No abstract

The ammonium bisulfate (ABS) widely exists at air preheater. The ABS may deposit and foul at the heating elements of air preheater because of the chemical reaction between SO3 at flue gas side and ammonia slip from SCR excess injection. The heat transfer equation between flue gas side and air side is constructed and simplified using physical and mathematical models accordingly. The finite difference method is applied to solve numerically by means of iterative computation. Based on the NH3 and SO3 concentration data from the real time data in the actual operation and the discrete calculation of the temperature field, the Radian number (Ra) is used to evaluate the possibility of ABS fouling and the developing trend of heating elements at the air preheater. A 1000MW ultra supercritical boiler is selected as example. The ABS deposit area is simulated under different working conditions 100%BMCR, 75% BMCR and 50% BMCR. The possible ABS deposition and fouling is analyzed for operators to evaluate the risk of cold-end and hot-end heating elements plate at air preheater. As the working load decreases lower than 50%BMCR, the deposition and fouling position could extend to the hot-end area of heating elements at air preheater.

Based on the research on the predator–prey model with Holling type response function, a delayed predator–prey system with diffusion term and habitat complexity effect is established, and the effects of time delay and diffusion on dynamical behavior of the system are studied. First, taking habitat complexity as the parameter, the dynamical properties of the system without time delay are studied. By eigenvalue analysis, the sufficient conditions for locally asymptotic stability of the positive equilibrium and globally asymptotic stability of the boundary equilibrium are given, the existence conditions of Hopf bifurcation induced by diffusion term are discussed. In an appropriate range, diffusion makes a family of spatially homogeneous and inhomogeneous periodic solutions bifurcate from the positive equilibrium. Second, taking production delay as the bifurcation parameter, the existence conditions of Hopf bifurcation are given, the method to determine the bifurcation direction and the stability of bifurcating periodic solutions is given by using the center manifold theory and normal form method. Finally, the biological interpretations of the results are given, and some numerical simulations are given to verify the theoretical analysis results.

Deep image denoisers achieve state-of-the-art results but with a hidden cost. As witnessed in recent literature, these deep networks are capable of overfitting their training distributions, causing inaccurate hallucinations to be added to the output and generalizing poorly to varying data. For better control and interpretability over a deep denoiser, we propose a novel framework exploiting a denoising network. We call it controllable confidence-based image denoising (CCID). In this framework, we exploit the outputs of a deep denoising network alongside an image convolved with a reliable filter. Such a filter can be a simple convolution kernel which does not risk adding hallucinated information. We propose to fuse the two components with a frequencydomain approach that takes into account the reliability of the deep network outputs. With our framework, the user can control the fusion of the two components in the frequency domain. We also provide a user-friendly map estimating spatially the confidence in the output that potentially contains network hallucination. Results show that our CCID not only provides more interpretability and control, but can even outperform both the quantitative performance of the deep denoiser and that of the reliable filter, especially when the test data diverge from the training data.Our code and models are publicly available at https://github.com/IVRL/CCID Related WorkImage denoising is a widely studied problem in the literature, particularly the fundamental AWGN removal. Classic image denoising methods make various assumptions to improve de-

No abstract