Merge branch 'develop' of github.com:deepmodeling/abacus-develop into symmetry-debug

Author: dyzheng

.github/workflows/test.yml

@@ -5,6 +5,12 @@ on:
     branches:
       - develop
       - tryEigen
+      - update_MD
+      - ABACUS_2.2.0_beta
+      - deepks
+      - planewave
+      - pw_refactor
+      - TDDFT
 
 jobs:
   test:

CMakeLists.txt

@@ -275,6 +275,7 @@ target_link_libraries(${ABACUS_BIN_NAME}
     ri
     driver
     xc
+    esolver
     -lm
 )
 

doc/examples/band-struc.md

@@ -5,7 +5,7 @@
 This example shows how to calculate the energy band structure. Similar to the [DOS case](#dos.md), we first, do a ground-state energy calculation as in [this example](#basic-lcao.md) ***with one additional keyword in the INPUT file***:
 
 ```
-out_charge              1
+out_chg              1
 ```
 
 this will produce the converged charge density, which is contained in the file SPIN1_CHG. Copy the file along with the `STRU` file, the pseudopotential file and the atomic orbital file (and the local density matrix file onsite.dm if DFT+U is used) to the new working directory where we will do a non-self-consistent calculation. In this example, the potential is constructed from the ground-state charge density from the proceeding calculation. Now the INPUT file is like:
@@ -20,17 +20,17 @@ read_file_dir   ./
 
 #Parameters (Accuracy)
 ecutwfc 60
-niter 50
-dr2 1.0e-9
-ethr 1.0e-7
+scf_nmax 50
+scf_thr 1.0e-9
+pw_diag_thr 1.0e-7
 
 #Parameters (File)
-start_charge file
+init_chg file
 out_band 1
 
 #Parameters (Smearing)
-smearing gaussian
-sigma 0.02
+smearing_method gaussian
+smearing_sigma 0.02
 ```
 
 Here the the relevant k-point file KPT looks like,

doc/examples/basic-lcao.md

@@ -52,7 +52,7 @@ basis:
  ---------------------------------------------------------
  SELF-CONSISTENT : 
  ---------------------------------------------------------
- ITER   ETOT(eV)       EDIFF(eV)      DRHO2      TIME(s)    
+ ITER   ETOT(eV)       EDIFF(eV)      SCF_THR      TIME(s)    
  GE1    -2.138667e+02  0.000000e+00   1.652e-01  4.690e+00  
  GE2    -2.139153e+02  -4.859216e-02  3.480e-02  4.970e+00  
  GE3    -2.139161e+02  -8.407097e-04  4.131e-03  4.760e+00  

doc/examples/basic-pw.md

@@ -15,9 +15,9 @@ For this example, the input files are:
     basis_type pw
     suffix Si2_diamond
     symmetry 1
-    niter 60
-    dr2 1.0e-9
-    out_charge 1
+    scf_nmax 60
+    scf_thr 1.0e-9
+    out_chg 1
     ```
 
     The meanings of the above parameters are:
@@ -48,11 +48,11 @@ For this example, the input files are:
 
         Use symmetry(=1) or not(=0) in the calculation. The default value is 0.
 
-    - niter
+    - scf_nmax
         The maximal iteration number for electronic-structure calculations.
-    - dr2
+    - scf_thr
         Tolerance of the difference of charge density, below which the self-consistent calculation is considered to be converged.
-    - out_charge
+    - out_chg
         Print out the charge density(=1) or not(=0).
 
     A complete list of INPUT keyewords can be found in the [instruction](../input-main.md).
@@ -131,7 +131,7 @@ The following typical output information will be printed to the screen:
  -------------------------------------------
  SELF-CONSISTENT : 
  -------------------------------------------
- ITER   ETOT(eV)       EDIFF(eV)      DRHO2      CG_ITER    TIME(S)    
+ ITER   ETOT(eV)       EDIFF(eV)      SCF_THR      CG_ITER    TIME(S)    
  CG1    -2.154524e+02  0.000000e+00   6.855e-02  1.000e+00  1.290e+00  
  CG2    -2.154992e+02  -4.673475e-02  2.378e-03  2.000e+00  8.400e-01  
  CG3    -2.155050e+02  -5.882715e-03  8.220e-05  2.594e+00  9.000e-01  

doc/examples/berry-phase.md

@@ -25,14 +25,14 @@ suffix        PbTiO3
 ntype         3
 nbands        25 // number of bands
 ecutwfc       50 // Ry
-niter         20
+scf_nmax         20
 symmetry      0 // turn off symmetry
-ethr          1e-10
-smearing      gaussian // gaussian smearing
-sigma         0.002 // Ry
+pw_diag_thr          1e-10
+smearing_method      gaussian // gaussian smearing
+smearing_sigma         0.002 // Ry
 calculation   nscf // non-self-consistent calculation
 basis_type    lcao // atomic basis
-start_charge  file // read charge from files
+init_chg  file // read charge from files
 berry_phase   1 // calculate Berry phase
 gdir          3 // calculate polarization along c axis
 ```

