Search

Abstract The consumption of coal in winter heating period in northern China is large, and the combustion generates greenhouse gases that pollute the environment. At the same time, wind abandonment is widespread in northern China, causing waste of energy. In order to solve these problems, this paper proposes to apply clean energy heating and waste wind power generation for heating, and build a multi‐objective optimal dispatching model under the goal of considering customer satisfaction and operating costs. Finally, taking a region in the north of China as an example, the improved genetic algorithm is used to solve the model, the improved genetic algorithm ensures the survival rate of excellent genes, which is more efficient than the traditional genetic algorithm. The example results verify that the use of clean energy heating can increase the wind power consumption space and reduce the heating cost in winter.

In this study, coix seed starch nanoparticles were prepared by nano-precipitation method using three different systems (alkali solution, water-ethanol system and dimethyl sulfoxide (DMSO)-ethanol system). The structures and physicochemical properties of the three nanoparticles were investigated by analysis of their particle characteristics, molecular characteristics, crystallization characteristics and thermal characteristics. The results showed that the nanoparticles prepared using alkaline solution had the highest proportion of nanoparticles (85.65%). Both the nanoparticles prepared using alkaline solution and water-ethanol system had an A-type crystalline structure, while the crystalline structure of the nanoparticles prepared using DMSO-ethanol system changed from A-type to amorphous. The nanoparticles prepared using alkali solution at 45 ℃ had the smallest particle size (348 nm), the highest polydispersity, higher degree of order, relative crystallinity and thermal stability, indicating that alkali solution was the most suitable to prepare coix seed starch nanoparticles. This study may provide a theoretical basis for the preparation of starch nanoparticles and the study of their structural properties as well as the expansion of the application of coix seed starch.

[Purpose/Significance] Smart libraries are considered to be the dominant mode and highest form of future libraries. The smart service it provides has a natural demand for self-service, and readers can obtain it under the condition of "unattended" 24 hours a day for more efficient, convenient and intelligent services.[Method/Process] This paper analyzed the feasibility of 24-hour self-service of smart libraries from three aspects: technology, economy, and operation, and the implementation of the service from five aspects: management system, personnel management, reader management, space transformation, and service innovation, and the constraints that impede the service in smart libraries. [Results/Conclusions] Several suggestions were put forward, such as building a smart space, encouraging and supporting social forces to participate in the construction of private libraries, using strategies such as rural revitalization and cultural poverty alleviation to promote equalization of public cultural services, scientific site selection, and precise services.

This paper aims to explore the molecular mechanism of Enhancer of Zeste Homolog 2 (EZH2)-mediated H3K27me3 in human corneal endothelial cells (HCEC) apoptosis by inhibiting Heme oxygenase-1 (HO-1) transcription to provide a potential target for the treatment of corneal apoptosis. HCECs were cultured in vitro and transfected with si-EZH2, pcDNA3.1-EZH2, pcDNA3.1-HO-1, GSK-J4 (an effective H3K27me3 demethylase inhibitor), and corresponding controls. Western Blot assay was used to detect the levels of EZH2, HO-1, H3K27me3, and apoptosis-related proteins (Bcl-2, Bax, and Cleaved-caspase-3) in HCECs; CCK-8 assay was conducted to detect cell viability and flow cytometry to analyze the apoptosis. HO-1 mRNA levels were detected by RT-qPCR and changes in H3K27me3 levels on the HO-1 promoter were detected by chromatin immunoprecipitation. HCECs transfected with si-EZH2 showed significantly lower EZH2 mRNA and protein levels, higher HCEC viability, lower apoptosis rates, higher antiapoptotic protein Bcl-2 expression, lower proapoptotic protein (Bax and Cleaved-caspase-3) levels, and significantly higher HO-1 expression. HCECs transfected with pcDNA3.1-EZH2 showed the opposite results. EZH2 repressed HO-1 transcription by mediating H3K27me3. H3K27me27 was enriched in the HO-1 promoter and overexpression of EZH2 increased H3K27me27 levels. Promotion of H3K27me3 partially reversed the mitigating effect of si-EZH2 on HCEC apoptosis. Overexpression of HO-1 partially reversed the apoptosis-promoting effects of EZH2 and H3K27me3 on HCECs. EZH2 promotes HCE cell apoptosis by mediating H3K27me3 to inhibit HO-1 transcription. [ABSTRACT FROM AUTHOR]

Story generation, namely, generating a reasonable story from a leading context, is an important but challenging task. In spite of the success in modeling fluency and local coherence, existing neural language generation models (e.g., GPT-2) still suffer from repetition, logic conflicts, and lack of long-range coherence in generated stories. We conjecture that this is because of the difficulty of associating relevant commonsense knowledge, understanding the causal relationships, and planning entities and events with proper temporal order. In this paper, we devise a knowledge-enhanced pretraining model for commonsense story generation. We propose to utilize commonsense knowledge from external knowledge bases to generate reasonable stories. To further capture the causal and temporal dependencies between the sentences in a reasonable story, we use multi-task learning, which combines a discriminative objective to distinguish true and fake stories during fine-tuning. Automatic and manual evaluation shows that our model can generate more reasonable stories than state-of-the-art baselines, particularly in terms of logic and global coherence.

To advance multi-domain (cross-domain) dialogue modeling as well as alleviate the shortage of Chinese task-oriented datasets, we propose CrossWOZ, the first large-scale Chinese Cross-Domain Wizard-of-Oz task-oriented dataset. It contains 6K dialogue sessions and 102K utterances for 5 domains, including hotel, restaurant, attraction, metro, and taxi. Moreover, the corpus contains rich annotation of dialogue states and dialogue acts on both user and system sides. About 60% of the dialogues have cross-domain user goals that favor inter-domain dependency and encourage natural transition across domains in conversation. We also provide a user simulator and several benchmark models for pipelined task-oriented dialogue systems, which will facilitate researchers to compare and evaluate their models on this corpus. The large size and rich annotation of CrossWOZ make it suitable to investigate a variety of tasks in cross-domain dialogue modeling, such as dialogue state tracking, policy learning, user simulation, etc.