Modeling and characteristic analysis of a marine condenser based on improved PSO-RBF algorithm

With the increase of usage time, the working state of equipment will be changed, thus the simple mechanism model is difficult to meet the requirement of precision. In accordance with this special characteristic, a 2 stage modeling of a certain type marine condenser based on mechanism and historical data is presented in this paper. Firstly, the operating mechanism of the condenser is analyzed, and then the mechanism model is established. Secondly, the model parameters are optimized by using the historical data and the proposed improved PSO-RBF algorithm, thus the RBF neural network model, which can reflect the input and output characteristics of the condenser, is obtained. Through the experiment, it is proved that the model is consistent with real dynamic response of the actual equipment. Through integrating the established condenser model with the marine power performance simulation model library, the dynamic response of each parameter of the condenser when cooling water mass flow rate and steam turbine speed reduced 20% are studied. The analysis result can provide reference for the optimization design and control strategy of the marine condenser.