Deep Learning in the Ubiquitous Human-Computer Interactive 6G Era: Applications, Principles and Prospects.

With the rapid development of enabling technologies like VR and AR, we human beings are on the threshold of the ubiquitous human-centric intelligence era. 6G is believed to be an indispensable cornerstone for efficient interaction between humans and computers in this promising vision. 6G is supposed to boost many human-centric applications due to its unprecedented performance improvements compared to 5G and before. However, challenges are still to be addressed, including but not limited to the following six aspects: Terahertz and millimeter-wave communication, low latency and high reliability, energy efficiency, security, efficient edge computing and heterogeneity of services. It is a daunting job to fit traditional analytical methods into these problems due to the complex architecture and highly dynamic features of ubiquitous interactive 6G systems. Fortunately, deep learning can circumvent the interpretability issue and train tremendous neural network parameters, which build mapping relationships from neural network input (status and specific requirements of a 6G application) to neural network output (settings to satisfy the requirements). Deep learning methods can be an efficient alternative to traditional analytical methods or even conquer unresolvable predicaments of analytical methods. We review representative deep learning solutions to the aforementioned six aspects separately and focus on the principles of fitting a deep learning method into specific 6G issues. Based on this review, our main contributions are highlighted as follows. (i) We investigate the representative works in a systematic view and find out some important issues like the vital role of deep reinforcement learning in the 6G context. (ii) We point out solutions to the lack of training data in 6G communication context. (iii) We reveal the relationship between traditional analytical methods and deep learning, in terms of 6G applications. (iv) We identify some frequently used efficient techniques in deep-learning-based 6G solutions. Finally, we point out open problems and future directions.

Co-digestion of primary sewage sludge with drinking water treatment sludge: A comprehensive evaluation of benefits.

Anaerobic co-digestion of primary sludge with two types of drinking water treatment sludge (DWTS), namely iron- or aluminum-rich DWTS (Fe- or Al-DWTS) were systematically evaluated by biochemical methane potential tests, kinetic modelling, downstream process parameters and microbial community analysis. Specific methane yields decreased approximately 19% to 123 mL·g-1 VS, while the hydrolysis constant kh decreased from 0.21 d-1 to 0.18 d-1 for Fe-DWTS at 10% to 40% dosages. On the contrary, specific methane yields decreased 45-55% for Al-DWTS, and kh decreased to 0.14 d-1 at 40% dosage. Significant removals (>95%) of phosphate and hydrogen sulfide were observed for Fe- and Al-DWTS additions at 40% dosage. Microbial community analysis revealed that Al-DWTS increased the abundance of most hydrogenotrophic methanogens, while Fe-DWTS increased the abundance of acetoclastic methanogens. Kinetic modelling further revealed that Fe- and Al-DWTS additions affected the hydrolysis and methanogenesis process kinetics and the methane yield differently.

Enhanced anaerobic digestion of primary sludge with additives: Performance and mechanisms.

Anaerobic digestion of primary sludge with different additives, namely nano magnetite, graphite powder, activated carbon powder and NiCl2/CoCl2, were evaluated by biomethane potential tests, kinetics modelling and microbial community analysis. Specific methane yields increased from 136 mL/g VS for primary sludge to 146 mL/g VS, 151 mL/g VS, and 152 mL/g VS for the addition of nano magnetite, graphite powder, and activated carbon powder at optimal dosages, respectively. The first order hydrolysis constant kh increased from 0.488 d-1 to 0.526 d-1, 0.622 d-1, and 0.724 d-1, respectively. Microbial community analysis revealed that the abundance of key bacterial and archaeal populations was positively correlated with hydrolysis and methane production. The enhanced methane production with activated carbon powder was due to shifting methane formation pathway from acetoclastic to hydrogenotrophic methanogenesis. In contrast, nano magnetite and graphite powder additives enhanced the direct interspecies electron transfer evidenced by increased abundance of Methanosaeta and Methanolinea.

Assessing the removal of organic micropollutants from wastewater by discharging drinking water sludge to sewers.

Discharging drinking water treatment sludge (DWTS) to sewers could be an efficient waste management strategy with the potential to replace chemical dosing for pollutant control. This study for the first time investigated the fate of 28 different organic micropollutants (MPs) due to the dosing of iron-rich and aluminum-rich DWTS in a pilot rising main sewer. Nine MPs had an initial rapid removal within 1-hr (i.e., 10-80%) due to Fe-DWTS dosing. The formation of FeS particles due to Fe-DWTS dosing was responsible for the removal of dissolved sulfides (80% reduction comparing to control sewer). Further particle characterization using SEM-EDS, XRD and ATR-FTIR confirmed that FeS particles formation played an important role in the removal of MPs from wastewater. Adsorption of MPs onto the FeS particles was likely the possible mechanism for their rapid removal. In comparison to iron-rich DWTS, aluminum-rich DWTS had very limited beneficial effects in removing MPs from wastewater. The degradability of degradable MPs, including caffeine, paraxanthine, paracetamol, metformin, cyclamate, cephalexin, and MIAA were not affected by the DWTS dosing. Some non-degradable MPs, including cotinine, hydroxycotinine, tramadol, gabapentin, desvenlafaxine, hydrochlorothiazide, carbamazepine, fluconazole, sulfamethoxazole, acesulfame, saccharin and sucralose were also not impacted by the DWTS dosing. This study systematically assessed the additional benefits of discharging Fe-DWTS to the sewer network i.e., the removal of MPs from the liquid phase thereby reducing its load to the treatment plant. The results corroborate the discharge of Fe-rich DWTS in sewers as an effective and beneficial way of managing the waste by-product.

