No abstract

Discrete, thick-film, combinatorial libraries of ceramics are made by computer controlled mixing of well-dispersed suspensions of commercially available powders followed by deposition of droplets on a substrate for drying and subsequent firing. The substrate influences the shape and size of the samples through spreading effects. Later, at the firing stage, interaction with the substrate can involve reactions that adulterate the composition or low adhesion that can result in sample loss during transport. We describe substrate pretreatment with a perfluorinated silane to provide control of droplet size and prevent spreading. Substrate compatibility during high temperature firing is more problematic and we discuss the limited material choices. One approach seeks a substrate that will support the library from beginning to end of the preparatory stage and during storage. An alternative method involves firing samples on low adhesion substrates followed by pneumatic transfer to a silver electroding paste and a low temperature firing. This method provides sufficiently strong adhesion to allow libraries of samples to be surface-ground a give a flat parallel configuration on a common electrode together with robust construction for storage and transport.

The characteristics of the massive data is inspected which is accumulated by the energy management system of industrial enterprises during the production process. They are real-time, massive, non-linear, and even unstructured, meanwhile some implicit correlation are hidden between them. These correlations have not yet been fully learned and unearthed. Here big data technology is proposed to improve the availability of energy data for the energy saving and consumption reduction. The general pattern of big data application is proposed in energy management systems. On this basis, some algorithms are applied to the actual management process, including parameter prediction, status monitoring, operation optimization and performance evaluation. Finally, two practical examples of the constructed model are presented including compressed air prediction and Operation prediction of air-conditioning system. The results show that big data technology can improve the efficiency of data acquisition, expand data applications, and exert greater data value.

No abstract