The wheel polygon is one of the common damage forms of the railway wheel, which could deteriorate the dynamic response of the vehicle/track system and decrease their stability and service life. The initiation and development of wheel polygon is the gradual accumulation of periodic uneven wear. In general, the greater the hardness of the wheel, the stronger its wear resistance, which may affect the wheel polygon. Laser quenching is one of the methods to improve the hardness of wheel materials. Although many scholars have carried out a large number of studies on the laser surface quenching, those mainly focus on the effect of laser surface quenching on wear and damage of wheel and rail materials. Additionally, few studies on the influence of laser quenching on wheel polygon were conducted. Therefore, it is necessary to investigate the influence of laser surface quenching on wheel polygon initiation and development. In this study, MMS-2A tester was used to carry out rolling and sliding wear experiments. During the test, the wheel polygon was studied from the aspects of the macro morphology of the wear surface, the overall contour of the wheel, the polygon order, the vibration frequency of the wheel rail system, the change of the amplitude of the polygon, the microstructure after the test, and the rolling contact fatigue to explore the influence of laser surface quenching on the wheel polygon. The results showed that after laser quenching, a certain depth of martensite was produced in the surface hardened area of the wheel, and the surface hardness was increased by about 2.6 times. The original hardness was maintained in the non-quenched area, and the matrix structure was pearlite. As for polygon evolution during the rolling test, when 12 and 16 points were quenched uniformly along the circumference direction on the wheel surface, due to the initial uneven hardness, 12 and 16 orders polygon generated respectively on the wheel surface, and serious wheel rail system vibration was excited, and finally a wheel polygon with large amplitude was generated at 150 000 cycles. When the wheel surface was fully quenched, the hardness of the wheel surface was greatly improved and evenly distributed, and its ability to resist plastic deformation was enhanced, thus resisting the occurrence of uneven wear, thereby inhibiting the initiation of wheel polygons. In order to study the influence of laser surface quenching on the development of the wheel polygon when it had been generated, the surface quenching was carried out on the trough of the wheel polygon after 60 000 cycles. The polygon development showed a phenomenon of slowing down first and then accelerating, the polygon amplitude became smaller first and then larger, and the trough turned to be the crest. When the whole surface of the wheel polygon was quenched after 60 000 cycles, due to the polygon had been generated at this time, and the amplitude of the polygon would not be affected significantly after quenching, so the characteristic frequency of the polygon was still excited, and the polygon would continue to develop. And the amplitude of the polygon slightly decreased after 150 000 cycles. After the rolling test, the plastic deformation in the non quenched position of the wheel surface was obvious, and there was no plastic deformation in the quenched position. In terms of fatigue damage, compared with the wheel without quenching treatment, whether it was point quenching or full surface quenching, the position of the crest or the trough, the fatigue cracks in the quenching position extended to the inside of the material in a way of large angle and large depth, while the fatigue cracks in the non quenching area showed the same length, smaller angle and smaller depth. Based on the above results, it could be found that the wheel polygons could be effectively restrained after the full surface hardening of the initial wheel (non-polygon wheel), but the development of the polygon could not be restrained after the full surface hardening of the polygonal wheel. The test results could provide practical reference for "selecting a suitable wheel state (or running time node) for laser full-surface quenching, and then inhibiting the initiation and development of wheel polygons".