Predicting traffic accident duration is necessary for ensuring traffic safety. Several attempts have been made to achieve high prediction accuracy, but researchers have not considered traffic accident text data in much detail. The limited text data of the first report on an incident describes the characteristics of an accident that are initially available. This paper uses text data fusing and ensemble learning algorithms to build a model to predict an accident’s duration, and a preprocessing scheme of accident duration text data is established. Next, the random forest (RF) algorithm is applied to select feature variables of text data related to the traffic incident duration. Last, a text feature vector is introduced to models such as decision tree, k nearest neighbor, support vector regression, random forest, Gradient Boosting Decision Tree, and Xtreme Gradient Boosting. Our results show that the improved RF model has good prediction accuracy with RMSE, MAPE and R2. From this, the textual factors important to determining the duration of the accident are identified. Further, we investigated that the cumulative importance of 60% is sufficient for traffic accident prediction using text data. These results provide insights into minimizing traffic congestion related to accidents and contribute to the input optimization in text prediction.
The estimates of total zenith delay are derived using Bernese GPS Software V4. 2 based on GPS data every 30 s from the first measurement experiment of a ground-based GPS network in Chengdu Plain of Southwest China during the period from July to September 2004. Then the estimates of 0.5 hourly precipitable water vapor (PWV) derived from global positioning system (GPS) are obtained using meteorological data from automatic weather stations (AWS). The comparison of PWV derived from GPS and those from radiosonde observations is given for the Chengdu station, with RMS (root mean square) differences of 3.09m. The consistency of precipitable water vapor derived from GPS to those from radiosonde is good. It is concluded that Bevis' empirical formula for estimating the weighted atmospheric mean temperature can be applicable in Chengdu area because the relationship of GPS PWV with Bevis' formula and GPS PWV with radiosonde method shows a high correlation. The result of this GPS measurement experiment is helpful both for accumulating the study of precipitable water vapor derived from GPS in Chengdu areas located at the eastern side of the Tibetan Plateau and for studying spatial-temporal variations of regional atmospheric water vapor through many disciplines cooperatively.
Smart transportation relies on data collection, transmission, processing, and release, involving various terminal devices, control systems, central platforms, and communication links, so its control process is more complicated. In order to improve the operation efficiency of the intelligent traffic control system, based on the open Internet of Things and machine learning, this paper builds an intelligent three-way intelligent traffic control system, sets various parameters, and builds a simulation model using cellular automata as a platform. Moreover, in order to study the performance of the model, the model constructed in this paper is compared with the model of the traditional road traffic control system. In addition, this paper analyzes the model constructed in this paper through the statistics of the highest vehicle flow on the road and the relationship between road occupancy and vehicle speed. The research results show that the model constructed in this paper has good performance and can be applied to intelligent traffic control.
In order to determine the importance of influencing factors of energy consumption in oilfield water injection systems, the distribution of energy loss in the water injection system was analyzed, the factors affecting the energy consumption of the water injection system were determined, and an evaluation index system for the energy consumption of the water injection system was established. This indicator system covers all links and all energy loss nodes of the energy loss of the water injection system, thereby an evaluation model for influencing factors of energy consumption in water injection system based on entropy weight - grey correlation method was built. Use the entropy weight method to get the ranking of the importance of energy consumption indicators; use the gray correlation method to determine the correlation between each water injection system and energy consumption factors. The application results show that the entropy weight-grey correlation method proposed in this paper can effectively obtain the importance of the energy consumption factors of the oilfield water injection system, and provide scientific guidance for the daily management and targeted optimization of the water injection system.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.