Leveraging Graph Neural Network with LSTM For Traffic Speed Prediction Lu, Zhilong; Lv, Weifeng; Xie, Zhipu ...
2019 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computing, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI),
2019-Aug.
Conference Proceeding
Accurate traffic forecasting plays an important role in the smart city and is of great significance for urban traffic planning, management, and traffic control. However, road speed prediction is a ...challenge task due to the complex topological structure of road networks and stochastic traffic patterns. For capturing spatial and temporal dependencies simultaneously, in this paper we propose a novel graph neural network based traffic speed forecasting model, the graph Long short term Memory (GLSTM) model which consists Graph neural network (GNN) and Long short term Memory. To the best of our knowledge, this is the first time to combine LSTM and GNN to feed graph-structured data as input for train models in traffic speed prediction. More specifically, at first we construct a unweighted directed graph from road network and feed LSTM cell with graph-structured data. After that, we train the whole model based on encoder-decoder architecture and message-passing mechanism of GNN. Experiments show that our proposed method is able to utilize road structure to capture spatial-temporal dependencies based on GNN while capture long-term dependencies based on LSTM. The result of real world dataset shows that proposed method outperform state-of-the-art baseline methods.
The dynamic stability problems caused by the broadband interfering characteristics of power electronics grid-connected systems have received wide attention due to the large-scale installation of ...renewable energy and intelligent distribution equipment. Measuring the port impedance of grid-connected converters and determining the system stability by using Nyquist stability criterion is the most commonly used method in broadband stability studies. This paper presents an impedance measurement method by generating series disturbance voltage, where the broadband disturbance is generated by means of model predictive control (MPC) with a frequency domain segmentation model. A comparison with theoretical calculations shows that the method enables accurate measurement of the port impedance of grid-connected converters over broadband frequency range, providing a basis for system stability analysis.
In order to meet the demand of broadband miniaturized solid-state power amplifiers for radar systems, a high efficiency power amplifier was designed and implemented using a 0.5um GaN HEMT process. ...The power amplifier is realized by using passive elements on substrates with different permittivity for reducing the size and a compact broadband matching topology for broadband. The experimental results at pulse excitation show that the power amplifier output power is larger than 60W, the gain is over 31dB, and the power added efficiency (PAE) is more than 55% in the 700-1250MHz frequency band, which has a record relative bandwidth over 56% in P/L band over 60W output power.
Highway Traffic Volume Prediction via Stacking KNN, SVR, MLP, RNN Lu, Zhilong; Lv, Weifeng; Xie, Zhipu ...
2018 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computing, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI)
Conference Proceeding
Highway traffic volume prediction is critical for urban planning and relieving traffic stress. In recent years, many researches have been focusing on it by using machine learning algorithms. However, ...different prediction methodologies are required for different evaluation metrics and the prediction performance of single machine learning model is limited. In this paper, we present our models which are respectively inspired from KNN, SVR, MLP and RNN, to predict highway traffic volume, aiming at different evaluation metrics. Based on the revised models, model stacking is used which can boost predictive accuracy by combining the predictions of multiple machine learning models. The experiments show that, when the metrics are specified, the revised models outperform the traditional ones. Moreover, one single model is not always superior to the others and each of them can have encouraging result in certain conditions. By stacking these models, significant improvements can be achieved.
Rare earth element (REE) contents, and Sr and Nd isotopic compositions were measured for three suites of mantle xenoliths from the Kuandian, Hannuoba and Huinan volcanoes in the north of the ...Sino‐Korean Platform. From the correlations of Yb contents with Al/Si and Ca/Si ratios, the peridotites are considered to be the residues of partial melting of the primitive mantle. The chondrite‐normalized REE compositions are diverse, varying from strongly LREE‐depleted to LREE‐enriched, with various types of REE patterns. Metasomatic alteration by small‐volume silicate melts, of mantle peridotites previously variably depleted due to fractional melting in the spinel peridotite field, can account for the diversity of REE patterns. The Sr/Ba versus La/Ba correlation indicates that the metasomatic agent was enriched in Ba over Sr and La, suggestive of its volatile‐rich signature and an origin by fluid‐triggered melting in an ancient subduction zone. The Sr and Nd isotopic compositions of these xenoliths, even from a single locality, vary widely, covering those of Cenozoic basalts in eastern China. The depleted end of the Sr‐Nd isotope correlation is characterized by clearly higher 143Nd/144Nd and a broader range of 87Sr/86Sr compared to MORB. The low‐143Nd/144Nd and high‐87Sr/86Sr data distribution at the other end of data array suggests the existence of two enriched mantle components, like EM1 and EM2. The correlations between 143Nd/144Nd and 147Sm/N4Nd ratios in these xenoliths suggest at least two mantle metasomatic events, i.e. events at 0.6–1.0 Ga and 280–400 Ma ago.
