The Synthesis Optimization Of Heat Exchanger Network And Its Control Strategy

The limited energy resources and low energy utilization efficiency will restrict economic development in China. To save energy and lower energy consumption is becoming a long term tendency. Process integration and optimization is among the main means of energy saving in industrial fields, and the optimization of heat exchanger network (HEN) design has been one of the important parts of process integration and optimization.There are some performance requirements in industrial HEN system. For example, the outlet temperature must meet the process requirements, reducing the influence of temperature fluctuation to the process; the heat exchange system can operate stably; improving the heat exchanging efficiency, reducing the energy consumption, and so on. All these requirements need real-time automatic control of the HEN. The control system design of a well behaved HEN must include two parts: the design of control structure, and the design of control algorithm. The former determines the sensibility of controlled variables to manipulated variables, the dynamic characteristics of the regulation channels, and the system operation cost, and restricts the latter by these aspects.Based on the above considerations, the paper focuses on the following research:At first, the structure of HEN is obtained by applying the pinch technology based on the stream properties, and the topology structure of HEN is designed aiming at least annual cost;Secondly, the control structure design method of HEN is proposed based on the theory of "downstream paths" and empirical rules. The operation cost of candidate control structure schemes is determined by a bypass optimal mathematical model. Therefore, control engineers can choose a control structure which is a trade-off between control performance and operation cost. Since some performances of control system can not be generally estimated by an index, and simultaneous optimization requires a more complicated mathematical model, the control structure design method applies a two steps optimization, considering the control performance and operation cost respectively.Finally, the multivariable dynamic matrix control algorithm is derived which contains the steady information. Then, it is illustrated by a HEN which has a given structure, parameters and control structure. Simulation results demonstrate better disturbance and noise rejection,servo response performance of this control strategy.