Building performance management requires massive data input; however, the relevant data are separated and heterogeneous; thus, it prevents a comprehensive building performance management. Building information modeling brings a new way to capture rich information of a building, and has great potential in data interoperability for building performance management. This article presents a scalable building information modeling–based cyber-physical platform for building performance monitoring to integrate heterogeneous data from different buildings. A smart sensor network based on Arduino and standard protocol is installed for data sensing and collection. A building information modeling–based sensing information model integrating heterogeneous data in a unified structure is proposed, and a scalable NoSQL database is established to store data in a cloud environment. A series of RESTful web services is developed to share data for building performance management applications. The proposed platform is developed taking the advantage of horizontal scalability of NoSQL database, and the data schema and services are generated automatically based on the unified data model. The platform has collected data from 77 buildings in China, and the results of a case study show the platform brings a new paradigm in collecting, storing, integrating, and sharing of sensor data and building information for building performance monitoring and analytics.
Buildings account for a majority of the primary energy consumption of the human society, therefore, analyses of building energy consumption monitoring data are of significance to the discovery of anomalous energy usage patterns, saving of building utility expenditures, and contribution to the greater environmental protection effort. This paper presents a unified framework for the automatic extraction and integration of building energy consumption data from heterogeneous building management systems, along with building static data from building information models to serve analysis applications. This paper also proposes a diagnosis framework based on density-based clustering and artificial neural network regression using the integrated data to identify anomalous energy usages. The framework and the methods have been implemented and validated from data collected from a multitude of large-scale public buildings across China.
1 September 2018 in Journal of Management Science and Engineering
We study the problem of training named entity recognition (NER) models using only distantly-labeled data, which can be automatically obtained by matching entity mentions in the raw text with entity types in a knowledge base. The biggest challenge of distantly-supervised NER is that the distant supervision may induce incomplete and noisy labels, rendering the straightforward application of supervised learning ineffective. In this paper, we propose (1) a noise-robust learning scheme comprised of a new loss function and a noisy label removal step, for training NER models on distantly-labeled data, and (2) a self-training method that uses contextualized augmentations created by pre-trained language models to improve the generalization ability of the NER model. On three benchmark datasets, our method achieves superior performance, outperforming existing distantly-supervised NER models by significant margins.
Nuclide migration poses a problem to the potential geological disposal site performance assessment of radioactive waste. Effective diffusivity of clay is an important factor in radionuclide retention. In this study, a set of diffusion experiments were performed with radioactive water tracers (HTO and 36Cl−) on the Téguline Clay using the in-diffusion method, and the effect of location was emphasised. The results show that the apparent diffusion coefficient (Da) did not vary linearly with the depth of the Téguline Clay. From the depth of 0–60 m, the effective diffusion coefficients of HTO and 36Cl− were 5.2–10.3 (×10−11 m²s−1) and 1.3 –6.3 (×10−11 m²s−1), respectively. Téguline Clay at different locations and the same depth showed similar results. The results showed that all the diffusion coefficients of 36Cl− in the Téguline Clay at different positions were lower than those of HTO. HTO exhibited lower anisotropy in the region, while HTO and 36Cl− showed anisotropies between 1 and 2.2. Numerical analysis predicted different initial concentrations and time affect the diffusion of contaminants in Téguline Clay. These findings can provide ideas for the prevention and control of contaminant in radioactive waste disposal. GRAPHICAL ABSTRACT
Relative to non-psychiatric controls, people with schizophrenia have higher rates of cannabis and tobacco use. Those who co-use cannabis and tobacco, may be particularly vulnerable to heightened cannabis cravings and withdrawal symptoms, resulting in a greater risk of addiction substitution, particularly, increases in cigarette use to compensate for reductions in cannabis use. We conducted a secondary analysis of 28-days of cannabis abstinence in participants with and without schizophrenia to investigate cannabis craving, withdrawal, and cigarette use following 7-days of abstinence.