Short-Term Household Load Forecasting Based on Attention Mechanism and CNN-ICPSO-LSTM

Accurate load forecasting forms a crucial foundation for implementing household demand response plans andoptimizing load scheduling. When dealing with short-term load data characterized by substantial fluctuations,a single prediction model is hard to capture temporal features effectively, resulting in diminished predictionaccuracy. In this study, a hybrid deep learning framework that integrates attention mechanism, convolution neuralnetwork (CNN), improved chaotic particle swarm optimization (ICPSO), and long short-term memory (LSTM), isproposed for short-term household load forecasting. Firstly, the CNN model is employed to extract features fromthe original data, enhancing the quality of data features. Subsequently, the moving average method is used for datapreprocessing, followed by the application of the LSTM network to predict the processed data. Moreover, the ICPSOalgorithm is introduced to optimize the parameters of LSTM, aimed at boosting the model’s running speed andaccuracy. Finally, the attention mechanism is employed to optimize the output value of LSTM, effectively addressinginformation loss in LSTM induced by lengthy sequences and further elevating prediction accuracy. According tothe numerical analysis, the accuracy and effectiveness of the proposed hybrid model have been verified. It canexplore data features adeptly, achieving superior prediction accuracy compared to other forecasting methods forthe household load exhibiting significant fluctuations across different seasons.

Building Indoor Dangerous Behavior Recognition Based on LSTM-GCN with Attention Mechanism

Building indoor dangerous behavior recognition is a specific application in the field of abnormal human recognition.A human dangerous behavior recognition method based on LSTM-GCN with attention mechanism(GLA)model was proposed aiming at the problem that the existing human skeleton-based action recognition methods cannot fully extract the temporal and spatial features.The network connects GCN and LSTMnetwork in series,and inputs the skeleton sequence extracted by GCN that contains spatial information into the LSTM layer for time sequence feature extraction,which fully excavates the temporal and spatial features of the skeleton sequence.Finally,an attention layer is designed to enhance the features of key bone points,and Softmax is used to classify and identify dangerous behaviors.The dangerous behavior datasets are derived from NTU-RGB+D and Kinetics data sets.Experimental results show that the proposed method can effectively identify some dangerous behaviors in the building,and its accuracy is higher than those of other similar methods.

Long noncoding RNAs in hepatocellular carcinoma: Novel insights into their mechanism

Hepatocellular carcinoma(HCC) is the predominant subject of liver malignancies which arouse global concern. Advanced studies have found that long noncoding RNAs(lnc RNAs) are differentially expressed in HCC and implicate they may play distinct roles in the pathogenesis and metastasis of HCC. However, the underlying mechanisms remain largely unclear. In this review, we summarized the functions and mechanisms of those known aberrantly expressed lncR NAs identified in human HCC tissues. We hope to enlighten more comprehensive researches on the detailed mechanisms of lncR NAs and their application in clinic, such as being used as diagnostic and prognostic biomarkers and the targets for potential therapy. Although studies on lncR NAs in HCC are still deficient, an improved understanding of the roles played by lncR NAs in HCC will lead to a much more effective utilization of those lnc RNAs as novel candidates in early detection, diagnosis, prevention and treatment of HCC.

Characteristics and Formation Mechanism of Giant Long-Runout Landslide: A Case Study of the Gamisi Ancient Landslide in the Upper Minjiang River, China

