372406 Optimal Integration of Industrial Wastewater Discharges through Dynamic Models
This work presents a dynamic model to determine the optimal location of a new industrial plant considering simultaneously the integration of the water network system inside of the new industrial facilities and its effect on the surrounding watershed to satisfy environmental regulations and constraints over the water quality. The model was formulated to predict the behavior for the watershed system impacted for the new polluted discharges. In this sense, the material flow analysis (MFA) technique is useful to properly track the pollutants throughout the watershed system. This model considers the variation of the existing discharges, as well as the fluctuation respect to the time of the extractions carried out over the watersheds and their surroundings. In this regard, the model accounts for all the types of inlet streams in the watershed (i.e., pluvial, residential, sanitary and industrial discharges, etc.); in addition to the natural extractions (i.e., evaporation, filtration) or not natural (i.e., irrigation). The mathematical formulation includes a disjunctive model for the optimal selection of new sites to install new facilities; each possible alternative is available to discharge new wastewater streams to the section nearest of the river satisfying the environmental regulations and all the environmental restrictions imposed over the water quality along the watershed. Thus, the new wastewater discharge is integrated to the existing processes and its compositions are tracked throughout the watershed. In this way, it can be determined the required treatment for the new wastewater stream, which operates dynamically considering the temporal variations, as well as the chemical and biochemical reactions carried out in the water bodies for the pollutants degradation.
Finally, the objective function consists in minimizing the total annual cost that is constituted by the cost of installation of the new plant (including the transportation for raw materials, products and services, as well as the land cost) and the wastewater treatment costs. The proposed model is a Mixed Integer Non Linear Programming problem and it was implemented in the software GAMS, where the solver DICOPT was used to solve a case study from Mexico. Results show that there is very important to consider the dynamic behavior for the system to track the pollutants through the watershed; this way, the model allows to yield sustainable solutions during the installation of a new industrial plant.