doc/examples/dispersion.md

@@ -25,18 +25,18 @@ Below is an example using DFT + dispersion calculation:
 INPUT_PARAMETERS
 ntype                         2
 ecutwfc                       20
-dr2                           1e-06
-niter                         400
+scf_thr                           1e-06
+scf_nmax                         400
 basis_type                    lcao
 ks_solver                     genelpa
-smearing                      gaussian
-sigma                         0.02
+smearing_method                      gaussian
+smearing_sigma                         0.02
 mixing_type                   pulay
 mixing_beta                   0.4
 vdw_method                    d2
 calculation                   scf
-force                         1
-stress                        1
+cal_force                         1
+cal_stress                        1
 ```
 
 - STRU:
@@ -125,7 +125,7 @@ The files (`Sn_ONCV_PBE-1.0.upf`,`Te_ONCV_PBE-1.0.upf`,`Sn_pbe_9.0au_100Ry_2s2p2
  ---------------------------------------------------------
  SELF-CONSISTENT : 
  ---------------------------------------------------------
- ITER   ETOT(eV)       EDIFF(eV)      DRHO2      TIME(s)    
+ ITER   ETOT(eV)       EDIFF(eV)      SCF_THR      TIME(s)    
  GE1    -4.263615e+03  0.000000e+00   1.030e-01  1.030e+00  
  GE2    -4.262783e+03  8.325232e-01   6.041e-02  1.010e+00  
  GE3    -4.262610e+03  1.722243e-01   2.263e-02  1.040e+00  

doc/examples/dos.md

@@ -5,7 +5,7 @@
 The main task of this example is to calculate the density of states (DOS) of the system. At first, do a ground-state energy calculation as in [this example](#basic-lcao.md) ***with one additional keyword in the INPUT file***:
 
 ```
-out_charge              1
+out_chg              1
 ```
 
 this will produce the converged charge density, which is contained in the file SPIN1_CHG. Copy the file along with the `STRU` file, the pseudopotential file and the atomic orbital file (and the local density matrix file onsite.dm if DFT+U is used) to the new working directory where we will do a non-self-consistent calculation. In this example, the potential is constructed from the ground-state charge density from the proceeding calculation. Now the INPUT file is like:
@@ -22,18 +22,18 @@ read_file_dir   ./
 #Parameters (Accuracy)
 ecutwfc 60
 symmetry 1
-niter 50
-dr2 1.0e-9
-ethr 1.0e-7
+scf_nmax 50
+scf_thr 1.0e-9
+pw_diag_thr 1.0e-7
 
 #Parameters (File)
-start_charge file
+init_chg file
 out_dos 1
 dos_sigma 0.07
 
 #Parameters (Smearing)
-smearing gaussian
-sigma 0.02
+smearing_method gaussian
+smearing_sigma 0.02
 
 ```
 
@@ -42,12 +42,12 @@ Some parameters in the INPUT file are explained:
 
     choose which kind of calculation: scf calculation, nscf calculation, structure relaxation or Molecular Dynamics. Now we need to do one step of nscf calculation.
     Attention: This is a main variable of ABACUS, and for its more information please see the [list of input variables](../input-main.md).
-- ethr
+- pw_diag_thr
 