The upper reaches of the Minjiang River are in the eastern margin of the Tibetan Plateau,where active faults are well developed and earthquakes frequently occur.Anomalous climate change and the extremely complex geomechanical properties of rock and soil have resulted in a number of geohazards.Based on the analysis of remote sensing interpretations,geological field surveys,geophysical prospecting and geological dating results,this paper discusses the developmental characteristics of the Gamisi ancient landslide in Songpan County,Sichuan Province,and investigates its geological age and formation mechanism.This study finds that the Gamisi ancient landslide is in the periglacial region of the Minshan Mountain and formed approximately 25 ka BP.The landslide initiation zone has a collapse and slide zone of approximately 22.65×106–31.7×106 m3 and shows a maximum sliding distance of approximately 1.42 km,with an elevation difference of approximately 310 m between the back wall of the landslide and the leading edge of the accumulation area.The landslide movement was characterized by a high speed and long runout.During the sliding process,the landslide body eroded and dammed the ancient Minjiang River valley.The ancient river channel was buried 30-60 m below the surface of the landslide accumulation area.Geophysical prospecting and drilling observations revealed that the ancient riverbed was approximately 80-100 m thick.After the dam broke,the Minjiang River was migrated to the current channel at the leading edge of the landslide.The Gamisi ancient landslide was greatly affected by the regional crustal uplift,topography,geomorphology and paleoclimatic change.The combined action of periglacial karstification and climate change caused the limestone at the rear edge of the landslide fractured,thus providing a lithological foundation for landslide occurrence.Intense tectonic activity along the Minjiang Fault,which runs through the middle and trailing parts of the Gamisi ancient landslide,may have been the main factor inducing landsliding.Studying the Gamisi ancient landslide is of great significance for investigating the regional response to paleoclimatic change and geomorphologic evolution of the Minjiang Fault since the late Pleistocene and for disaster prevention and mitigation.

Dynamical Mechanisms of Effects of Landslides on Long Distance Oil and Gas Pipelines

According to the investigations on the oil and gas pipelines such as the Lan-Cheng-Chong pipeline and the Southwest pipeline, there are two ways of laying pipeline： pipelines paralleling （approximately） to the main slide direction and pipelines perpendicular （approximately） to the main slide direction. If earth-retaining walls have been built for pipelines paralleling to the main slide direction, they will prevent the lands from sliding; On the contrary, without earth-retaining walls, the sharp broken rocks in the backfilling soil will scratch the safeguard of the pipeline when the landslides take place. Pipelines perpendicular to the main slide direction can be classified into four types according to the relative positions between pipelines and landslides： Pipelines over the slide planes, pipelines inside the fracture strips of slide planes, pipelines below the slide planes and pipelines behind the backsides of landslides. The different dynamical mechanisms of the process in which landslide acts against pipelines are analyzed based on whether the pipelines are equipped with fixed frusta, because the sliding resistance depends on whether and how many fixed frusta are equipped and the distance between frusta.

Regulatory mechanisms of long non-coding RNAs

Long non-coding RNAs(lncRNAs) belong to a large and complex family of RNAs, which play many important roles in regulating gene expression. However, the mechanism underlying the dynamic expression of lncRNAs is still not very clear. In order to identify lncRNAs and clarify the mechanisms involved, we collected basic information and highlighted the mechanisms underlying lncRNA expression and regulation. Overall, lncRNAs are regulated by several similar transcription factors and protein-coding genes. Epigenetic modification(DNA methylation and histone modification) can also downregulate lncRNA levels in tissues and cells. Moreover, lncRNAs may be degraded or cleaved via interaction with miRNAs and miRNAassociated protein complexes. Furthermore, alternative RNA splicing(AS) may play a significant role in the post-transcriptional regulation of lncRNAs.

The Long-Term Effective Mechanism of Rural Poverty Alleviation in China from the Perspective of Ecological Management

Based on the ecological environmental situation of poverty-stricken areas in China and the domestic and foreign research results,the long-term effective mechanism of sustainable poverty alleviation in China is established(the ideological premise is ecological culture,the material base is ecological economy and the basic guarantee is ecological system) from the perspective of ecological management.To be specific,ecological culture,the ideological premise of rural sustainable poverty alleviation in China,includes two aspects:the first one is upholding the Marxism,passing on the Chinese traditional ecological wisdom and fostering ecological consciousness;the second one is mirroring the thought of western modern ecological ethics,emphasizing ecological criticism and redoubling the ecological education.As for ecological economy,the material guarantee for sustainable poverty alleviation in Chinese rural poverty-stricken areas,also contents two aspects.The first one is promoting the way of "clean production" and developing ecological industry;the second one is building ecological concept of consumption and establishing the ecological lifestyle.In addition,ecological system,the basic guarantee of sustainable poverty alleviation in Chinese rural poverty-stricken areas,covers three terms.The first one is implementing the ecological policies and stipulating ecological plan;the second one is establishing perfect ecological legislation and ecological system and intensifying their implementation;the third one is enforcing ecological management in the process of sustainable poverty alleviation.Through the establishment of the above mechanism,we look forward to realizing sustainability of economic development and poverty alleviation effects in the process of poverty alleviation in Chinese rural areas,as well as ecological management of the poverty-stricken areas.

