 week ago
1
Update .NET version to 8.0?

#132 steinelg opened 1 week ago
0
Create CODE_OF_CONDUCT.md

#128 steinelg closed 5 months ago
0
Go through warnings related to docfx and missing "snippets"

#127 steinelg closed 8 months ago
0
docfx workflow builds but does not deploy to pages

#126 steinelg closed 7 months ago
1
Use MinVer

#125 steinelg opened 8 months ago
0
GettingStarted ex4 simulation/identification no longer works as expected

#124 steinelg closed 8 months ago
1
ProcessControl examples runs, but errounously has warning "PlantSimulatorInitializer: computational loop init failed"

#123 steinelg opened 8 months ago
0
GainSchedulingIdentification: Better automatic selection of tresholds

#122 steinelg opened 8 months ago
0
Gain-schedueld model identification sometimes give too high gains if there are too many thresholds

#121 steinelg opened 8 months ago
0
Handover GainSchedIdentify

#120 gunnska closed 8 months ago
1
extrapolation for U outside Umin/Umax in tuning set should be only linear

#119 steinelg opened 1 year ago
0
ClosedLoopIdentifier does not attempt to estimate time delays

#118 steinelg opened 1 year ago
0
ClosedLoopUnitIdentifier.Ident() does not correctly identify gains or disturbances when the unit model has multiple inputs (MISO-closed loop systems)

#117 steinelg closed 1 year ago
2
Pid identification currently does not work for gain-scheduled systems

#116 steinelg opened 1 year ago
0
process identification does not work for processs controlled by gain-scheduling

#115 steinelg opened 1 year ago
0
ClosedLoopUnitIdentifier only works for a single PID-controller

#114 steinelg opened 1 year ago
0
BasicPID-setpointstep example estiamtes far too high a gain when using "gain global search"

#113 steinelg closed 1 year ago
1
docfx documentation artifacts should be created by github actions

#112 steinelg opened 1 year ago
0
Test job fails - NUnit Exception "System.InvalidCastException"

#111 steinelg closed 8 months ago
1
need to verify how the disturbance estimation performs when there is a setpoint change

#110 steinelg closed 8 months ago
2
ClosedLoopDisturbanceEstimator currently does not try to identify model dynamic terms, only finds static models

#109 steinelg closed 8 months ago
1
Sign of process sometimes wrong in closed-loop estimation

#108 steinelg closed 1 year ago
2
PidIdentifier should estimate u0 for P-only controllers

#107 steinelg closed 8 months ago
2
PidIdentifier is vulnerable to noise in ymeas

#106 steinelg closed 1 year ago
1
make it possible to adjust if sas scales kp from input object (so that it can be affected from front-end)

#105 steinelg opened 1 year ago
0
verify that identification is able to take the effect of specified scaling into account

#104 steinelg opened 1 year ago
0
PidIdentifier: lacking unit test to verify edge cases and detection/warning

#103 steinelg closed 8 months ago
1
PidIdentfier should give out uncertainty on paramters like UnitIdentifier

#102 steinelg opened 1 year ago
0
PidIdentifier does not give nice output like UnitIdentifer

#101 steinelg closed 1 year ago
1
The use of special strings in signal names to denote their type excludes directly mapping signals to tagnames for real-world data

#100 steinelg closed 1 year ago
1
Normalization (Unorm) different from one causes incorrect plant-simulations

#99 steinelg closed 8 months ago
2
Disturbance estimation should be able to estimate disturbance for a given model in a PlantSimulator, rather than only UnitSimulator-models

#98 steinelg closed 1 year ago
1
UnitIdentifier should separate between inputdatasets and simulated datasets

#97 steinelg closed 1 year ago
1
All methods require disturbance to be zero at start of data set

#96 steinelg closed 8 months ago
2
PlantSimulator.Simulate error text should be collected in a form suitable for sending over web-api

#95 steinelg closed 1 year ago
1
UnitModel and PidModel have timbeBase_S =0 when de-serialized

#94 steinelg closed 2 years ago
1
simData output of PlantSimualtor.Simulate does not have correct timestamps

#93 steinelg closed 2 years ago
0
Need ability to map "tagnames" in IMS or a csv-input file to internal logical names.

#92 steinelg closed 1 year ago
1
Port DisturbanceEstimator class for PidIdent repo

#91 steinelg closed 2 years ago
1
PlantSimulator configuration should add serialiation/desirialization to json

#90 steinelg closed 2 years ago
1
Create the ability to automatically divide a TimeSeriesDataSet to many UnitDataSets,each identified with UnitIdenfier and then producing a PlantSimulator

#89 steinelg closed 8 months ago
1
non-observable gains in the model tend to take on extreme values-add regularization to force these to zero instead.

#88 steinelg closed 2 years ago
1
Move everything indice-related form Vec to new class "Indices"

#87 steinelg closed 2 years ago
0
Vec. MakeIndexArray forgets the last value

#86 steinelg closed 8 months ago
1
As N increases, time-constant estimates suffer (noise enters model through diff-)

#85 steinelg opened 2 years ago
1
Add a plot which explains the logic of how PlantSimulator initializes simulations in steady-state

#84 steinelg opened 2 years ago
0