Mechanism Analysis and Self-Adaptive RBFNN Based Hybrid Soft Sensor Model in Energy Production Process: A Case Study

Abstract

Despite hard sensors can be easily used in various condition monitoring of energy production process, soft sensors are confined to some specific scenarios due to difficulty installation requirements and complex work conditions. However, industrial process may refer to complex control and operation, the extraction of relevant information from abundant sensors data may be challenging, and description of complicated process data patterns is also becoming a hot topic in soft-sensor development. In this paper, a hybrid soft sensor model based mechanism analysis and data-driven is proposed, and ventilation sensing of coal mill in a power plant is conducted as a case study. Firstly, mechanism model of ventilation is established via mass and energy conservation law, and object-relevant features are identified as the inputs of data-driven method. Secondly, radial basis function neural network (RBFNN) is used for soft sensor modeling, and genetic algorithm (GA) is adopted for quick and accurate determination of the RBFNN hyper-parameters, thus self-adaptive RBFNN (SA-RBFNN) is proposed to improve the soft sensor performance in energy production process. Finally, effectiveness of the proposed method is verified on a real-world power plant dataset, taking coal mill ventilation soft sensing as a case study.

An Algorithm for Soft Sensor Development for a Class of Processes with Distinct Operating Conditions

“…Initial widths of radial functions are set according to the spatial distribution of determined clusters. For all cluster centers, the same initial value is used, calculated with the expression [ 11 , 27 ], which ensures that initial individual radial functions are not too peaked or too flat.…”

“…There are different approaches to soft sensor development, but the complexity of the production process and uncertainties in determining the connection between laboratory values and signals that are measured in the process are reasons why soft sensors in the process industry are mainly based on black box or gray box models. The black box approach is successfully applied to different processes, from the cement industry [ 1 , 2 , 3 , 4 ] and chemical processes [ 5 , 6 , 7 ] to water treatment [ 8 , 9 , 10 ], energy production [ 11 ] and the oil industry [ 12 , 13 , 14 ]. Since it proved appropriate for the development of soft sensors for a variety of industrial processes, the black box approach has the potential to be used as a basis for a wide industrial implementation of soft sensors.…”

“…Different model types have been used for the development of black box models for soft sensors in industrial processes, from models based on different types of Neural Networks [ 1 , 3 , 6 , 8 , 9 , 10 , 11 , 14 ], which represent the most common class of models, to Support Vector Regression [ 2 , 9 ], Neuro Fuzzy [ 13 ], Fuzzy Modeling [ 15 ], Linear Regression [ 16 ], and various other types of models. Determining the model type that is suitable for developing soft sensors for all industrial processes is, of course, not possible.…”

Separable Gaussian Neural Networks: Structure, Analysis, and Function Approximations

“…Assume the size of the input dataset is m. Using Equations ( 5) to (8), we can estimate the number of floating-point operations (FLOPs) of the forward pass in SGNNs. More specifically, the number of FLOPs to calculate the output of the k-th layer with the input from the previous layer is…”

“…Radial-basis functions have many important applications in the fields such as function interpolation [1], meshless methods [2], clustering classification [3], surrogate models [4], Autoencoder [5], dynamic system design [6], network event detection [7], and modeling in energy production processes [8], to name a few. The Gaussian-radial-basis function neural network (GRBFNN) is a neural network with one hidden layer and produces output in the form…”