Search results

The purpose of this paper is to predict quantitative level of stray current leaking to the buried metallic structure by establishing convolution neural network (CNN) model.

First, corrosion experimental system of buried metallic structure is established. The research object of this paper is the polarization potential within 110 min, CNN model is used to predict the quantitative level of stray current leakage using the data from corrosion experimental system further. Finally, results are compared with the method using BP neural network.

Results show that the CNN model has better predictive effect and shorter prediction time than the BP model, the accuracy of which is 82.5507 per cent, and the prediction time is shortened by more than 10 times.

The established model can be used to forecast the level of stray current leakage in the subway system effectively, which provides a new theoretical basis for evaluating the stray current corrosion hazard of buried metallic structure.

This paper aims to thrash out friction and wear properties of automobile brake lining reinforced by lignin fiber and glass fiber in braking process.

ABAQUS finite element software was used to analyze thermo-mechanical coupled field of friction materials. XD-MSM constant speed friction testing machine was used to test friction and wear properties of friction material. Worn surface morphology and mechanism of friction materials were observed by using scanning electron microscope.

The results show that when the temperature was below 350°C, worn mechanism of MFBL was mainly fatigue wear and abrasive wear, and worn mechanism of GFBL was mainly fatigue wear because MFBL contained lignin fiber. Therefore, it exhibits better mechanical properties and friction and wear properties than those of GFBL.

Lignin fiber can improve mechanical properties and friction and wear properties of the automobile brake lining.

The aim of this article is to present a PIN (pedestrian inertial navigation) solution that incorporates altitude error correction, which eliminates the altitude error accurately without using external sensors. The main problem of PIN is the accumulation of positioning errors due to the drift caused by the noise in the sensors. Experiment results show that the altitude errors are significant when navigating in multilayer buildings, which always lead to localization to incorrect floors.

The PIN proposed is implemented over an inertial navigation systems (INS) framework and a foot-mounted IMU. The altitude error correction idea is identifying the most probable floor of each horizontal walking motion. To recognize gait types, the walking motion is described with angular rate measured by IMU, and the dynamic time warping algorithm is used to cope with the different dimension samples due to the randomness of walking motion. After gait recognition, the altitude estimated with INS of each horizontal walking is checked for association with one of the existing in a database.

Experiment results show that high accuracy altitude is achieved with altitude errors below 5 centimeters for upstairs and downstairs routes in a five floors building.

The main limitations of the study is the assumption that accuracy floor altitude information is available.

Our PIN system eliminates altitude errors accurately and intelligently, which benefits from the new idea of combination of gait recognition and map-matching. In addition, only one IMU is used which is different from other approach that use external sensors.

Using Shanghai as an example, the purpose of this paper is to perform grade evaluation and zoning for different land use spaces by GIS by identifying the major restrictive factors in current socio-economic development.

Based on short plate theory, 11 major restrictive indicators that will restrict socio-economic development in Shanghai are identified, and urban land is divided into four subspaces and the restrictive grade evaluation of urban land subspace is achieved with GIS spatial analysis; then, land development zoning is processed according to the results of the evaluation.

In all, 11 major restrictive indicators that will restrict socio-economic development in Shanghai are identified. The restrictive grades of the agricultural production, urban construction and ecological protection subspaces are mainly common, weak and weaker, and the relatively strong restrictive grade of industrial development subspace is mainly concentrated in the more developed industrial districts (counties). The areas of the common and good regions of constructive development and ecological development zones account for 87.4 and 98.3 per cent of each total area, respectively, and urban land still has significant development potential in Shanghai.

This paper proposes various urban land space evaluations and zoning strategies based on restrictive indicators and perspectives, enriching the ideas and methods of urban land use evaluation.

The vulnerability of multinational corporation (MNC) knowledge network is one of the major causes for the failure and even the death of MNCs in the fierce global market competition. Employee turnover and knowledge loss are the triggers for the MNC knowledge network vulnerability and a matter of serious concern in the evolution and development of MNC knowledge network. The purpose of this work is to propose a valid and quantitative measurement method to investigate the influence of employee loss and knowledge loss on the vulnerability of MNC knowledge network.

MNC knowledge network is inherently a heterogeneous network where there are mainly two types of units: employees and their knowledge. Therefore, this paper establishes a weighted super-network model for MNC knowledge network to depict its heterogeneous composition. On the basis of the weighted MNC knowledge super-network, the static and dynamic vulnerability measurement methods are further proposed to investigate and evaluate MNC knowledge network vulnerability.

A real case is given to illustrate the applicability of the proposed weighted MNC knowledge super-network model and the network vulnerability measurement methods. The results show the super-network model proposed in this paper can effectively embody the complex features of MNC knowledge network, and the vulnerability measurement methods can effectively investigate the influence of employee loss and knowledge loss on network vulnerability.

