Two important and large non-point sources of nitrogen in residential areas that adversely affect water quality are stormwater runoff and effluent from on-site treatment systems. These sources are challenging to control due to their variable flow rates and nitrogen concentrations. Denitrifying bioreactors that employ a lignocellulosic wood chip medium contained within a saturated (anoxic) zone are relatively new technology that can be implemented at the local level to manage residential non-point nitrogen sources. In these systems, wood chips serve as a microbial biofilm support and provide a constant source of organic substrate required for denitrification. Denitrifying wood chip bioreactors for stormwater management include biofilters and bioretention systems modified to include an internal water storage zone; for on-site wastewater, they include upflow packed bed reactors, permeable reactive barriers, and submerged wetlands. Laboratory studies have shown that these bioreactors can achieve nitrate removal efficiencies as high as 80–100% but could provide more fundamental insight into system design and performance. For example, the type and size of the wood chips, hydraulic loading rate, and dormant period between water applications affects the hydrolysis rate of the lignocellulosic substrate, which in turn affects the amount and bioavailability of dissolved organic carbon for denitrification. Additional field studies can provide a better understanding of the effect of varying environmental conditions such as ambient temperature, precipitation rates, household water use rates, and idle periods on nitrogen removal performance. Long-term studies are also essential for understanding operations and maintenance requirements and validating mathematical models that integrate the complex physical, chemical, and biological processes occurring in these systems. Better modeling tools could assist in optimizing denitrifying wood chip bioreactors to meet nutrient reduction goals in urban and suburban watersheds.

No abstract

This paper aims at integrating three powerful techniques namely Deep Learning, Approximate Computing, and Low Power Design into a strategy to optimize logic at the synthesis level. We utilize advances in deep learning to guide an approximate logic synthesis engine to minimize the dynamic power consumption of a given digital CMOS circuit, subject to a predetermined error rate at the primary outputs. Our framework, Deep-PowerX 1 , focuses on replacing or removing gates on a technology-mapped network and uses a Deep Neural Network (DNN) to predict error rates at primary outputs of the circuit when a specific part of the netlist is approximated. The primary goal of Deep-PowerX is to reduce the dynamic power whereas area reduction serves as a secondary objective. Using the said DNN, Deep-PowerX is able to reduce the exponential time complexity of standard approximate logic synthesis to linear time. Experiments are done on numerous open source benchmark circuits. Results show significant reduction in power and area by up to 1.47× and 1.43× compared to exact solutions and by up to 22% and 27% compared to state-of-the-art approximate logic synthesis tools while having orders of magnitudes lower run-time.