Forecasting of turbine heat rate with online least squares support vector machine based on gravitational search algorithm

Generalized refined composite multiscale fuzzy entropy and multi-cluster feature selection based intelligent fault diagnosis of rolling bearing

Extracting the failure related information from vibration signals is a very important aspect of vibration-based fault detection for rolling bearing Multiscale entropy and its improvement, multiscale fuzzy entropy (MFE), are significant complexity measure tools of time series. They have been successfully applied to extract vibration failure features for rolling bearing condition monitoring . However, MFE over different scales will fluctuate with increase of scale factor. A new nonlinear dynamic parameter termed generalized refined composite multiscale fuzzy entropy (GRCMFE) is firstly developed to enhance the performance of MSE and MFE in data complexity measurement. Then three algorithms are developed and compared with MSE and MFE, as well as two algorithms of generalized MFE to verify the availability and superiority by analyzing two kinds of noise signals. In addition, based on three algorithms of GRCMFE, a novel fault diagnosis approach for rolling bearing is proposed with linking multi-cluster feature selection for supervised learning and the gravitational search algorithm optimized support vector machine for failure pattern recognition. Last, the proposed fault diagnostic approach was utilized to analyze two kinds of bearing test data sets. Analysis results indicate that our proposed fault diagnosis approach could effectively extract nonlinear dynamic complexity information and gets the highest identifying rate and the best performance among the comparative approaches.

Research on the Adaptive Control in Sugar Evaporative Crystallization Using LSSVM and SaDE-ELM

The process of sugar evaporative crystallization is a nonlinear process with large time lag and strong coupling. It is difficult to establish a reasonable mechanism model. In this paper, we use the data driving modeling method to establish an Adaptive Control model for batch boiling sugar crystallization process. First, by analyzing the main influencing factors of the evaporative crystallization process of intermittent boiling sugar, the most important two parameters, brix and liquid level, are selected as the control object. The self-adaptive differential evolution Extreme Learning Machine (SaDE-ELM) is used to construct the control model. A least squares support vector machine (LSSVM) is established and connected in the control loop to control the opening of the feed valve so that to control the feed flowrate according to the objective values of syrup Brix and liquid level. Experiments are conducted and the obtained data are used to train and verify the learning machines. Experiments indicate that the learning machines are able to realize adaptive control to key parameters of the crystallization process. Comparison of different neural networks indicates that the LSSVM performs better than BP, RBF and ELM and SaDE-ELM with prediction error of below 0.01, and training time of below 0.05 s.

A Hybrid Short-Term Traffic Flow Prediction Model Based on Singular Spectrum Analysis and Kernel Extreme Learning Machine

Short-term traffic flow prediction is one of the most important issues in the field of intelligent transport system (ITS). Because of the uncertainty and nonlinearity, short-term traffic flow prediction is a challenging task. In order to improve the accuracy of short-time traffic flow prediction, a hybrid model (SSA-KELM) is proposed based on singular spectrum analysis (SSA) and kernel extreme learning machine (KELM). SSA is used to filter out the noise of traffic flow time series. Then, the filtered traffic flow data is used to train KELM model, the optimal input form of the proposed model is determined by phase space reconstruction, and parameters of the model are optimized by gravitational search algorithm (GSA). Finally, case validation is carried out using the measured data of an expressway in Xiamen, China. And the SSA-KELM model is compared with several well-known prediction models, including support vector machine, extreme learning machine, and single KLEM model. The experimental results demonstrate that performance of the proposed model is superior to that of the comparison models. Apart from accuracy improvement, the proposed model is more robust.

An Intelligent Weather Station

Accurate measurements of global solar radiation, atmospheric temperature and relative humidity, as well as the availability of the predictions of their evolution over time, are important for different areas of applications, such as agriculture, renewable energy and energy management, or thermal comfort in buildings. For this reason, an intelligent, light-weight, self-powered and portable sensor was developed, using a nearest-neighbors (NEN) algorithm and artificial neural network (ANN) models as the time-series predictor mechanisms. The hardware and software design of the implemented prototype are described, as well as the forecasting performance related to the three atmospheric variables, using both approaches, over a prediction horizon of 48-steps-ahead.