Recently, charge-momentum-energy-conserving schemes for the relativistic Vlasov-Maxwell system and the relativistic Fokker-Planck operator have been developed. Errors of the conservation laws can be amplified exponentially, so numerical simulations are often unstable with nonconservative algorithms. In the development of the structure-preserving schemes, the finite-difference of the Lorentz factor plays an important role to derive the discrete conservation laws. The mathematical background and the results of numerical experiments will be discussed in the presentation.