Konrad Wojdan

Immune inspired optimizer of combustion process in power boiler

The article presents an optimization method of combustion process in a power boiler. This solution is based on the artificial immune systems theory. A layered optimization system is used, to minimize CO and NOx emission. Immune inspired optimizer SILO is implemented in each of three units of Ostroleka Power Plant (Poland). The results from this implementation are presented. They confirm that presented solution is effective and usable in practice.

Stochastic immune layer optimizer: efficient tool for optimization of combustion process in a boiler

The article presents a method for optimization of combustion process in a boiler. This solution is based on the artificial immune system. A layered optimization system is used, to minimize CO and NOx emission. This solution is implemented in real power plant. The results from this implementation have been presented. They confirm that presented solution is effective and usable in practice.

New Improvements of Immune Inspired Optimizer SILO

The article presents new improvements of an immune inspired optimization method, used to control a combustion process in a steam generating, coal fired, large scale boiler. Immune Inspired Optimizer SILO is implemented at each of three units of Ostroleka Power Plant (Poland) and at one unit in Newton Power Plant (USA). The results from Newton Power Plant are presented. They confirm that presented solution is effective and usable in practice and it can be treated as a good alternative to MFC controllers. The main goal of this solution is CO and NOx emission minimization.

Immune Inspired System for Chemical Process Optimization using the example of a Combustion Process in a Power Boiler

The article presents an optimization method of combustion process in a power boiler. Immune inspired optimizer SILO is used to minimize CO and NOx emission. This solution is implemented in each of three units of Ostroleka Power Plant (Poland) and in the Newton Power Plant (USA). The result from the second SILO implementation in Newton Power Plant is presented. The results confirm that this solution is effective and usable in practice and it can be a good alternative to MPC controllers.

Transition state layer in the immune inspired optimizer

The SILO (Stochastic Immune Layer Optimizer) system is a novel, immune inspired solution for an on-line optimization of a largescale industrial processes. Three layers of optimization algorithm were presented in previous papers. Each layer represents a different strategy of steady state optimization of the process. New layer of the optimization algorithm is presented in this paper. The new Transition State layer is responsible for efficient operation of the optimization system during essential process state transitions. New results from SILO implementation in South Korean power plant are presented. They confirm high efficiency of the SILO optimizer in solving technical problems.

Approximation of HRPITS results for SI GaAs by large scale support vector machine algorithms

For the first time large-scale support vector machine algorithms are used to extraction defect parameters in semi-insulating (SI) GaAs from high resolution photoinduced transient spectroscopy experiment. By smart decomposition of the data set the SVNTorch algorithm enabled to obtain good approximation of analyzed correlation surface by a parsimonious model (with small number of support vector). The extracted parameters of deep level defect centers from SVM approximation are of good quality as compared to the reference data.

The method for optimisation of gas compressors performance in gas storage systems

We present a computational method for achieving an optimal operation of compressor units used in industrial gas storage systems. The proposed method is capable to operate with a mix of compressor types (i.e., with different operational parameters, different power drives, and different types of construction, e.g., reciprocal and turbocompressors). The goal of the optimisation is to find an optimal compressors configuration and distribution of the compressor loads for each instance of time. The proposed method is based on conversion of a multidimensional discrete-continuous optimisation problem into a set of independent combinatorial and nonlinear optimisation problems. We derive the mathematical foundations of the algorithms. The exemplary results of the application are presented. [Received: May 4, 2016; Accepted: December 2, 2016]

Conditioning of Model Identification Task in Immune Inspired Optimizer SILO

Methods which provide good conditioning of model identification task in immune inspired, steady‐state controller SILO (Stochastic Immune Layer Optimizer) are presented in this paper. These methods are implemented in a model based optimization algorithm. The first method uses a safe model to assure that gains of the process’s model can be estimated. The second method is responsible for elimination of potential linear dependences between columns of observation matrix. Moreover new results from one of SILO implementation in polish power plant are presented. They confirm high efficiency of the presented solution in solving technical problems.