The detection of gravitational waves (GWs) by pulsar timing arrays (PTAs) is not only a very important supplementation of the verification of general relativity but also a new window to study the evolution of supermassive binary black holes and the early universe. However, so far the detection sensitivity of PTAs is not good enough to catch signals of GWs due to the disturbance of various noises. In this paper we explore the influences of the correlated noises caused by the reference clock errors and solar system ephemeris errors in pulsar timing on the detection of stochastic gravitational waves background (GWB). We demonstrate the power-law integrated sensitivity curves of GWB detection under the impacts from these correlated noises. From the simulated data, we find that the influence of different reference time-scale is non-negligible, and the influence is even quite huge if one uses a very old version of solar system ephemeris. However, the impact from these correlated noises on the sensitivity curve is very limited for the real observational data released by international pulsar timing arrays (IPTA). We also calculate the signal-to-noise ratios based on the theoretical GW amplitude permitted by observations. Moreover, we study how the detection sensitivity increases with more pulsar number and longer observation.

脉冲星定时阵列（PTA）探测引力波（GWs）不仅是验证广义相对论的重要补充，也是研究超大质量双黑洞和早期宇宙演化的新窗口。然而，由于各种噪声的干扰，到目前为止，PTA的检测灵敏度还不足以捕捉GW的信号。在本文中，我们探讨了脉冲星计时中的参考时钟误差和太阳系星历误差引起的相关噪声对随机引力波背景（GWB）检测的影响。我们展示了在这些相关噪声的影响下 GWB 检测的幂律集成灵敏度曲线。从模拟数据中，我们发现不同参考时间尺度的影响是不可忽略的，如果使用非常古老的太阳系星历表，其影响甚至是相当大的。然而，对于国际脉冲星计时阵列（IPTA）发布的真实观测数据，这些相关噪声对灵敏度曲线的影响非常有限。我们还根据观测允许的理论 GW 幅度计算信噪比。此外，我们研究了探测灵敏度如何随着脉冲星数量的增加和观察时间的延长而增加。我们还根据观测允许的理论 GW 幅度计算信噪比。此外，我们研究了探测灵敏度如何随着脉冲星数量的增加和观察时间的延长而增加。我们还根据观测允许的理论 GW 幅度计算信噪比。此外，我们研究了探测灵敏度如何随着脉冲星数量的增加和观察时间的延长而增加。