Industrial park heat integration considering centralized and distributed waste heat recovery cycle systems

The waste heat recovery cycle, such as Organic Rankine cycle and absorption refrigeration cycle, can convert waste heat to other forms of energy (heat, cooling and electricity) efficiently. Meanwhile, besides the traditional intra-plant energy utilization, the cluster of enterprises in an industrial park brings extra opportunities and benefits for the total site energy recovery and conversion. Therefore, in order to improve the energy utilization efficiency and reduce the overall cost, an optimization-based framework that enables the simultaneous integration of heat exchanger network, utility and waste heat recovery cycle systems of industrial parks is proposed in this study. Three heat integration technologies are simultaneously considered to achieve the design purpose, including the direct heat integration inside plants, the indirect heat integration across different plants, and the integration of heat exchanger network with waste heat recovery cycle systems. Two waste heat recovery cycle systems configuration modes, centralized and distributed, are proposed to achieve the flexible utilization of waste heat in industrial parks. A mathematical model in mixed integer nonlinear programming is formulated to optimize the design scheme subjecting to the objective of minimum total annualized cost. The effectiveness of the proposed method is demonstrated by two cases studies, and the applicable scenarios for the two waste heat recovery cycle modes are revealed by comparing and analyzing the economic performance of the total site systems. It is concluded that the two waste heat recovery cycle configuration modes do have specific conditions of applicability, that is, the distributed mode is the better choice if the utility cost plus transportation cost of the entire system dominates the total cost, conversely, the centralized mode is recommended. The results also indicate that the inter-plant distance is an important factor affecting decision making, so sensitivity analysis is subsequently conducted and indicates that the distributed mode is preferable only if a specific threshold of interplant distance is exceeded.