Week 7 State of charge estimation

Aim: To simulate the test cases using the harness dashboard and write a detailed report, explain coulomb counting method.

Objective:

1. Study the model.
2. Simulate the 3 test cases from the harness dashboard.
3. Study the output results.
4. Write a detailed report.
5. Explain coulomb counting method.
6. Explain how coulomb counting method used in the above model.

Body of the content:

Steps to open the file:

1. Using open from the MATLAB window open the folders were the files are extracted.
2. Open plant folder.
3. Open Battery Model.slx file.
4. Click on open project and model.
5. Now after simulink window is open, from the current folder open Battery_System.prj file.
6. Now open the BMS System from the top bar.
7. Now open the harness dashboard as: Test->System->BMS_ClosedLoop_Harness_Dashboard

Simulation of BMS cases:

1. The model represents the basic Battery managment system used for Lithium ion batteries.
2. The simulation is run for three different cases of the vehicle states(driving, charging and standby) with respect to time.
3. There are two battery pack, one consists of the 6 cells and another battery pack contains 96 cells.
4. The below figure shows the harness dashboard:
   • 
   • It mainly consists of:
     1. Test sequence variant: Used to select the case.
     2. BMS Closed loop:
     3. Fault state indicators: Will indicate any faults/warning.
5. The below figure shows the BMS system and Battery pack configuration selection:
   • 
   • The BMS closed loop consists of:
     1. BMS ECU.
     2. PLANT.
   • 
6. The BMS ECU consists of the following:
   • 
     1. Current power limit calculation.
     2. State machine: Defines the state of operation of the BMS.
        1. There are four states:
           1. State one defines the state of BMS i.e. standby, driving, charging and fault.
           2. The second switches the fault state if there are any unsafe parameters(voltage, temperature etc)
           3. The third state defines the contractors ON and OFF for the precharge and high voltage circuits.(avoid any large inrush current at start)
           4. The fourth defines the contractors switching of invertor.
     3. SOC estimation.
        1. 
        2. Three State of Charge estimation methods are used:
           1. Coulomb Count Method.
           2. Unscented Kalman filter (UKF).
           3. Extended Kalman filter (EKF).
     4. Balancing Logic.
7. The PLANT consists of the following arrangment as follows:
   • 
8. Case 1
   1. Here the first test sequence case is selected:
      • 
   2. First case is selected:
      • 
   3. For the Test case 1 these are the following states:
      1. Driving - Till 3000 seconds.
      2. Balancing - next 1000 seconds.
      3. Charging and Balancing - next 5000 seconds.
      4. Post charge balancing _ throughout
      5. 
   4. Results of Case 1:
      • 
      • Cell 6 shows the highest temperature and cell 1 shows the lowest temperature.
      • During driving and charging the temperature rises exponentially.
      • During driving (0-3000sec) the voltages and current fluctuates as per the conditions.
      • During standby(3000-4000sec) the voltage and current remains constant.
      • During charging (4000-9000 seconds) the voltage and current attains peaks (cell 2 voltage -4.2 volts).
      • SOC (CCM) rises to 100%.
      • After 9000 seconds the state is standby, the current flows is zero and SOC remians same.
9. Case - 2
   1. Second sequence is selected.
      • 
   2. For this following case these are the states;
      1. Driving - Till 10000 seconds.
      2. Charging - after 10000 seconds throughout.
   3. Here the battery SOC drops to 0 before 9000 seconds, so no further charging occurs as it encounters error.
      1. 
   4. Results of Case 2:
      • 
      • Here Driving state occurs till 10000 seconds.
      • After than charging occurs.
      • Cell 6 shows the highest temperature and cell 1 shows the lowest temperature.
      • The voltage decreases with respect to time.
      • Voltage varies from 3.9volts to 3volts.
      • Max current discharge is 70A.
      • The SOC is initially 80% and goes to 0% a 8692 seconds.
      • So no charging occurs as the SOC becomes 0% the error is encountered.
10. Case - 3
    1. Selecting third sequence.
       • 
    2. For this following case these are the states;
       • 
    3. Results of case 3
       • 
       • 
       • 
       • In this case only charging is done.
       • Initially the battery pack is at 80 % of SOC and then its charged to 100%, 94% and 93%(CCM, UKF, EKF respectively) SOC in 2500 seconds.
       • All six cells temperatures rise(according to the rate of charge) to peaks till the battery pack is fully charged.
       • Peak charge current is around 30A.
       • Cell 6 shows the highest temperature and cell 1 shows the lowest temperature.

SOC Estimation:

There are various methods used for the estimation of SOC of the battery:

• 

In the present model three methods are used for the SOC calculation:

1. Coulomb Count Method.
2. Unscented Kalman filter (UKF).
3. Extended Kalman filter (EKF).

2. Coulomb Counting:

1. It is a method of estimation of SOC, in which the discharging current of the battery is measured and the value of current is integrated over the time to estimate the remaning SOC.
2. Here the Current discharge rate (I(t)) and the previous SOC(t-1) value is taken to calculate the estimates present SOC(t).
3. The accuracy of the coulomb counting method is greatly affected by temperature, battery history,cycle life and discharge current.
4. SOC is calculated by the equation:
   • `SOC(t) = SOC(t-1) + I(t)*n*deltat`

SOC estimation in the given model:

1. The below figure shows the arrangment of the BMS ECU:
   • 
2. The SOC estimation module is as follows:
   • 
3. In coulomb count method:
   1. 
   2. 
   3. Working:
      1. Here the current and temperature are given as inputs.
      2. Then by using current value the discharge current per hour is calculated by using the gain block(1/3600).
      3. Then a lookup table is used to find the capacity at specific temperature.
      4. Then two of these values are divided and the values are integrated(by discreate time integrator) to estimate the SOC.
      5. The upper and lower limit are 0 to 1.
   4. Note: The Coulomb counting methods icnorporates errors in the current value calculation which leads to the error in the calculations, it may not be tursted if used as standalone method.

Conclusions:

1. The three test cases are simulated and studied and the detailed report is given.
2. Methods of SOC estimation are studied and Coulomb count method is explianed.
3. The SOC estimation processs in the BMS using Coulomb counting is explained.

References:

1. Wen-Yeau Chang, "The State of Charge Estimating Methods for Battery: A Review", International Scholarly Research Notices, vol. 2013, Article ID 953792, 7 pages, 2013. https://doi.org/10.1155/2013/953792
2. Matúš Danko, Juraj Adamec, Michal Taraba, Peter Drgona,
   Overview of batteries State of Charge estimation methods,
   Transportation Research Procedia,
   Volume 40,
   2019,
   Pages 186-192,
   ISSN 2352-1465,
   https://doi.org/10.1016/j.trpro.2019.07.029