Journals for active magnetic bearing related work (AMB)

1. International Journal of Control, Automation and Systems, monthly, 55 days review, Link
2. Journal of Dynamical and Control Systems, once in 3 months, 5 days decisionLink
3. International Journal of Control, Automation and Systems, monthly, 55 days review, Link
4. IFAC Journal of Systems and Control, 2 weeks review,Link
5. International Journal of Dynamics and Control, once in 2 months, Link
6. Mechanisms and machine theory, monthly journal,3 weeks, Link
7. Mechanical systems and signal processing, monthly journal, Link
8. Sadhana,Once in 3 months, 40 days decision, Link
9. Journal of Vibration Engineering and Technologies, irregular publishing, First decisiton 5 days, Link
10. Journal of dynamics systems measurement and control, monthly journal, Link
11. Mechatronics, 3 months review, Link
12. Journal of Verification Validation and Uncertainity quantification, Once in 3 monthsLink
13. Archive of Mechanical engineering, once in 3 months, Link
14. The Journal of Vibration and Acoustics , once in 2 months, Link
15. International Journal of Systems, Control and Communications, once in 3 months, Link
16. International journal of control, automation and systems, open access, monthly journal,Link
17. European Journal of Control , 2 weeks publication, Link
18. Control engineering practice,once a year, Link
19. IEEE Transactions on Control Systems Technology , once in 2 months, paid Link
20. Frontiers in control engineering, 77 days review, Paid,Link
21. Journal of control science and engineering, paid, 31 days decision, Link

Paper Ideas to follow up

Current paper 2 plan (29 Oct 2022)

1. Show why 2D gain scheduling is required  +  2D gain scheduling considering speed and levitated height + LQR filter  for designing the PID controller in the gain scheduling + as you go lower, as you go closer to the bottom actuator, make the objective function take sign of the error into consideration, rather than simple xQx, quadratic under the effect of 1X disturbing force.

Notch filter based(29 Oct 2022)

1.  Instead of notching filter (Herzog 1996) + Tracking control for machining spindles, shift the notch filter central frequency around to balance between current consumption and vibration control. Also, based on the response to the 1X excitation force with the shifted notch filter, change the reference input adaptively: Called input trajectory optimization in https://ieeexplore.ieee.org/document/6859290, Nanu Chen

System identification + health monitoring (2 Nov 2022)

1. Do unbalance identification for health monitoring when the plant is originally controlled by LPV/gain scheduled control where RPM itself is a scheduling variable.

Active disturbance rejection control(29 Oct 2022)

ADRC is uses extended state observer ESO

ESO can be made time dependent, TADRC

Can you make ESO as parameter dependent (RPM ) dependent.

If ESO is itself state dependent (position) of plate, then does ESO become non linear. Can ADRC control it? "On disturbance rejection in magnetic levitation", Wei Wei, Wenchao Xue.

Algebraic successive integration scheme(29 Oct 2022)

1. FEM model with flexible rotor with algebraic successive integration for unabalance estimation and shaft model identification
2. Rigid rotor on AMB with algebraic successive integration for unbalance identification + km identification
3. Rigid rotor + AMB + algebraic successive integration for unbalance identification + Changing RPM + Notch filter tracking based on Herzog (03 Nov 2022)
4. Rigid rotor + AMB + algebraic successive integration for unbalance identification + Changing RPM + Notch filter tracking based on Herzog + With run out error on sensor disk (03 Nov 2022)
5. Impart thrust disk face out and thrust disk tilt on the 4 actuator supported thrust AMB, with rotation, in the first paper frame work on algebraic successive integration
   

 Also see older ideas:

https://balajimitplane.blogspot.com/2022/10/dyncont7-paper-publication-ideas-to.html

DynCont#10 Linearising closed loop in simulink, Equilibrium point, trim point.

1.  When linearising a non linear plant, about a trim point, always open the feed back loop
2. Do not place input perturbation and output pickup in simulink in the closed loop
3. Linearising is applicable for unstable systems, so open the feedback loop and linearise the plant alone.
4. Important : Trim point is not Equilibrium point
5. At equilibrium point, all state vector derivatives are zero
6. At trim point, only the constrained state vector derivatives are zero. Example, linearising the aircraft model at a constant forward speed.. 

