Abstract: A de-noising method of hydropower unit throw signals based on ensemble empirical mode decomposition (EEMD) and approximate entropy is presented. This method decomposes the throw signals into several intrinsic mode functions (IMFs) and calculates the approximate entropy for each function. Those functions of approximate entropy less than a certain preset threshold, then are reconstructed so as to eliminate the abnormal noises of the signals. It also compares the two methods of wavelet transform and EEMD in terms of their calculations of root mean square error, correlation coefficient, and signal-to-noise ratio. The results show that the EEMD method is superior in its adaptability and de-noising performance and very suitable for online de-noising of hydropower unit throw signals.