Evaluation of data-driven models for predicting the service life of concrete sewer pipes subjected to corrosion.

Concrete corrosion is one of the most significant failure mechanisms of sewer pipes, and can reduce the sewer service life significantly. To facilitate the management and maintenance of sewers, it is essential to obtain reliable prediction of the expected service life of sewers, especially if that is based on limited environmental conditions. Recently, a long-term study was performed to identify the controlling factors of concrete sewer corrosion using well-controlled laboratory-scale corrosion chambers to vary levels of H2S concentration, relative humidity, temperature and in-sewer location. Using the results of the long-term study, three different data-driven models, i.e. multiple linear regression (MLR), artificial neural network (ANN), and adaptive neuro fuzzy inference system (ANFIS), as well as the interaction between environmental parameters, were assessed for predicting the corrosion initiation time (ti) and corrosion rate (r). This was performed using the sewer environmental factors as the input under 12 different scenarios after allowing for an initiation corrosion period. ANN and ANFIS models showed better performance than MLR models, with or without considering the interactions between environmental factors. With the limited input data available, it was observed that ti prediction by these models is quite sensitive, however, they are more robust for predicting r as long as the H2S concentration is available. Using the H2S concentration as a single input, all three data driven models can reasonably predict the sewer service life.

[Sludge performances and membrane pollution in aerobic granular sludge membrane bioreactor].

The aerobic granular sludge membrane bioreactor worked for 75 days. In this operation, the removal efficiencies for COD, NH4+ -N and TN were 91%-95%, 89.57%-100% and 48.25%-90.84%, respectively. The granular sludge appeared disintegration to some extent and the sludge settlement performances deteriorated. Sludge specific resistance from initial resistance of 1.085 x 10(13) m/kg rose to 2.712 x 10(13) m/kg at the end; extracellular polymeric substances maintained between 44.71 and 72.23 mg/g. Then, pollution characteristics of the membrane module were studied. It was found that membrane surface cake layer resistance and membrane pore plugging resistance were 5.32 x 10(11) m(-1) and 2.34 x 10(11) m(-1), accounted for the total resistance of 67.42% and 29.66%, respectively. Through the analysis of infrared spectrum for membrane fouling matters found that the main fouling matters were proteins and carbohydrate materials for membrane.

[Flow field test on the tangential section of polypropylene tubular membrane module annular gap in rotating linear tangential flow].

A new type of polypropylene tubular membrane apparatus of rotating cross flow was designed to study experimentally the flow field characteristics of the tangential section of the membrane annular gap. The authors designed rotary linear tangential flow tubular membrane separator and its test system for the first time. Through the system, the flow field of rotary linear tangential flow with the advanced Particle Image Velocimetry (PIV) was tested for the first time. A lot of streamlines and vorticity maps of the tangential section of separator in different operation conditions were obtained. The velocity distribution characteristics were analyzed quantitatively: 1. At non-vortex area, no matter how the operation parameters change, the velocity near to rotary tangential flow entrance was higher than the velocity far from entrance at the same radial coordinates. At vortex area, generally the flow velocity of inner vortex was lower than the outer vortex. At the vortex center, the velocity was lowest, the tangential velocity were equal to zero generally. At the vortex center zone, the tangential velocity was less than the axial velocity. 2. Under test operations, the tangential velocity and axial velocity of vortices borders are 1-2 times of average axial velocity of membrane module annular gap. The maximum tangential velocity and axial velocity of ellipse vortices were 2-6 times of average axial velocity of membrane module annular gap. 3. The vortices that are formed on the tangential section, there existed mass transfer between inner and outer parts of fluid. Much fluid of outer vortices got into the inner ones, which was able to prevent membrane tube from particles blocking up very soon.

[Analysis on energy consuming of polypropylene tubular membrane microfiltration in rotating tangential flow].

A new type of polypropylene tubular membrane module in rotating tangential flow was designed to experimentally study energy consuming of the polypropylene PA tubular membrane microfiltration. Through the experimental system of rotating tangential flow which was set up to enhance the microfiltration, the polypropylene PA tubular membrane fluxes and energy efficiency of rotating tangential flow and axial flow were measured under different fluid pressures. The experimental results demonstrate that rotary tangential flow tubular membrane microfiltration was a high efficiency technology of membrane separation. At test conditions, the stable energy efficiency of linear and circular tangential flow were higher than 11.185 times and 14.591 times of the stable energy applying efficiency of axial flow respectively.