Due to the characteristics of large field of view, small size and low cost, infrared refractive lenses have been widely used in aerospace remote sensing, military reconnaissance, biological detection and other fields. With the development of infrared imaging technology, higher requirements are put forward for the technical specifications of imaging systems. Therefore, improving the imaging performance of lenses becomes the key in the alignment process. At present, the high-precision alignment of infrared refractive lenses at home and abroad mainly adopts the method of centering error measurement, which makes the centering error of each optical element within the tolerance range by adjusting its position. However, due to the preparation process characteristics of infrared optical materials, the refractive index homogeneity of most materials is difficult to guarantee (Fig.2), which must introduce additional irregular aberrations into the optical system, and lead to the degradation of lens imaging performance. The core of precise centering error measurement and alignment is the control of optical axis consistency and spacing of optical elements, which can't do anything to correct the irregular aberrations of the system. For this reason, a system wavefront compensation method combining the iterative adjustment position and the surface modification for the optical elements was presented to realize infrared lenses alignment based on high performance.Based on the precise centering error measurement of lens, an online device with lenses alignment and image quality measurement was designed (Fig.3), which can switch the measurement and alignment conditions through the planar mirror switching device to cut in and out. According to the measured wavefront results of the lens and the computer aided alignment technology, the positions of the optical elements were iteratively adjusted by the clamping devices and jackscrews to correct the first-order aberrations (Fig.4). The residual medium and high-order aberrations of the system were compensated by repairing the optical element surface at the pupil and introducing the anti-residual wave aberrations, on the basis of comprehensive analysis and calculation of system wavefronts measured in several fields of view (Fig.6).A medium wavelength infrared refractive lens was aligned by means of systematic wavefront compensation. Due to the refractive index inhomogeneity of materials, the RMS (λ=3.39 μm) of the system wavefronts for the lens's three fields of view were 0.162λ, 0.118λand 0.166λ respectively after precision centering (Tab.2). Based on the sensitivity matrix, the first-order aberrations were corrected by iteratively adjusting the positions of the optical elements, which decreased the RMS (λ=3.39 μm) of the three fields of view to 0.090λ, 0.083λ and 0.098λ (Tab.3). Then, the residual medium and high-order aberrations were compensated by repairing the surface of the optical windows (Tab.4), and the RMS (λ=3.39 μm) of the three fields of view were reduced to 0.064λ, 0.040λ and 0.067λ at last (Tab.5). The results show that this alignment technology can effectively compensate the system wavefronts of the infrared lenses and greatly improve the image performance, which has important engineering application value. Due to the high deviation of the refractive index homogeneity of infrared optical materials, it was very difficult to align infrared refractive lenses with high performance. In this study, the alignment technology for infrared refractive lenses based on high performance is proposed. Through the introduction of online device with lenses alignment and image quality measurement, the lens alignment and measurement were integrated, and the positions of the optical elements were iteratively adjusted to correct the first-order aberrations based on the system sensitivity matrix. At the same time, the residual medium and high-order aberrations of the system were compensated by repairing the optical element surface at the pupil and introducing the anti-residual wave aberrations. Through an example of lens alignment, it was verified that this technology can significantly improve the imaging performance of infrared refractive lenses, break through the limitations of traditional methods of lens alignment, and provide a feasible way for developing high performance infrared lens.