 

From Analyzing Models (Getting Started) (urv.es)

DGI direct gasoline injection system drive details

1. Purchase link: GDI_CRDI_Direct_Injector_Tester (autodiagnosticsandpublishing.com)
2.  Good article on current sense and drive circuit : Link
3. NCS 333 Operational amplifier.. Link
4. BSS123 Mosfet: Link
5. Using pressurised nitrogen bottle and to pressurise the fuel line instead of using a DGI pump: Mdpi paper on bike engines DGI optimisation.
6. A good reference with details on the circuit design of the DGI injector drive circuit and the power stage: 2011 Paper (Wenchang Sai)
7. High pressure DGI injector drive circuit using 3 Mosfets : 

   

   High pressure 3 stage pulses injector drive circuit PCB from MDPI paper

   
   
8. Study of the injector drive circuit for a high pressure GDI injector, Simulation study: Link
9. Some photos of the experimental setup in Professor Krishna Sahu's Lab, thermodynamics:

10. 

    Ford injector 

    
    

    

    Fuel rail with the injector

    
    

    

    Cam driven DGI pump

    
    

    

    Encoder for triggering the solenoid of the cam driven DGI pump

    
    

Specifying design point of a gas turbine: Lesson learnt.

1. Design point of the compressor for a gas turbine is specified at 100% RPM, 101325 Pascal inlet pressure and 288.15 at  0 Mach = ISA SLS
2. Design point performance of the compressor is specified as 3 numbers
1. Corrected mass flow rate
2. pressure ratio
3. efficiency
3. The inlet face of the compressor is at ISA SLS
4. If you have inlet pressure drop in the bell mouth inlet at ISA SLS, then actual mass flow rate seen by the compressor will be less than the corrected mass flow rate. 
5. For design point calculation, do not give inlet pressure drop in the bell mouth inlet.
6. If inlet pressure drop in the bell mouth inlet is zero, the actual mass flow rate and corrected mass flow rate will be same and it is easier to generate the compressor characteristics curve from CFD.  
7. Commercial softwares such as GSP change the design point Mach number automatically when you specify the inlet pressure drop. For example, if you give design point ISA SLS inlet pressure drop as 0.98, it will automatically raise Mach number to 0.17 and reduce the ambient temperature by 2 degrees. 
8. The raise in total pressure at the inlet due to Mach number compensates for the inlet pressure drop and the compressor corrected mass flow rate and actual mass flow rate are same. inlet face of compressor sees 101325 after the total pressure rise due to mach number and total pressure drop due to inlet pressure loss.
9. The reduction of ambient temperature by -2 degrees compensates for ram temperature rise and compressor inlet temperature  becomes 288.15 after Mach number total temperature rise 
10. If you want zero mach number with inlet pressure drop, use the "flight" option in GSP and not the design point calculation option. 

DynCont#9 Simple Looped AI and AO using PXIe, nidaqmx, Python, Labview, NIMAX

1. It took 3 days of poking, trial and error to get this python program working. This can only be run with hardware (PXIe system, with NI 6356 card. It has 8 diff analog inputs and 2 analog outputs
2. No, you cannot use 8 differential as 16 ground referenced single ended channels. Pay more for that to Hungary.
   
3.  LABVIEW not essential
4. NIMAX drivers required for configuring connections, checking hardware using test panels
5. nidaqmx python library used instead of labview.
6. LABVIEW is a mind melting mixed veg noodles, that is 2 days old.
7. This program is a simple looped AI and AO 
8. It is not yet software timed. To do software timed loop, use a while loop inside the outer loop to check if your control loop update time has elapsed and then continue with the next outer loop pass.
9. In software timed loop, if one outer loop pass takes more time than your control update time, then increase control update time.
10. In hardware timed loop, use call backs from input task that is triggered after samples are read from the buffer to the computer.
    
11. Program below samples  4 channels of analog in TCV, BCV, Excitation and Eddy. Writes 2 outputs TCV and BCV. TCV and BCV are looped into analog in also.
12. No hardware or software timing, no processing using the sampled values to calculate the outputs. Only seeing how fast simple loop runs. This is the bare minimum control update possible. This uses stream readers and stream writers. Uses shared clock between analog in and analog out. Uses a trigger from analog in to start analog out.
    

DynCont#8 NIDAMX PID control Labview NI PXI

Tasks start and stop behavior:

1.  If you are using finite samples, no need to stop after reading input. Always close only at the end of the complete program
2. Create task once in the beginning of the program
3. Start task once when using finite samples.
4. If you allow regeneration, call backs will not work: Source = https://forums.ni.com/t5/Multifunction-DAQ/Continuous-write-analog-voltage-NI-cDAQ-9178-with-callbacks/td-p/4036271/page/2?profile.language=en

Don't use  USB/ETHERNET

If you are using a USB DAQ, keep in mind that the devices are optimised for data throughput not response latency. You don't want to service tasks (e.g. pull data off the board) more than about 5 times a second. Consequently you should plan to perform larger operations on a USB DAQ than on a PCI or PCIe-based device.

