This paper simply depicts knowledge related to BP network and the algorithm first, then introduces BP tool functions supplied by MATLAB for BP neural network research and how to program within the functions; finally explains the advantages supplied by BP tool functions for BP neural network research with BP neural application in pattern identification and curve imitation.

The application of small-pore chabazite-type SSZ-13 zeolites, key materials for the reduction of nitrogen oxides (NO x ) in automotive exhausts and the selective conversion of methane, is limited by the use of expensive N,N,N-trimethyl-1ammonium adamantine hydroxide (TMAdaOH) as an organic structure-directing agent (OSDA) during hydrothermal synthesis. Here, we report an economical and sustainable route for SSZ-13 synthesis by recycling and reusing the OSDA-containing waste liquids. The TMAdaOH concentration in waste liquids, determined by a bromocresol green colorimetric method, was found to be a key factor for SSZ-13 crystallization. The SSZ-13 zeolite synthesized under optimized conditions demonstrates similar physicochemical properties (surface area, porosity, crystallinity, Si/Al ratio, etc.) as that of the conventional synthetic approach. We then used the waste liquid-derived SSZ-13 as the parent zeolite to synthesize Cu ion-exchanged SSZ-13 (i.e., Cu-SSZ-13) for ammonia-mediated selective catalytic reduction of NO x (NH 3 -SCR) and observed a higher activity as well as better hydrothermal stability than Cu-SSZ-13 by conventional synthesis. In situ infrared and ultraviolet−visible spectroscopy investigations revealed that the superior NH 3 -SCR performance of waste liquid-derived Cu-SSZ-13 results from a higher density of Cu 2+ sites coordinated to paired Al centers on the zeolite framework. The technoeconomic analysis highlights that recycling OSDA-containing waste liquids could reduce the raw material cost of SSZ-13 synthesis by 49.4% (mainly because of the higher utilization efficiency of TMAdaOH) and, meanwhile, the discharging of wastewater by 45.7%.

Lipids represent a large family of compounds with highly diverse structures that are involved in complex biological processes. A photocatalytic technique of on-tissue epoxidation of C=C double bonds has been developed for in situ mass spectrometric identification and spatial imaging of positional isomers of lipids. It is based on the plasmonic hot-electron transfer from irradiated gold nanowires to redox-active organic matrix compounds that undergo bond cleavages and generate hydroxyl radicals in nanoseconds. Intermediate radical anions and negative fragment ions have been unambiguously identified. Under the irradiation of a pulsed laser of the third harmonic of Nd3+:YAG (355 nm), the hydroxyl radical-driven epoxidation of unsaturated lipids with different numbers of C=C bonds can be completed in nanoseconds with high yields of up to 95%. Locations of C=C bonds were recognized with diagnostic fragment ions that were produced by either collision with an inert gas or auto-fragmentation resulting from the impact of energetic hot electrons and vibrational excitation. This technique has been applied to the analysis of breast cancer tissues of mice models without extensive sample processes. It was experimentally demonstrated that C=C bonds may be formed at different positions of not only regular mono- or poly-unsaturated fatty acids but also other odd-numbered long-chain fatty acids.

No abstract

The PWARX model is a widely used modeling approach in hybrid systems. In this paper, a PWARX method based on LSTM is proposed. Firstly, the optimal control problem is clarified in the PWARX method. Secondly, the mechanism of PWARX based on the clustering method is described, and the LSTM algorithm is used to optimize the parameters and determine the region boundary conditions. Finally, the validity and reliability of the proposed method is verified by the engine universal characteristic curve, which solves the problem that the region boundary has gaps in the traditional method.