Countermeasures for Establishing a Long-term Security Mechanism for Landless Farmers

In the process of urbanization,farmers will inevitably lose their land,which is resulted from the large-scale requisition of rural collective land.The imperfect social security system in rural areas has resulted in a large number of landless farmers being transformed into urban poor people.Therefore,proper resettlement of landless farmers and the establishment of a long-term security mechanism for ensuring the basic livelihood of landless farmers are important issues that need to be resolved in the process of China’s social and economic development.

通用人工智能中的Long-Context技术分析

阐述通用人工智能中的Long-Context技术原理,提出该技术的解决方案,包括位置编码外推、注意力机制计算优化、模型结构优化和检索增强生成。分析硬件系统的瓶颈,探讨Long-Context在通用人工智能时代的发展趋势。

Countermeasures on improving the charge rules and long-acting operation mechanism for the treating residential domestic sewage and household garbage in rural Yunnan

Improving the rural living environment in rural China is one of the key tasks for the country to accomplish its goal of building a moderately prosperous society by 2020. Yunnan has highly focused on the task of the treatment of residential domestic sewage and household garbage in rural area. Many efforts and resources have been put into this field in Yunnan since 2016. Progress has been made to increase the coverage rate of the sewage and household garbage treatment facilities. Seventy-five percent of total administrative villages have built up garbage transportation system and treatment facilities. Sixty-three percent of towns have collected and treated the residential domestic sewage by constructing various scale sewage treatment stations. However, the lack of the long-acting operation mechanism and the imperfection of the charge rules for the sewage and garbage treatment facilities have become problems that would hinder the achieving of the environmental goals in Yunnan. The reasons were elaborated on the basis of the local actual situations. Therefore, it is significant to improve the charge rules and frame the long-acting operation mechanism by strengthening the governance capacity, frame an overall mechanism and encourage the mass to be involved in the improvements of the living environment in rural Yunnan.

Alkali-Silica Reaction Inhibited by LiOH and Its Mechanism

A high alkali reactive aggregate zeolitization perlite was used to test the long term effectiveness of LiOH in inhibiting alkali silica reaction.In this paper,the rigorous conditions were designed that the mortar bars had been cured at 80℃ for 3 years after autoclaved 24 hours at 150℃.Under this condition,LiOH was able to inhibit the alkali silica reaction long term effectiveness.Not only the relationship between the molar ratio of n(Li)/(Na) and the alkali contents in systems was established, but also the governing mechanism of such effects was also studied by SEM.

Evolution of Microstructure and Mechanical Property during Long-Term Aging in Udimet 720Li

Thermal stabilities of microstructure and mechanical property have been investigated on superalloy U720Li, which is of great interest of application for jet engine and land-based turbine disc. The results showed that, the primary and secondary gamma' particles maintain good thermal stability at 650 and 700 degreesC with aging time up to 3000 h, while the tertiary gamma' is apparently dependent on aging temperature and time. The tertiary gamma' particles undergo a procedure of coarsening, dissolution and eventually complete disappearance with the increasing of aging time and temperature. They exhibit unusual high sensibility upon aging temperature, which is attributed to the lattice misfit between the gamma' precipitates and the matrix in the alloy. The grain boundary phase M23C6 remains stable without forming of sigma phase even with aging time up to 3000 h at 700 degreesC. Microhardness decreases apparently with increasing aging time and aging temperature. Theoretical analysis based on dislocation mechanism indicates that the change of microhardness should be attributed to the evolution of the tertiary gamma' during aging.

