| Among the process of industry heat-exchange , fouling causes great cost and low efficiency. Because its occurring conditions, factors and mechanism are variable and complex, it is very difficult to develop a precise mathematical model to make the overall description. Accordingly, it is necessary to apply artificial intelligence to the on-line fouling monitoring technique. The automatic on-line monitor of fouling thermal resistance is developed and much fouling datum is obtained . So it is important to analyze these fouling datum and research nonlinear character of fouling. The main content and innovation of this thesis is as following:The change rate of fouling thermal resistance is employed as a single to reflect primary characteristic of fouling process. Utilizing PCA and power spectrum and Lyapunov exponent means, experimental data is analyzed and fouling process taking on chaos under some condition is found in this paper Then, one or more kinds of parameters, which describes complicated coupling process about mass-exchange and quantity of heat-exchange and flow in fouling process, may lead to chaos phenomenon. Water bath temperature is key parameter, whose stochastic fluctuate may lead to the time series of fouling thermal resistance with chaos character. The consumption characteristic of fouling is researched and the consumption function of fouling is attained. The time series of fouling thermal resistance is analyzed and compared by utilizing multivariate nonlinear regression,Logistic,S model . Accordingly, a new fouling process model is attained, which fitted many enhancement of heat transfer surface. The monitor parameters influencing extent on fouling thermal resistance is gained by deflection correlation coefficient method calculating metrical fouling thermal resistance datum .The result is :exit temperature,entry temperature,downriver pipeline surface temperature,water bath temperature,velocity of flow,upriver pipeline surface temperature. |
PDF Full Text Request