Source: https://github.com/tenss/Python_DAQmx_examples

Software timed IO:

Source = https://www.ni.com/docs/en-US/bundle/ni-daqmx/page/mxcncpts/controlappcase5_2.html

1. For software timing, the software and operating system determines the rate at which the loop executes. Software timing is not deterministic. Controlling a while loop and using the Wait Until Next ms Multiple VI to handle timing is an example of a software-timed loop. 
   
2. IO is triggered by the python code in the PC
3. Use this mode when hardware time IO is not available
4. Timing will have jitter due to software errors
5. Configure the Timed Loop to run at the desired rate. 
   

Hardware timed IO:

source = https://www.ni.com/docs/en-US/bundle/ni-daqmx/page/mxcncpts/controlappcase1.html

1. The current iteration's output samples are guaranteed to be aligned with the next iteration's input samples. 
2. Use the DAQmx Wait For Next Sample Clock function
3. Read, process, and write operations are confined to the time available between the moment the device starts acquiring data and the moment the next sample clock edge arrives.  
4. Does exactly what we need = An example of this kind of application is an analog control loop that reads samples from a specific number of analog input channels, processes the data using a control algorithm (such as PID), and writes new control values to the analog output channels.  

Software develop helps

1. https://github.com/mjablons1/twingo
2. https://github.com/toastytato/DAQ_Interface
3. https://github.com/czimm79/MuControl-release
4. Relevant = https://github.com/tenss/Python_DAQmx_examples
5. Relevant = https://github.com/mjablons1/nidaqmx-continuous-analog-io
6. Relevant = https://github.com/mjablons1/nidaqmx-continuous-analog-io 
   
7. https://github.com/tenss/SimplePyScanner
8. https://github.com/petebachant/daqmx
   
9. https://github.com/tenss/Python_DAQmx_examples/blob/master/pynidaqmxegs/mixed/AOandAI_sharedClock.py
   

DynCont#7 Paper publication Ideas to follow up.

 Controls

1. PD control steady stater error in servo control analysis and physical meaning
2. PD / PID + Mapping of the initial trajectory required + Using preknown offset to modify the desired path beforehand from system identification +  Tune this controller using LQR using closed form algebraic solution of Ricatti equation 
3. Weight the negative error more than the positive error to simulate the non allowable tool plunging while optimizing LQR (x^3-x^2-x)^2 rather than x'Qx. 
4. Model reference adaptive controller . 
5. Mode predictive controller for AMB 
6. RAMB. Flexible mode control using RAMB. How to damp the flexible modes.
7. Simulation model of shaft on AMB with Dynamic modal decomp, sparse SINDY, DMD with control , DMD without control!...
8. Kalman filter post on linked in https://bitbucket.org/tremaineconsultinggroup/observer_luenberger/src/master/
   

 Gas turbines

1. PID controller with gain scheduling for gas turbine
2. LQR / LQG controller for gas turbine
3. MPC controller for gas turbine
4. Model reference adaptive controller for gas turbine

Foil bearings

1.  Link model FEM from IISC course+ Psedospectral method
2. Orbit simulation
3. Stability analysis
   
4. FDM  / FVM / Commercial solver comparison for Foil bearing
5. Software for foil beairng load capacity estimation

DynCont#6 Optimal control resources for preparation

• Video lectures =  AA4CC lectures (Lecture 2 missing)
• Video lectures = Neuro match academy, good explanation, https://www.youtube.com/playlist?list=PLkBQOLLbi18NSMbvuDS2WaXLWctg3QArY
• Video lectures = Barjeev Tyagi, IIT Roorkee, https://nptel.ac.in/courses/108107098
• Video lectures - Shyam Kamal, IIT BHU, EE564,  https://www.youtube.com/playlist?list=PLHEIL5MR2cEGqWdrFvzxDwR6Lw-D5ETWY
• Video lectures = G D Ray, IIT KGDP, https://www.youtube.com/playlist?list=PLbMVogVj5nJQNzJT6sYZpB7H1G6WF0FZ4
  
• Optimal control 2022 https://www.youtube.com/playlist?list=PLZnJoM76RM6Iaf59ICcU9-DzztGZvK_52
  
• Kirk Introduction to optimal control book