NOx Reduction Based on an Improved Orthogonal Particle Swarm Optimization

Qingwei Li

Abstract


The optimization of operation parameters to cut NOx emissions was studied. First, the orthogonal experimental design(OED) study strategy based on the optimum particle and suboptimum particle was introduced to improve the performance of particle swarm optimization(IPSO). To overcome the premature problem and save the computer resources at the same time, linear increasing probability for OED was proposed. Next, a forecasting model for NOx emissions was established based on an improved ensemble support vector machine whose parameters were optimized by IPSO. Afterwards, the operation parameters were optimized by IPSO to cut NOx emissions. Finally, the new method was applied to a power plant. Simulation results show that IPSOSVM can predict NOx emissions more accurately than the existing methods. IPSO can find desirable parameters within a reasonable time in the parameter selection process. NOx emission optimized by IPSO are the lowest among the state-of-the-art PSOs and other intelligent algorithms.


Full Text:

PDF

Refbacks

  • There are currently no refbacks.