Abstract. Flash floods have occurred frequently in the urban areas of southern China. An effective process-oriented urban flood inundation model is urgently needed for urban storm-water and emergency management. This study develops an efficient and flexible cellular automaton (CA) model to simulate storm-water runoff and the flood inundation process during extreme storm events. The process of infiltration, inlets discharge and flow dynamics can be simulated with little preprocessing on commonly available basic urban geographic data. In this model, a set of gravitational diverging rules are implemented to govern the water flow in a rectangular template of three cells by three cells of a raster layer. The model is calibrated by one storm event and validated by another in a small urban catchment in Guangzhou of southern China. The depth of accumulated water at the catchment outlet is interpreted from street-monitoring closed-circuit television (CCTV) videos and verified by on-site survey. A good level of agreement between the simulated process and the reality is reached for both storm events. The model reproduces the changing extent and depth of flooded areas at the catchment outlet with an accuracy of 4 cm in water depth. Comparisons with a physically based 2-D model (FloodMap) show that the model is capable of effectively simulating flow dynamics. The high computational efficiency of the CA model can meet the needs of city emergency management.

No abstract

Low grade heat still widely exists in energy-intensive industrial parks, although good energy integration has been accomplished for individual processes or plants. Low grade heat is notably large but difficult to utilize because of the limitation of heat transfer and the scarcity of low grade heat sinks. Large scale utilization of low grade heat is very challenging for energy-intensive industries or industrial parks. A large scale low grade heat recovery, refrigeration, and utilization network system is introduced in this study to improve energy performance for industrial parks. In order to model the large scale system, the system is decomposed into three levels: pipe networks, refrigeration stations and absorption chillers. A mixed integer nonlinear programming model is presented that considers mass and energy networks, pipes, refrigeration stations, absorption chillers, and economic performance. The mathematical model is applied to the optimization and economic analysis for the low grade heat utilization in a petrochemical industrial park in China. The model can be solved in available time using the global solver. The solution results demonstrate the good economic performance of the new low grade heat recovery, refrigeration, and utilization network system for the industrial park.

No abstract

Though toluene disproportionation is an important process for producing para-xylene, it is heavily energy intensive because of its high reaction temperature and the need to separate close boiling-point components. Pinch analysis is often used to target utility requirements for process systems. Nevertheless, the supply and final temperatures of process streams are all predetermined according to the sequential method indicated in the onion model. Therefore, the sequential method ignores the influences of outer level facilities on inner level facilities, which leads to suboptimal solutions. To tackle this problem, variable temperatures of process streams are taken into account in this study to simultaneously target the utility requirements of columns and heat exchanger networks in a toluene disproportionation plant. To this end, relevant equations representing the relationships between feed temperatures and heat duties of columns are first obtained based on simulation data. Second, the equations are integrated into a transshipment model. Meanwhile, variable temperatures are introduced into temperature intervals. As a result, a mixed integer nonlinear programming problem is formulated to minimize the utility requirement in the whole toluene disproportionation plant. Third, the solution results are discussed, providing insights into the optimal results and the sensitivity of utility requirement caused by process streams and separation columns.