EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3557. Demand response control of wind-based electrolyzed oxygen for advanced activate sludge wastewater treatment
Invited abstract in session WD-24: Sustainable supply chains, stream Circular Economy, Remanufacturing and Recycling .
Wednesday, 14:30-16:00Room: 83 (building: 116)
Authors (first author is the speaker)
| 1. | Qipeng Liu
|
| Chemical and Biochemical Engineering, Technical University of Denmark | |
| 2. | Yi Zheng
|
| College of Smart Energy, Shanghai Jiaotong University | |
| 3. | You Shi
|
| Wind and Energy Systems, Technical University of Denmark | |
| 4. | Krist V.Gernaey
|
| Chemical and Biochemical Engineering, Technical University of Denmark |
Abstract
Green hydrogen has been regarded as a promising substance for sustainable energy conversion and storage. However, the by-product pure oxygen, from water electrolysis, has not yet been widely utilized. Meanwhile, the municipal wastewater treatment system is facing great challenges to upgrade its original capacity and the advanced tertiary treatment. This work proposes to integrate electrolysed oxygen with activated sludge aeration and effluent post-oxidation. An onsite wind/multi-electrolysers’ water electrolysis/oxygen model is simulated to satisfy the potential oxygen demands from the Benchmark Simulation Model No.2 (BSM2). The oxygen productivity fluctuates because of the uncertainty from intermittent wind energy; meanwhile, the oxygen demands of aeration and effluent oxidation are also non-linear dynamic variables. The imbalance of day-ahead supply and demands of oxygen is solved by a virtual oxygen storage unit in the model. The optimization goal is to: maximize the real-time oxygen applications of aeration or effluent oxidation; minimize the volume requirements, and usage counts of oxygen storage units. To simplify the simulation system by making the oxygen production and demands deterministic, the short-term forecasting intelligence of linear regression (LR) algorithm is employed. Then the LR is incorporated with reinforcement learning (RL) for scheduling flexible control of oxygen dispatches, so as to make wastewater treatment processes more efficient and accurate.
Keywords
- Process Systems Engineering
- Water Management
- Sustainable Development
Status: accepted
Back to the list of papers