True Value Estimation of Centrifugal Fan Vibration Data Based on Fusion Method

Poor information means incomplete and insufficient information, such as unknown probability distributions and trends. In this case, many statistical methods may become ineffective. So a fusion method based on the information poor theory is proposed. The fusion method contains five methods, three concepts, and one rule. The five methods include the rolling mean method, the membership function method, the maximum membership grade method, the moving bootstrap method, and the arithmetic mean method. The three concepts comprise the solution set on the estimated true value, the fusion series, and the terminal estimated true value. The rule is the range rule. The centrifugal fan vibration data is evaluated using the above methods and theory. The experimental investigation shows that the fusion method is effective in evaluating true value with small sample sizes without any prior information of the probability distribution.