The invention provides a predicted control adaptive filtering algorithm based on posterior probability distribution, and belongs to the automation control field. The predicted control adaptive filtering algorithm comprises steps that a process measured value sequence is constructed, and the standard deviation of the process measured value sequence is acquired; a probability density function expression is determined according to the noise type of the process measured value, and the current measured value and the measured value at a previous moment with respect to the current measured value of the process measured value sequence are selected, and by combining with the standard deviation, a filtering coefficient is determined; filtering calculation is carried out according to the filtering coefficient. By introducing a Bayes probability calculation way, the measured value at the previous moment is used as a distribution expected mean value, and the prior variance of the noise is used as the distribution variance, and posterior probability of a filtering parameter estimated value is calculated, and then the filtering coefficient is determined, and the filtering processing is carried out. By comparing with the prior art, adaptive adjusting of an actual filtering process is carried out, and artificial parameter setting is not required, and therefore under a complicated noisy condition, a strong adaptive performance is provided, and a strong filtering effect is displayed.