Study Of Multi-band Observations And Search Of Pulsars And Chemically Peculiar Stars

Pulsar is the evolution product of the massive stars.It is one of the main observation objects of the radio and high-energy telescopes in the word.To date,more than 30000 pulsars have been discovered and studied.However,the radiation mechanism of pulsars is still a open question.For some chemically peculiar(CP)stars,such as the magnetic chemically peculiar(CP2)stars,they show some pulsar-like radio radiation phenomena.The radio radiation mechanism of CP stars is also an unsolved question.The construction and operation of the Five-hundred-meter Aperture Spherical radio Telescope(FAST)and Guo Shoujing telescope(LAMOST)and other major scientific projects provide a historical opportunity to further study the radiation properties of pulsar and CP stars.This thesis will focus on the radiation mechanism of these two kinds of stars.On the one hand,the radiation properties of several pulsars have been studied by using multi-band observations of the radio,X-ray bands;On the other hand,with the machine learning method,the CP stars have been searched from the massive low resolution spectra released by the LAMOST,which provides a large sample data support for the further studies of the radio radiation and rotation etc.The main research contents of this thesis are as follows:In order to study the correlation between the radiation variation and timing noise of pulsar,the long-term changes of pulse profile of Crab pulsar were studied by using the observations of the Rossi X-ray Timing Explorer(RXTE),Hard X-ray Modulation Telescope(HXMT)and Neutron star Interior Composition Explorer(NICER).The data analysis shows that the phase separation?between the first left peak(P1)and the second right peak(P2)is a key parameter that shows the variations of pulse profiles for the Crab pulsar.The evolution of?has a tendency with increasing rates of 0.82±0.25,0.80±0.54,and 0.77 ± 0.28 degrees per century for the 2-6 ke V,6-15 ke V,and 15-60 ke V bands,respectively.To study the physical origin of the pulse changes,the high-SNR X-ray pulse profiles were simulated in the annular gap model,and two model parameters(e.g.,the maximum emission heights of the two peaks)were observed to slightly affect the variations of peak separation.We fitted the long-term variations of emission heights of the two peaks and discovered that the emission height has an increasing tendency with time.Variations of these emission heights induced a characteristic period derivative,and a complete formula for both the magnetic dipole radiation and wind-particle-induced variations of the moment of inertia was used for the spin-down to obtain the variation rate of the magnetic inclination angle,which was-1.6 deg per century.Intrinsic timing noise is observed to be mainly induced by the variations of pulse profiles,which might correlate with a characteristic spin period derivative arising from the fluctuations of the emission regions.This work will lay a foundation for understanding the origin of intrinsic timing noise and making high-precision timing models in the future.The multi-band observations are very helpful for constraining the pulsar radiation models.To study the originate problem of radio radiation of PSR J1848-0123,the newly observed 5 GHz data with Shanghai Tianma 65 meter radio telescope(TMRT)and the archive data provided by the European Pulsar Network(EPN)and CSIRO ATNF Data Archives databases were used to study the broadband evolution property of pulse profiles for this pulsar.It was found that the pulse profiles of PSR J1848-0123 narrows with the increase of emission frequency.The emission components of each pulse profile were separated with the multi-Gaussian fitting method,and the beaming angle and emission radius of each component were calculated with the basic radiation geometry of pulsar.With calculated beam angles and radiation radii,the radius-frequency mapping of this pulsar was simulated with an inverse Compton scatting model(ICS).It was found that the observed radio radiation could be generated in the core gap(CG)region above the polar cap of the pulsar.The beaming angle and emission radius versus frequency for PSR J1848-0123 can be explained well with the ICS model.The radiation properties of single pulse of pulsar was studied.This thesis reported the“Bi-drifting” subpulses observed in PSR J0815+0939 using the FAST.The observation at bandpass 1050-1450 MHz is evenly divided into two bands,i.e.,the bands at center frequencies 1150 MHz and 1350 MHz.The mean pulse profiles and the “Bi-drifting”subpulses at this two bands were investigated.It is found that the pulse profiles at this two frequencies show four emission components,and the peak separations between four emission components decrease with the increase of frequency.In addition,the ratio of peak intensity of each component to the intensity of component IV at 1150 MHz is larger than that at 1350 MHz.An analysis of the longitude resolved fluctuation spectrum and two-dimensional fluctuation spectrum for each emission component was carried out.It is found that the P3 of components I,II and III are about 10.56,10.57,and 10.59 s at both bands.However,the reliable measurements of P3 of component IV and P2 for these four components were not obtained due to the low signal-to-noise ratio of observation data.The pulse energy distributions at frequencies 1150 and 1350 MHz were presented and found that no nulling phenomenon have been found in this pulsar.With the FAST observations,the “Bi-drifting” subpulse phenomenon of PSR J0815+0939 was expanded from 400 MHz to 1350 MHz,which is helpful for the relevant researchers to test and constrain the pulsar emission model,especially the model of “Bi-drifting” subpulse.To solve the problem of the search,spectral classification and atmospheric parameter calculation of CP stars,this thesis report a searching results of CP stars from the low-resolution spectra of early-type stars of the LAMOST DR8.With this big volume of these spectra,we tried machine learning algorithms to search for class-one and class-two Chemical Peculiars(CP1 and CP2)and intended to find spectral features to distinguish the two classes in low-resolution spectra.We compared the classification performance of three machine-learning ensemble algorithms and found that the XGBoost algorithm was the most efficient.Using XGBoost and identified by the visual investigation,we presented a CP1 and CP2 catalogue having 20,694 sources,including 17,986 CP1 and 2,708 CP2,in which 6,917 CP1 and 1,652 CP2 are newly discovered.By XGBoost algorithm,we also obtained the spectral features for classifying CP1&2.The effective temperature(Teff),surface gravity(log g),and metallicity [Fe/H] of the CP1 and CP2 in the catalog were calculated,and the spatial distribution in galactic coordinates and colour-magnitude were investigated as well.The distribution of Teff for CP1 is from 6,000 to 8,500 K,while from 7,000 to 13,700 K for CP2.The distributions for log g of CP1 and CP2 are from 2.8 to4.8 dex peaking at 4.5dex,and from 2.0 to 5.0dex peaking at 3.6dex,respectively.The[Fe/H] of CP1 and CP2 are from-1.4 to 0.4dex,and the [Fe/H] of most CP1 are higher than that of CP2.Almost all targets in our sample locate around the galactic plane.This work provides a large sample support for further detailed research of radiation and rotation of CP stars.