Fouling is a principal constraint of membrane bioreactors (MBRs). It blocks the wide use of MBRs and aggravates the ability of MBRs. Trans-membrane pressure (TMP) is measured simply from MBRs and is a useful factor for evaluating fouling phenomena such as fouling mechanisms. Fouling mechanism diagnosis based on a measured TMP was used to evaluate MBRs operation conditions. However, diagnosis of MBR conditions is difficult due to the dynamic conditions of MBRs. Therefore, we used differential calculus, exponential weighted moving average (EWMA) and fast Fourier transform (FFT) to determine a periodic pattern for diagnosing fouling mechanisms in the dynamic operating conditions of MBRs. The periodic pattern was reflected in the operating conditions of MBRs, based on the fouling mechanism using TMP. We used two data sets obtained from pilot-scale MBR to suggest a periodic pattern and validated the proposed method using a lab-scale MBR experiment. Consequently, the suggested periodic pattern can diagnose fouling mechanisms using the proposed method, because the methods can be adjusted under the dynamic conditions of MBRs.