From the perspective of super-network, researchers and practitioners can get a more systematic and deeper understanding of the MNC knowledge network and its human and knowledge resource constitute which are vital for the evolution and development of MNC. Moreover, the MNC knowledge network vulnerability measurement methods can effectively measure and analyze the influence of resource loss on network vulnerability, which can provide a helpful decision support for monitoring and managing of MNC knowledge network vulnerability to reduce its adverse effects.

Online employee reviews have emerged as a crucial information source for business managers to evaluate employee behavior and firm performance. The purpose of this paper is to test the relationship between employee personality traits, derived from online employee reviews and job satisfaction and turnover behavior at the individual level.

The authors apply text-mining techniques to extract personality traits from online employee reviews on Indeed.com based on the Big Five theory. They also apply a machine learning classification algorithm to demonstrate that incorporating personality traits can significantly enhance employee turnover prediction accuracy.

Personality traits such as agreeableness, conscientiousness and openness are positively associated with job satisfaction, while extraversion and neuroticism are negatively related to job satisfaction. Moreover, the impact of personality traits on overall job satisfaction is stronger for former employees than for current employees. Personality traits are significantly linked to employee turnover behavior, with a one-unit increase in the neuroticism score raising the probability of an employee becoming a former employee by 0.6%.

These findings have implications for firm managers looking to gain insights into employee online review behavior and improve firm performance. Online employee review websites are recommended to include the identified personality traits.

This study identifies employee personality traits from automated analysis of employee-generated data and verifies their relationship with employee satisfaction and employee turnover, providing new insights into the development of human resources in the era of big data.

Recognizing every worker's working status instead of only describing the existing construction activities in static images or videos as most computer vision-based approaches do; identifying workers and their activities simultaneously; establishing a connection between workers and their behaviors.

Taking a reinforcement processing area as a research case, a new method for recognizing each different worker's activity through the position relationship of objects detected by Faster R-CNN is proposed. Firstly, based on four workers and four kinds of high-frequency activities, a Faster R-CNN model is trained. Then, by inputting the video into the model, with the coordinate of the boxes at each moment, the status of each worker can be judged.

The Faster R-CNN detector shows a satisfying performance with an mAP of 0.9654; with the detected boxes, a connection between the workers and activities is established; Through this connection, the average accuracy of activity recognition reached 0.92; with the proposed method, the labor consumption of each worker can be viewed more intuitively on the visualization graphics.

With this proposed method, the visualization graphics generated will help managers to evaluate the labor consumption of each worker more intuitively. Furthermore, human resources can be allocated more efficiently according to the information obtained. It is especially suitable for some small construction scenarios, in which the recognition model can work for a long time after it is established. This is potentially beneficial for the healthy operation of the entire project, and can also have a positive indirect impact on structural health and safety.

Night lighting reflects the prosperous development of economic and the increasingly rich and colorful cultural life. Currently, various technical standards, protocols and management specifications have been developed to build a safe, comfortable and economical lighting environment. However, prevailing evaluation systems focus on objective indexes of illumination and have ignored environmental characteristics and subjective feelings and lacked consideration of regional culture, economic benefit, management and maintenance. In this context, a lighting evaluation system combining subjective and objective is proposed for the first time in this study to explore approaches to guide the development of a healthy and comfortable urban night-time environment.

Existing research and relevant lighting standards are analyzed and an evaluation model with a logical hierarchy is constructed by combining with the evaluation theory that is set based on people and the environment. The index weights were scientifically determined on the basis of the analytic hierarchy process (AHP) and fuzzy comprehensive evaluation method. The rationality and validity of the proposed evaluation system is verified in accordance with field projects and case studies.

Taking into account traditional and cultural factors, the evaluation model established has an acceptable accuracy. Evaluation based on subjective-objective combination can provide a scientific basis for the management and optimization of night lighting.

The proposed evaluation system can serve as a guiding reference for other areas of cultural identity and esthetic perspective.

The purpose of this paper is to propose a approach for data visualization and industrial process monitoring.

A deep enhanced t-distributed stochastic neighbor embedding (DESNE) neural network is proposed for data visualization and process monitoring. The DESNE is composed of two deep neural networks: stacked variant auto-encoder (SVAE) and a deep label-guided t-stochastic neighbor embedding (DLSNE) neural network. In the DESNE network, SVAE extracts informative features of the raw data set, and then DLSNE projects the extracted features to a two dimensional graph.

The proposed DESNE is verified on the Tennessee Eastman process and a real data set of blade icing of wind turbines. The results indicate that DESNE outperforms some visualization methods in process monitoring.

This paper has significant originality. A stacked variant auto-encoder is proposed for feature extraction. The stacked variant auto-encoder can improve the separation among classes. A deep label-guided t-SNE is proposed for visualization. A novel visualization-based process monitoring method is proposed.