In this technical report, we present TeleChat, a collection of large language models (LLMs) with parameters of 3 billion, 7 billion and 12 billion. It includes pretrained language models as well as ...fine-tuned chat models that is aligned with human preferences. TeleChat is initially pretrained on an extensive corpus containing a diverse collection of texts from both English and Chinese languages, including trillions of tokens. Subsequently, the model undergoes fine-tuning to align with human preferences, following a detailed methodology that we describe. We evaluate the performance of TeleChat on various tasks, including language understanding, mathematics, reasoning, code generation, and knowledge-based question answering. Our findings indicate that TeleChat achieves comparable performance to other open-source models of similar size across a wide range of public benchmarks. To support future research and applications utilizing LLMs, we release the fine-tuned model checkpoints of TeleChat's 7B and 12B variant, along with code and a portion of our pretraining data, to the public community.
With the rapid development of electronic technology and computer technology, the intelligent development of community system becomes a trend. Intelligent community is an intelligent, digital, network ...and convenient residential. In this paper, the design of intelligent community system based on LonWorks technology is introduced, focused on the specific method of remote meter reading and fire alarm monitoring. This article also described the design of Intelligent node based on LonWorks technology and the design of LonMaker control network based on LonWorks technology.
A compressive sensing based circular polarization snapshot spectral imaging system is proposed in this paper to acquire two-dimensional spatial, one-dimensional circular polarization (the right and ...left circular polarization), and one-dimensional spectral information, simultaneously. Using snapshot can collect the entire four-dimensional datacube in a single integration period. The dispersion prism in the coded aperture snapshot spectral imager is replaced by the combination of an Amici prism and a Wollaston prism to implement the spectral shifting along two orthogonal directions, which greatly improves the spectral resolution of the image. The right and left circular polarization components of objects are extracted by the assemble with an achromatic quarter wave-plate and a Wollaston prism. The encoding and reconstruction are illustrated comprehensively. The feasibility is verified by the simulation. It provides us an alternative approach for circular polarization spectral imaging such as defogging, underwater imaging, and so on.
Rare earth element (REE) contents, and Sr and Nd isotopic compositions were measured for three suites of mantle xenoliths from the Kuandian, Hannuoba and Huinan volcanoes in the north of the ...Sino-Korean Platform. From the correlations of Yb contents with Al/Si and Ca/Si ratios, the peridotites are considered to be the residues of partial melting of the primitive mantle. The chondrite-normalized REE compositions are diverse, varying from strongly LREE-depleted to LREE-enriched,with various types of REE patterns. Metasomatic alteration by small-volume silicate melts, of mantle peridotites previously variably depleted due to fractional melting in the spinel peridotite field, can account for the diversity of REE patterns. The Sr/Ba versus La/Ba correlation indicates that the metasomatic agent was enriched in Ba over Sr and La, suggestive of its volatilerich signature and an origin by fluid-triggered melting in an ancient subduction zone. The Sr and Nd isotopic compositions of these xenoliths, even from a single locality, vary widely, covering those of Cenozoic basalts in eastern China. The depleted end of the Sr-Nd isotope correlation is characterized by clearly higher 143Nd/144Nd and a broader range of 87Sr/86Sr compared to MORB. The low-143Nd/144Nd and high-87Sr/86Sr data distribution at the other end of data array suggests the existence of two enriched mantle components, like EM1 and EM2. The correlations between 143Nd/44Nd and 147Sm/N4Nd ratios in these xenoliths suggest at least two mantle metasomatic events, i.e. events at 0.6-1.0 Ga and 280-400 Ma ago.
Non-contact monitoring method for overvoltage based on Pockels effect Deng, Jun; Qian, Hai; Chu, Jinwei ...
2018 IEEE International Symposium on Electromagnetic Compatibility and 2018 IEEE Asia-Pacific Symposium on Electromagnetic Compatibility (EMC/APEMC)
Conference Proceeding
To measure the overvoltage accurately, a non-contact monitoring method based on Pockels effect is proposed. Firstly, the stray capacitance between the energized conductor and the transducer is ...calculated and connected in series with the dividing capacitor of the transducer in order to constitute a capacitive voltage divider, which can convert the high-amplitude over-voltage signal into a low voltage signal. Secondly, adopt the Pockels electro-optic crystal (Bismuth Germanium Oxide, BGO) with the design of dual-light-channel compensation to measure the low voltage signal, then the value of the high-amplitude overvoltage can be calculated. The feasibility and effectiveness of this method are verified by comparing with the contact measurement.