基于LIBSVM和智能算法的电站锅炉飞灰含碳量优化

Optimizing the Exhaust Carbon Contented of Fly Ash of Power Station Boiler Based On LIBSVM and Artificial Intelligence Algorithm

飞灰含碳量是影响锅炉热效率的一个重要因素,影响燃煤锅炉飞灰含碳量的因素很多而且复杂。借助电站燃煤锅炉的实际运行数据,建立了基于改进的支持向量机算法的大型电站燃煤锅炉的飞灰含碳量模型,并利用现场数据对模型进行了训练和验证,结果表明模型具有很高的预测精度,可以应用于实际的工程预测。然后,结合全局寻优的遗传算法、粒子群算法和微分进化算法,以锅炉的运行调节参数为优化目标函数的自变量,对锅炉飞灰含碳量进行寻优,并获得了具体工况下的最佳操作参数。所得结果表明,LIBSVM与上述智能算法相结合在电站锅炉飞灰含碳量优化方面具有很高的应用价值。

Carbon content of fly ash is an important factor to impact boiler efficiency,there are many complex factors affecting the coal-fired boiler carbon content of fly ash. The model of large-scale power station coal-fired boiler carbon content of fly ash is based on improved Support vector machine algorithm and actual operation data of the power station coal-fired boiler. Using the local data train and check the model. The result show that the model has a very high prediction accuracy and it could be used to predict the practical engineering. And then combining with the global optimal Genetic algorithm、Particle swarm algorithm and Differential evolution algorithm,the power station coal-fired boiler carbon content of fly ash is optimized using the boiler operation adjustable parameters as the argument of optimization objective function,and get the best operating parameters in the specific conditions. Results show that the libsvm combining with the above-mentioned intelligent algo-rithms has a very highly application value in the power plant boiler carbon content of fly ash optimization.