Comparison between the preparation,structure and mechanical properties of long fiber reinforced thermoplastics and short fiber reinforced thermoplastic

This article summarizes the comparison between the preparation, structure and mechanical properties of long fiber reinforced thermoplastics （LFT） and short fiber reinforced thermoplastics （SFT）. Both of the experiment and theory results showed that the mechanical properties of long glass fiber reinforced thermoplastics pellets （LGFRT） have been enhanced better than that of short glass fiber reinforced thermoplastics pellets （SGFRT） manufactured by molding procession. After regulation of the relative humidity by 50 % , the mechanical properties of 30 % （ weight percent） short glass fiber content in SFT （ SFT-PA6-SGF30 ） are similar to that of 40 % long glass fiber content in LFT. Howev- er, the density of the latter is about 17 % lower than that of the former. Thus, the corresponding weight of products is reduced by 13 % ;output rate is increased by 21% , and the cost is therefore significantly lowered. And it has the fol- lowing advantages： impact strength is increased by 87 % ; the proportion is reduced by 20 % ; molding cycle is short- ened by 10 % ;materials cost is saved by 20 % -30 % and the final total cost is saved by 30 % -40 %. So LFT （LFT-PP-LGF40） can replace SFT （SFT-PA6-SGF30） with the similar basic mechanical properties under normal tem- perature or 160 ℃ lower.

Attention-based long short-term memory fully convolutional network for chemical process fault diagnosis

A correct and timely fault diagnosis is important for improving the safety and reliability of chemical processes. With the advancement of big data technology, data-driven fault diagnosis methods are being extensively used and still have considerable potential. In recent years, methods based on deep neural networks have made significant breakthroughs, and fault diagnosis methods for industrial processes based on deep learning have attracted considerable research attention. Therefore, we propose a fusion deeplearning algorithm based on a fully convolutional neural network(FCN) to extract features and build models to correctly diagnose all types of faults. We use long short-term memory(LSTM) units to expand our proposed FCN so that our proposed deep learning model can better extract the time-domain features of chemical process data. We also introduce the attention mechanism into the model, aimed at highlighting the importance of features, which is significant for the fault diagnosis of chemical processes with many features. When applied to the benchmark Tennessee Eastman process, our proposed model exhibits impressive performance, demonstrating the effectiveness of the attention-based LSTM FCN in chemical process fault diagnosis.

MICROSTRUCTURE AND MECHANICAL PROPERTIES OF Fe-Ni-Co-BASE SUPERALLOY AFTER LONGTERM AGING AT 650℃

The microstructure and mechanical properties of a Ni-Fe-Co base superalloy with low thermal expansion coefficient after aging at 650℃ for 200,500,1000 and 2000 h have been investigated.It was found that they are stable at 650℃.This alloy is believed to.fill the re- quirements for long time service to aircraft engine.

Design,construction and mechanical behavior of relics of complete large Longyou rock caverns carved in argillaceous siltstone ground

This paper presents a comprehensive summary of data, analyses and findings from the investigations over the past twelve years about the relics of large Longyou rock caverns carved about 2 000 years ago at shallow depths in argillaceous siltstone. The paper presents the typical features associated with the rock caverns. They include structures, large spans, portals, extreme shallow-buried depths, imprints, drainages, inclined ceiling, inclined sidewalls, slender rock pillars, rock staircases, site and strata selections, caving lighting, carving method, and underground construction surveying. They are used to reconstruct and highlight the design and construction methods adopted by the ancients. The paper further demonstrates that the relics of the complete large rock caverns are a consequence of coincidental combinations of ancient human effort and natural factors. The full occupation of water with weak acidity in the large rock caverns with the soft surrounding rocks of weak alkalinity is found to be the main factor ensuring and preserving the caverns to have been stable and integral over 2 000 years. However, the five unwatered complete rock cavern relics have been experiencing various deteriorations and small failures including cracks, seepage, small rock falls and delaminating ceiling rocks. Although these deteriorations have been repaired and stabilized effectively, the paper demonstrates that an entire roof collapse failure is highly possible in the near future to each of the five unwatered rock cavern relics. The findings presented in this paper are also invaluable both to the long-term protection and preservation of the large rock cavern relics of national and international interests and importance, and to extend and enrich our experience and knowledge on the long-term stability and integrity of man-made underground rock cavern engineering projects.

