Lightweight Super-Resolution with Self-Calibrated Convolution for Panoramic Videos.

Panoramic videos are shot by an omnidirectional camera or a collection of cameras, and can display a view in every direction. They can provide viewers with an immersive feeling. The study of super-resolution of panoramic videos has attracted much attention, and many methods have been proposed, especially deep learning-based methods. However, due to complex architectures of all the methods, they always result in a large number of hyperparameters. To address this issue, we propose the first lightweight super-resolution method with self-calibrated convolution for panoramic videos. A new deformable convolution module is designed first, with self-calibration convolution, which can learn more accurate offset and enhance feature alignment. Moreover, we present a new residual dense block for feature reconstruction, which can significantly reduce the parameters while maintaining performance. The performance of the proposed method is compared to those of the state-of-the-art methods, and is verified on the MiG panoramic video dataset.

Generalized prediction and optimal operating parameters of PCDD/F emissions by explainable Bayesian support vector regression.

The current derived models for predicting polychlorinated dibenzo-p-dioxins and -furans (PCDD/F) emissions from incineration can only be applied to a specific incinerator due to high deviation or systematic errors. And the models fail to provide quantized guidance for the operation of full-scale municipal solid waste incinerators. To address the problem, explainable Bayesian support vector regression (E-BSVR) has been established to generalized predict and maximumly reduce the PCDD/F emissions. First, forty-two PCDD/F samples were determined from a whole year experiment in a full-scale incinerator. Meanwhile, 1,2,4-trichlorobenzene(1,2,4-TrCBz), carbon monoxide, sulfur dioxide, oxynitride, particulate matter, fluoride, and hydrogen chloride were measured, as input features. Second, after box-cox transformation normalization, and hyperparameters tuning, the R-Squared and root mean square error (RMSE) of the proposed method are 0.983 and 0.044, exhibiting high accuracy. The high accuracy (R-Squared = 0.992) and generalization are also proven on the dataset with high PCDD/F emissions. Then, the performances of BSVR are compared with kernel ridge regression, multiple linear regression, and unary linear regression, indicating afar smaller RMSE of BSVR. Finally, the optimal operating parameters are calculated through local interpretable model-agnostic explanations and the partial dependence plot. Results indicate that reducing the content of organic chlorine in municipal solid waste and inhibiting the deacon reaction are important methods for reducing PCDD/F emissions. The optimal operating parameters for the maximal reduction of PCDD/F emissions are 1,2,4-TrCBz < 0.098 ug/m3, fluoride > 0.452 mg/m3. As a whole, the E-BSVR method can be used as a reliable and accurate approach for the prediction and reduction of PCDD/F emissions.

Stable and Effective Online Monitoring and Feedback Control of PCDD/F during Municipal Waste Incineration.

For the long-term operation of municipal solid waste incineration (MSWI), online monitoring and feedback control of polychlorinated dibenzo-p-dioxin and dibenzofuran (PCDD/F) can be used to control the emissions to national or regional standards. In this study, 500 PCDD/F samples were determined by thermal desorption gas chromatography coupled to tunable-laser ionization time-of-flight mass spectrometry (TD-GC-TLI-TOFMS) for 168 h. PCDD/F emissions range from 0.01 ng I-TEQ/Nm3 to 2.37 ng I-TEQ/Nm3, with 44% of values below 0.1 ng I-TEQ/Nm3 (the national standard). In addition, the temperature of the furnace outlet, bed pressure, and oxygen content are considered as key operating parameters among the 13 operating parameters comprising four temperature parameters, four pressure parameters, four flow parameters, and oxygen content. More specifically, maintaining the furnace outlet temperature to be higher than 800 °C, or bed pressure higher than 13 kPa, or the oxygen content stably and above 10% are effective methods for reducing PCDD/F emissions. According to the analysis of the Pearson coefficients and maximal information coefficients, there is no significant correlation between operating parameters and PCDD/F I-TEQ. Only when there is a significant change in one of these factors will the PCDD/F emissions also change accordingly. The feedback control of PCDD/F emissions is realized by adjusting the furnace outlet temperature, bed temperature, and bed pressure to control the PCDD/F to be less than 0.1 ng I-TEQ/Nm3.

Online predicting PCDD/F emission by formation pathway identification clustering and Box-Cox Transformation.

The composition of the fuel and operational conditions change dramatically under the long-term operation of municipal solid waste incineration (MSWI). Therefore, it is difficult to provide effective rapid feedback to control PCDD/F emissions, presenting as International Toxic Equivalent Quantity (I-TEQ). To address this problem, a PCDD/F emission prediction method is developed, based on formation pathway identification clustering (FPIC) and Box-Cox transformation (BCT). Meanwhile, 1,2,4-trichlorobenzene is measured by the thermal desorption gas chromatography coupled to tunable-laser ionization time-of-flight mass spectrometry (TD-GC-TLI-TOFMS). In the method, FPIC includes de novo synthesis, chlorobenzene(CBz)-route synthesis, chlorophenol (CP)-route synthesis, and the chlorination of dibenzofuran (DD) or dibenzodioxin (DF). The PCDD/F emission data was divided into Cluster 1 (I-TEQ>0.1 ng/Nm3) and Cluster 2 (I-TEQ<0.1 ng/Nm3) by FPIC due to PCDD/F in Cluster 1 main from CP-route and PCDD/F in Cluster 2 main from de novo synthesis and CBz-route synthesis. Also, the BCT was used to transform the I-TEQ and 1,2,4-trichlorobenzene data and to construct effective models. The accurate and precise PCDD/F emissions are predicted with the vast majority of error percentage within [ -40%, 40% ], and errors within [ -0.126, 0.016 ] I-TEQ (ng/Nm3). The absolute value of the relative difference between predicted I-TEQ and measured I-TEQ (|RD|) of the linear model constructed by the method has a significant reduction to 20.28%. FPIC and BCT can be used as an effective method to online predict PCDD/F emission in long-term operation thereby allowing the rapid operational feedback to control PCDD/F emission from the incinerator.

Comprehensive diagnosis of PCDD/F emission from three hazardous waste incinerators.

Comprehensive diagnosis of polychlorinated dibenzo-p-dioxin and dibenzofuran (PCDD/F) emissions was systematically conducted on three hazardous waste incinerators (HWIs). Results indicated that PCDD/F mainly existed in the solid phase before the bag filter. This was especially true for higher chlorinated dioxin and furan congeners (hexa-, hepta- and octa-). The aged bag filters tended to increase the gas-phase PCDD/F. Emissions also increased due to PCDD/F desorption from circulated scrubbing solution and plastic packing media used in the wet scrubber. The PCDD/F concentrations were elevated during the start-up process, reaching up to 5.4 times higher than those measured during the normal operating period. The ratios of PCDFs/PCDDs revealed that the surface-catalysed de novo synthesis was the dominant pathway of PCDD/F formation. Installation of more efficient fabric filters, intermittent replacement of circulated scrubbing solution will result in reduced PCDD/F emission. Additionally, 2,3,4,7,8-PeCDF correlated well with the international toxic equivalent quantity (I-TEQ) value, which suggests that 2,3,4,7,8-PeCDF could act as an I-TEQ indicator.