面向分段计划负荷的耦合配烧优化模型与硫分约束界调整OA

In response to the issues of low efficiency and high cost in manual coal co-firing scheme formulation for thermal power plants,this paper systematically conducts a coupled blending optimization model and sulfur con-straint boundary adjustment research for segmented planned loads for the first time.Firstly,in order to ensure faster and more accurate formulation of the blending scheme,a coupling relationship between segmented planned loads and blending optimization model is established through the switching of grinding vectors of the coal mill unit,achieving digital co-firing mainly based on computer optimization models.Then,to address the problem that the sulfur constraint in the coupled coal co-firing optimization model is subject to multiple uncertainties(e.g.,coal sul-fur conversion and desulfurization),a sulfur constraint boundary feedforward compensation model based on an im-proved stochastic configuration network is established on the basis of inverse calculation of the sulfur constraint boundary.Next,in order to ensure the optimality of the historical training sample case database for the feedfor-ward compensation model,the case database is iteratively and dynamically updated throughout the entire cycle by monitoring the real-time production data of units,thereby improving the accuracy of the feedforward compensation model＇s full cycle prediction.Application results show the developed system cuts co-firing scheme formulation time from 40 to under 5 minutes per instance.Moreover,while meeting environmental standards,it lowers the cost by an average of 21 CNY/t compared to manual operations,yielding significant efficiency and economic gains.