Application of incremental support vector regression based on optimal training subset and improved particle swarm optimi
- PDF / 3,078,267 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 92 Downloads / 173 Views
Application of incremental support vector regression based on optimal training subset and improved particle swarm optimization algorithm in real-time sensor fault diagnosis Dongdong Zhang 1 & Wenguo Xiang 1
&
Qiwei Cao 1 & Shiyi Chen 1
# Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Attracted by the advantages of support vector regression and incremental learning approach, it is proposed in this work that an incremental support vector regression (ISVR) model optimized by particle swarm optimization (PSO) algorithm, and some improvements are made to be more suitable for sensor faults on-line diagnosis. To reducethe training time of ISVR model, an optimal training subset (OTS) method is adopted to reduce the size of training data set of the model. Then, in order to solve the problem of slow convergence of standard PSO algorithm, an incremental PSO (IPSO) algorithm is proposed to accelerate the model convergence through adjusting the inertial weight of each particle, which is gained by comparing the current position of each particle and the optimal position of the last incremental training. Based on the above improvements, a hybrid model, IPSO-OTS-ISVR model is presented finally. Experimental results based on actual operational data of a gas turbine shows that, under the premise of ensuring accuracy, the proposed IPSO-OTS-ISVR has much better performance in model response time and convergence performance over the comparison models. The experimental results based on an UCI data set indicate that the proposed hybrid model can also be extended to solve other prediction problems. Keywords Sensor fault diagnosis . Support vector regression (SVR) . Particle swarm optimization (PSO) . Optimal training subset (OTS)
1 Introduction In order to monitor the actual running conditions of an equipment in real time, a large number of sensors are widely used in modern industrial gas turbine system. As the main devices to acquire information, the accuracy and reliability of the sensors directly affect the reliability, stability and safety of the entire system. Nevertheless, due to the needs of practical industrial applications, many sensors must operate in poor working environments of high temperature, high pressure, and high mechanical and thermal stress [1], which makes them more vulnerable to damage than other parts of the system. When the relevant monitoring, fault diagnosis and control system receives the error information Wenguo Xiang and Shiyi Chen are corresponding authors and both contributed equally * Wenguo Xiang [email protected] 1
Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education, School of Energy and Environment, Southeast University, Nanjing 210096, China
generated by the faulty sensor, it may cause false alarm, misdiagnosis, mistaken operation and even causing incalculable loss [2, 3]. Therefore, it is important to study how to identify and diagnose the sensor fault. A common method for sensor fault diagnosis is to generate the residual through redunda
Data Loading...
