CPLUOS Astrophysics team meeting

Monday 18 Oct 2021, 16:30 → 17:55 Asia/Seoul

물리학과

https://uos-ac-kr.zoom.us/j/8264902652

16:30 → 17:45 COSMOLOGY

16:30 Dr.Sangnam Park's Report - FDM_Offset

Speaker: Sangnam Park

16:45 Hyeonmo's report - CDM Halo Evolution for Gadget2 Simulation

Speaker: Hyeonmo Koo (University of Seoul)

• Setting CDM Halo
  • Setting Radius and Half-mass Radius
    • M(r_h) = (1/2)M(R)
    • r_h = 1.51a
  • Setting Rotational Velocity
    • v(r) = (1-ϵ/175)((G(M)/r)^(1/2))
    • When ϵ > 0, damping occurs in the half-mass radius to cancel the gradual increase in the half-mass radius after relaxation.
    • In Time[Gadget Unit] vs r_(1/2)[kpc] graph, 1.0 <= ϵ <= 3.0
  • Displacement of Center of Mass(COM)
    • Average of COM shifts for 50 halos for each ϵ
      • Zeroth Evolution
      • First Evolution
  • Half-Mass Radius Evolution
    • Zeroth Evolution
    • First Evolution
  • Conclusion
    • ϵ = 2.5

17:00 Young's report. FoF and MST algorithm

Speaker: Young Ju

Random Data

• Comparison of Clustering Finding Algorithms in Random Data
  • 4 algorithms
    • MGS : fResolution = 10
    • MST : Linking length = 10
    • DBSCAN: eps = 10, min_samples = 10
    • Hierarchical : distance_threshold = 10
  • Results of clustering
    • (n, MGS, MST, DBSCAN, Hierarchical)
      • (1, 49, 50, 49, 49)
      • (2, 49, 50, 49, 49)
      • (4, 47, 50, 46, 46)
      • (6, 47, 48, 43, 43)
      • (8, 48, 48, 42, 42)
      • (10, 47, 44, 40, 40)
  • Results
    • 1. Result of simulated data with standard deviation = 1
    • 2. Result of simulated data with standard deviation = 2
    • 3. Result of simulated data with standard deviation = 4
    • 4. Result of simulated data with standard deviation = 6
    • 5. Result of simulated data with standard deviation = 8
    • 6. Result of simulated data with standard deviation = 10

Noise Data

• Comparison of Clustering Finding Algorithms in Noise Data
  • 4 algorithms
    • MGS : fResolution = 10
    • MST : Linking length = 10
    • DBSCAN: eps = 10, min_samples = 10
    • Hierarchical : distance_threshold = 10
  • Noise = 1000
    • (minimum number, MGS, MST, DBSCAN, Hierarchical)
      • (3, 81, 76, 40, 71)
      • (10, 47, 44, 40, 39)
      • (20, 47, 44, 40, 39)
      • (50, 47, 44, 40, 39)
  • Noise = 5000
    • (minimum number, MGS, MST, DBSCAN, Hierarchical)
      • (3, 570, 228, 40, 226)
      • (10, 140, 57, 39, 55)
      • (20, 54, 30, 39, 28)
      • (50, 47, 18, 39, 16)
  • Noise = 10000
    • (minimum number, MGS, MST, DBSCAN, Hierarchical)
      • (3, 616, 27, 167, 27)
      • (10, 357, 1, 148, 1)
      • (20, 157, 1, 66, 1)
      • (50, 52, 1, 33, 1)
  • Noise = 50000
    • (minimum number, MGS, MST, DBSCAN, Hierarchical)
      • (3, 504, 1, 1, 1)
      • (10, 480, 1, 1, 1)
      • (20, 459, 1, 1, 1)
      • (50, 346, 1, 1, 1)
      • From 50000 noise data, only MGS computed small clusters. The other three algorithms returned erroneous results where one huge cluster contained all data points.

17:15 CHOA(Cosmology of High-Order Statistics)

Speaker: Se Yeon Hwang

17:30 Dr.Sabiu's Report - 21cm paper referee report

Speaker: Dr Cristiano Sabiu (University of Seoul)

17:45 → 17:55 Discussion