Gravitational wave(GW)observations have provided a new avenue to explore the universe.It has opened a unique window to test General Relativity(GR)and investigate the nature of gravity.In this thesis,through the powerful tool of GWs,we test gravity theories with different methods and in different frequency bands.Our discussion includes constraints given by full-Bayesian inference with real data from current ground-based detectors,a forecasting study of constraining capabilities of future spacebased detectors,and a test on non-GR polarization with Pulsar Timing Array(PTA).Firstly,we consider the limits obtained through real GW data.Although,various model-independent tests have been performed by LIGO-Virgo-KAGRA(LVK)collaboration in previous works.Those theory-agnostic constraints cannot always be directly projected to constraints on specific theories.It is still necessary to perform new parameter estimations to place constraints on parameters in specific models.In Chapter 3,we consider three models of scalar-tensor theories:the Brans-Dicke theory,the theory with spontaneous scalarization phenomena,screened modified gravity,and use neutron star-black hole GW events to place constraints on these three specific models.The results show that for the theory with spontaneous scalarization phenomena,the constraints given by GWs are comparable to those previously obtained by pulsar timing,but for the other two models,the constraints given by GWs are not competitive with the current solar system constraints.A compromise between tests on specific models and most general modelindependent searches for signals beyond GR,is testing gravity theories through the viewpoint of the effective field theory,which can combine the advantages of the two previous approaches.The constraints obtained this approach not only have explicit physical implications,but can also cover different theories simultaneously.In chapter 4,we consider the gauge-invariant linearized-gravity sector of the Standard Model Extension(SME),which is a piece of the effective field theory,to investigate the Lorentz invariance during propagation of GWs.In the presence of Lorentz violation,effects of anisotropy,birefringence,and dispersion can be induced during propagation of GWs,which will deform the GW received by detectors.Employing the waveform with deformation induced by Lorentz violation,we perform the Bayesian analysis with confident candidates in the last GW events catalog.We consider two cases associated with the lowest mass dimension d=5,6,the constraints on the coefficients of d=5 are in the order of 10-15m,and constraints for d=6 are in the order of 10-11m2.In the near future,space-borne GW detectors will further expand the horizon of GW astronomy in low frequency.In Chapter 5,we investigate the potential constraints given by future space-borne detectors on screened modified gravity which is a unified theoretical framework describing the scalar-tensor gravity with screening mechanism.We assume an extreme mass ratio inspiral(EMRI)system located in the Virgo cluster as the GW source.For an EMRI composed by a massive black hole and a neutron star,we find that its GW signal can be detected at high significance levels and the screen parameter of neutron star can be constrained to O(10-5),but for an EMRI composed by a massive black hole and a white dwarf,it is difficult to be detected.For specific SMG models,including the chameleon,symmetron,and dilaton models,we find that the constraints given by GWs are complementary with those given by Cassini experiment,but weaker than those given by lunar laser ranging and pulsar observations due to the strong gravitational potential in the surface of neutron stars.For lower frequency band,pulsar timing array(PTA)provides an effective way to detect GWs at nanoHertz.Because pulsars distribute in different directions,the directional complexity makes PTA especially suitable for probing different polarization modes of GWs.In chapter 6,we investigate the capabilities of current and future PTA to constrain additional polarization modes of GWs.We consider monochromatic GWs from an individual source and use the limits of amplitude parameters(cb,csn,cse,cl)to quantify the capabilities of PTA to constrain the four additional polarization.Assuming a supermassive binary black hole system with chirp mass M=8.77 × 108M☉ and frequency f=10-9Hz located in the Virgo cluster as the GW source,we find that only if cb>0.00106,cl>0.00217,cse>0.00271,csn>0.00141,the additional polarization modes can be detected by current PTA.For future PTA,the limits can be enhanced by two orders of magnitude. |