PortiK: A computer vision based solution for real-time automatic solid waste characterization - Application to an aluminium stream.

In Material Recovery Facilities (MRFs), recyclable municipal solid waste is turned into a precious commodity. However, effective recycling relies on effective waste sorting, which is still a challenge to sustainable development of our society. To help the operations improve and optimise their process, this paper describes PortiK, a solution for automatic waste analysis. Based on image analysis and object recognition, it allows for continuous, real-time, non-intrusive measurements of mass composition of waste streams. The end-to-end solution is detailed with all the steps necessary for the system to operate, from hardware specifications and data collection to supervisory information obtained by deep learning and statistical analysis. The overall system was tested and validated in an operational environment in a material recovery facility. PortiK monitored an aluminium can stream to estimate its purity. Aluminium cans were detected with 91.2% precision and 90.3% recall, respectively, resulting in an underestimation of the number of cans by less than 1%. Regarding contaminants (i.e. other types of waste), precision and recall were 80.2% and 78.4%, respectively, giving an 2.2% underestimation. Based on five sample analyses where pieces of waste were counted and weighed per batch, the detection results were used to estimate purity and its confidence level. The estimation error was calculated to be within ±7% after 5 minutes of monitoring and ±5% after 8 hours. These results have demonstrated the feasibility and the relevance of the proposed solution for online quality control of aluminium can stream.

Removal of MS2, Qß and GA bacteriophages during drinking water treatment at pilot scale.

The removal of MS2, Qß and GA, F-specific RNA bacteriophages, potential surrogates for pathogenic waterborne viruses, was investigated during a conventional drinking water treatment at pilot scale by using river water, artificially and independently spiked with these bacteriophages. The objective of this work is to develop a standard system for assessing the effectiveness of drinking water plants with respect to the removal of MS2, Qß and GA bacteriophages by a conventional pre-treatment process (coagulation-flocculation-settling-sand filtration) followed or not by an ultrafiltration (UF) membrane (complete treatment process). The specific performances of three UF membranes alone were assessed by using (i) pre-treated water and (ii) 0.1 mM sterile phosphate buffer solution (PBS), spiked with bacteriophages. These UF membranes tested in this work were designed for drinking water treatment market and were also selected for research purpose. The hypothesis serving as base for this study was that the interfacial properties for these three bacteriophages, in terms of electrostatic charge and the degree of hydrophobicity, could induce variations in the removal performances achieved by drinking water treatments. The comparison of the results showed a similar behaviour for both MS2 and Qß surrogates whereas it was particularly atypical for the GA surrogate. The infectious character of MS2 and Qß bacteriophages was mostly removed after clarification followed by sand filtration processes (more than a 4.8-log reduction) while genomic copies were removed at more than a 4.0-log after the complete treatment process. On the contrary, GA bacteriophage was only slightly removed by clarification followed by sand filtration, with less than 1.7-log and 1.2-log reduction, respectively. After the complete treatment process achieved, GA bacteriophage was removed with less than 2.2-log and 1.6-log reduction, respectively. The effectiveness of the three UF membranes tested in terms of bacteriophages removal showed significant differences, especially for GA bacteriophage. These results could provide recommendations for drinking water suppliers in terms of selection criteria for membranes. MS2 bacteriophage is widely used as a surrogate for pathogenic waterborne viruses in Europe and the United States. In this study, the choice of MS2 bacteriophage as the best surrogate to be used for assessment of the effectiveness of drinking water treatment in removal of pathogenic waterborne viruses in worst conditions is clearly challenged. It was shown that GA bacteriophage is potentially a better surrogate as a worst case than MS2. Considering GA bacteriophage as the best surrogate in this study, a chlorine disinfection step could guaranteed a complete removal of this model and ensure the safety character of drinking water plants.

Assessment of the removal and inactivation of influenza viruses H5N1 and H1N1 by drinking water treatment.

Since 2003, there has been significant concern about the possibility of an outbreak of avian influenza virus subtype H5N1. Moreover, in the last few months, a pandemic of a novel swine-origin influenza A virus, namely A(H1N1), has already caused hundreds of thousands of human cases of illness and thousands of deaths. As those viruses could possibly contaminate water resources through wild birds excreta or through sewage, the aim of our work was to find out whether the treatment processes in use in the drinking water industry are suitable for eradicating them. The effectiveness of physical treatments (coagulation-flocculation-settling, membrane ultrafiltration and ultraviolet) was assessed on H5N1, and that of disinfectants (monochloramine, chlorine dioxide, chlorine, and ozone) was established for both the H5N1 and H1N1 viruses. Natural water samples were spiked with human H5N1/H1N1 viruses. For the coagulation-settling experiments, raw surface water was treated in jar-test pilots with 3 different coagulating agents (aluminum sulfate, ferric chloride, aluminum polychorosulfate). Membrane performance was quantified using a hollow-fiber ultrafiltration system. Ultraviolet irradiation experiments were conducted with a collimated beam that made it possible to assess the effectiveness of various UV doses (25-60 mJ/cm2). In the case of ozone, 0.5 mg/L and 1 mg/L residual concentrations were tested with a contact time of 10 min. Finally, for chlorine, chlorine dioxide and monochloramine treatments, several residual oxidant target levels were tested (from 0.3 to 3 mg/L) with contact times of 5-120 min. The infectivity of the H5N1 and H1N1 viruses in water samples was quantified in cell culture using a microtiter endpoint titration. The impact of coagulation-settling on the H5N1 subtype was quite low and variable. In contrast, ultrafiltration achieved more than a 3-log reduction (and more than a 4-log removal in most cases), and UV treatment was readily effective on its inactivation (more than a 5-log inactivation with a UV dose of 25 mJ/cm2). Of the chemical disinfection treatments, ozone, chlorine and chlorine dioxide were all very effective in inactivating H5N1 and H1N1, whereas monochloramine treatment required higher doses and longer contact times to achieve significant reductions. Our findings suggest that the water treatment strategies that are currently used for surface water treatment are entirely suitable for removing and/or inactivating influenza A viruses. Appropriate preventive actions can be defined for single disinfection treatment plants.