Deep Learning Applied to Computational Mechanics:A Comprehensive Review,State of the Art,and the Classics

Three recent breakthroughs due to AI in arts and science serve as motivation:An award winning digital image,protein folding,fast matrix multiplication.Many recent developments in artificial neural networks,particularly deep learning(DL),applied and relevant to computational mechanics(solid,fluids,finite-element technology)are reviewed in detail.Both hybrid and pure machine learning(ML)methods are discussed.Hybrid methods combine traditional PDE discretizations with ML methods either(1)to help model complex nonlinear constitutive relations,(2)to nonlinearly reduce the model order for efficient simulation(turbulence),or(3)to accelerate the simulation by predicting certain components in the traditional integration methods.Here,methods(1)and(2)relied on Long-Short-Term Memory(LSTM)architecture,with method(3)relying on convolutional neural networks.Pure ML methods to solve(nonlinear)PDEs are represented by Physics-Informed Neural network(PINN)methods,which could be combined with attention mechanism to address discontinuous solutions.Both LSTM and attention architectures,together with modern and generalized classic optimizers to include stochasticity for DL networks,are extensively reviewed.Kernel machines,including Gaussian processes,are provided to sufficient depth for more advanced works such as shallow networks with infinite width.Not only addressing experts,readers are assumed familiar with computational mechanics,but not with DL,whose concepts and applications are built up from the basics,aiming at bringing first-time learners quickly to the forefront of research.History and limitations of AI are recounted and discussed,with particular attention at pointing out misstatements or misconceptions of the classics,even in well-known references.Positioning and pointing control of a large-deformable beam is given as an example.

基于注意力机制LSTM的电离层TEC预测

电离层总电子含量(Total Electron Content,TEC)的监测与预报是空间环境研究的重要内容,对卫星通讯和导航定位等有重要意义.TEC值影响因素较多,很难确定精确物理模型来对其进行预测.本文设计了基于注意力机制的LSTM模型(Att-LSTM),采用过去24小时TEC观测数据对未来TEC进行预测.选择北半球东经100°上,每2.5°纬度选择一个位置,共计36个位置来验证本文提出模型的性能,并与主流的深度学习模型如DNN、RNN、LSTM进行对比实验.取得了如下成果:(1)在选定的36个地区未来2小时单点预测上,基于本文的Att-LSTM模型的TEC预测性能明显优于其他对比模型;(2)讨论了纬度对Att-LSTM预测未来2小时TEC值时性能的影响,发现在北纬0°到60°之间,Att-LSTM预测性能随着纬度的升高而略有降低,在北纬62.5°~87.5°之间,模型预测性能出现扰动,预测效果略差;(3)讨论了磁暴期和磁静期模型的预测性能,发现无论是磁暴期还是磁静期,本文模型预测性能均较好;(4)还讨论了对未来多时点预测效果,实验结果表明,本文所提出的模型对未来2、4个小时的预测拟合度R-Square均超过0.95,预测结果比较可靠,对未来6、8、10个小时预测拟合度最高为0.7934,预测拟合度R-Square下降迅速,预测结果不可靠.

融合多源异构气象数据的光伏功率预测模型

高精度光伏功率预测对提高电力系统运行效率具有重要意义。光伏功率受多种因素影响,其中云层的变化是最主要的不确定因素。传统光伏功率预测方法没有充分考虑云的3维结构和气象要素对光伏功率的影响。因此,该文提出一种融合多源异构气象数据的多源变量光伏功率预测模型(MPPM)。MPPM的核心包括时空条件扩散模型(STCDM)、注意力堆叠LSTM网络(ASLSTM)和多维特征融合模块(MFFM)。STCDM模型通过对2维卫星云图进行精确预测,消除了云层边界处的模糊现象。ASLSTM模型则提取了3维天气研究与预报模式(WRF)气象要素特征。MFFM模块将2维卫星云图特征和3维WRF气象要素特征进行融合,以得到未来1 h光伏功率预测结果。该文分别利用STCDM模型和MPPM模型开展卫星云图预测实验和光伏功率预测实验。实验结果显示,STCDM模型预测1 h内卫星云图的结构相似性指数(SSIM)达到0.914,MPPM模型预测1 h内光伏功率的相关系数(CORR)达到0.949,优于所有对比算法。

公立医院高质量发展背景下医院运营管理工作的探索与思考

加强运营管理是公立医院高质量发展的重要内容,也是实现公立医院高质量发展的必然要求。坚持和加强党对公立医院的全面领导,建立不断完善的运营管理体系和长效机制,健全运营管理人才学习培养机制,以智慧医院建设为契机,提高医院智慧运营信息化管理水平,对公立医院高质量发展具有重要意义。