Co-learning with a locally weighted partial least squares for soft sensors of nonlinear processes
نویسندگان
چکیده
A method to improve adaptivity of soft sensors is investigated in this paper. Soft sensors have become very important in the chemical industry to achieve a highly efficient, high-quality and safe production system. Among the various methods, partial least squares (PLS) method is the most used for soft sensors. In this research, a co-learning style locally weighted PLS method which utilizes a semi-supervised regression is proposed to estimate a process value. The method is applied to a simulated reactor process, and the results clearly show an improvement in the estimation accuracy compare with the conventional method.
منابع مشابه
Imitation-based Learning of Bipedal Walking Using Locally Weighted Learning
Walking is an extremely challenging problem due to its dynamically unstable nature. It is further complicated by the high dimensional continuous state and action spaces. We use locally weighted projection regression (LWPR) as a locally structurally adaptive nonlinear function approximator as the basis for learned control policies. Empirical evidence suggests that control policies for high dimen...
متن کامل5 Approximate Nearest Neighbor Regression in Very High Dimensions
Fast and approximate nearest-neighbor search methods have recently become popular for scaling nonparameteric regression to more complex and high-dimensional applications. As an alternative to fast nearest neighbor search, training data can also be incorporated online into appropriate sufficient statistics and adaptive data structures, such that approximate nearestneighbor predictions can be acc...
متن کاملDetermination of Stability Domains for Nonlinear Dynamical Systems Using the Weighted Residuals Method
Finding a suitable estimation of stability domain around stable equilibrium points is an important issue in the study of nonlinear dynamical systems. This paper intends to apply a set of analytical-numerical methods to estimate the region of attraction for autonomous nonlinear systems. In mechanical and structural engineering, autonomous systems could be found in large deformation problems or c...
متن کاملLocally Weighted Least Squares Temporal Difference Learning
This paper introduces locally weighted temporal difference learning for evaluation of a class of policies whose value function is nonlinear in the state. Least squares temporal difference learning is used for training local models according to a distance metric in state-space. Empirical evaluations are reported demonstrating learning performance on a number of strongly non-linear value function...
متن کاملEvaluation of spectral pretreatments, partial least squares, least squares support vector machines and locally weighted regression for quantitative spectroscopic analysis of soils
ISSn: 0967-0335 © IM publications llp 2010 doi: 10.1255/jnirs.883 all rights reserved the measurement of physical and chemical parameters of soil is an important step toward sustainable farming practices, landscaping management and, more generally, the understanding of terrestrial ecosystem processes. Standard soil analytical procedures are often complex, time-consuming, and expensive for many ...
متن کامل