-    threshold for the CG method which diagonalizes the Hamiltonian to get eigenvalues and eigen wave functions. If one wants to do nscf calculation, ethr needs to be changed to a smaller account, typically smaller than 1.0e-3. Note that this parameter only apply to plane-wave calculations that employ the CG method to diagonalize the Hamiltonian.
+    threshold for the CG method which diagonalizes the Hamiltonian to get eigenvalues and eigen wave functions. If one wants to do nscf calculation, pw_diag_thr needs to be changed to a smaller account, typically smaller than 1.0e-3. Note that this parameter only apply to plane-wave calculations that employ the CG method to diagonalize the Hamiltonian.
 
     For LCAO calculations, this parameter will be neglected !
-- start_charge
+- init_chg
 
     the type of starting density. When doing scf calculation, this variable can be set ”atomic”. When doing nscf calculation, the charge density already exists(eg. in SPIN1_CHG), and the variable should be set as ”file”. It means the density will be read from the existing file SPIN1_CHG. For more information please see the [list of input variables](../input-main.md).
 

doc/examples/force.md

@@ -1,11 +1,11 @@
 # Force calculation and structure relaxation
 [back to main page](../../README.md)
 
-To calculate the atomic forces for a given structure without ion relaxation, set ‘calculation’ to ‘scf’, set input parameter ‘force’ to 1.
+To calculate the atomic forces for a given structure without ion relaxation, set ‘calculation’ to ‘scf’, set input parameter ‘cal_force’ to 1.
 
 ```
 calculation scf
-force 1
+cal_force 1
 ```
 
 To relax the atom position without change cell shape, one needs to add a few more parameters
@@ -15,23 +15,23 @@ default.
 ```
 calculation relax
 gamma_only 1
-nstep 100
+relax_nmax 100
 force_thr_ev 0.01
-move_method cg
+relax_method cg
 out_stru 1
-trust_radius_ini 0.5
+relax_bfgs_init 0.5
 ```
 
 - `calculation` relax
 
     relax atom positions with fixed lattice vectors.
-- `nstep`
+- `relax_nmax`
 
     the maximal number of ionic iteration steps.
 - `force_thr_ev`
 
     the threshold for the force, below which the geometry relaxation is considered to be converged. The unit is eV/Angstrom.
-- `move_method`
+- `relax_method`
 
     the algorithm used for geometry optimization. Possible choices are:
 - `cg`
@@ -44,7 +44,7 @@ trust_radius_ini 0.5
 
 - `cg_bfgs`
 
-    A mixed cg-bfgs method. For detail description, check out the variable cg_threshold in the [list of input keywords](../input-main.md#cg-threshold).
+    A mixed cg-bfgs method. For detail description, check out the variable relax_cg_thr in the [list of input keywords](../input-main.md#cg-threshold).
 
 - `sd`
 
@@ -54,7 +54,7 @@ trust_radius_ini 0.5
 
     output the structure of each step or not.
 
-- `trust_radius_ini`
+- `relax_bfgs_init`
 
     the initial radius of the relaxation. We advise you not to change this parameter, unless you are sure that the initial structure is close to the final structure.
 

doc/examples/md.md

@@ -17,22 +17,22 @@ gamma_only          1
 calculation         md
 symmetry            0
 
-nstep               10
 out_level           m
-move_method         cg
+relax_method         cg
 
-smearing            gaussian
-sigma               0.02
+smearing_method            gaussian
+smearing_sigma               0.02
 #Parameters (3.PW)
 ecutwfc             30
-dr2                 1e-5
-niter               100
+scf_thr                 1e-5
+scf_nmax               100
 
 #Parameters (5.LCAO)
 basis_type          lcao
 mixing_beta         0.4
-charge_extrap       second-order
+chg_extrap       second-order
 
+md_nstep          10   // md steps
 md_type           1    //choose ensemble
 md_dt               1    //time step
 md_tfirst           700  //the first target temperature