Abstract
To alleviate the greenhouse gas emissions by the chemical industry,
electrification has been proposed as a solution where electricity from
renewable sources is used to power processes. The adoption of renewable
energy is complicated by its spatial and temporal variations. To address
this challenge, we investigate the potential of distributed
manufacturing for electrified chemical processes installed in a
microgrid. We propose a multi-scale mixed-integer linear programming
model for locating modular electrified plants, renewable-based
generating units, and power lines in a microgrid that includes monthly
transportation and hourly scheduling decisions. We propose a K-means
clustering-based aggregation disaggregation matheuristic to solve the
model efficiently. The model and algorithm are tested using a case study
with 20 candidate locations in Western Texas. Additionally, we define a
new concept, “the Value of the Multi-scale Model”, to demonstrate the
additional economic benefits of our model compared with a single-scale
model.