Prediction Of Free Calcium Oxide In Cement Clinker Based On Integrated Optimization Algorithm

Calcination of cement clinker is an important process in the cement production process,and the operation of the process will directly affect the final quality of the cement.Free lime in cement clinker(fCaO)is mainly produced in this process.The level of this indicator directly reflects the cement stability and energy consumption of cement production.The traditional cement clinker fCaO content detection is poor in real-time,and it is difficult to guide production in real time.Therefore,through in-depth study of the heat transfer mechanism,group intelligence algorithm,evolutionary algorithm and machine learning algorithm inside the cement rotary kiln firing process,the cement clinker fCaO content prediction model is established to improve the quality of cement clinker and promote energy saving and emission reduction in the cement industry.And it plays an important role in improving cement production efficiency.The specific research work of this paper is described as follows:Firstly,in view of the current multi-phase flow model,it is difficult to trace the pulverized coal particles,the pulverized coal combustion reaction cannot be accurately simulated,and the calculation is difficult to converge.A method for modeling the material watershed and the high-temperature smoke airflow in the rotary kiln is proposed.Then use the interface between the two basins for heat transfer calculation,which simplifies the heat transfer process in the rotary kiln.In order to make the heat transfer process of the simplified model not affected by the simplification,this paper establishes the heat transfer compensation model at the interface according to the principle of heat transfer,and establishes the pulverized coal in the rotary kiln according to the internal firing environment of the rotary kiln.Combustion model,radiation model and turbulence model.The simulation experiment of the cement rotary kiln material firing system was carried out by using the Computational Fluid Dynamics(CFD)simulation software Fluent,and the influence of different rotary kiln conditions on the material temperature was analyzed.Secondly,aiming at the problems of differential evolution(DE)algorithm,this paper proposes an adaptive composite differential evolution(SCDE)algorithm that can automatically select experimental vector generation strategies based on optimization problems.In order to integrate the advantages of various algorithms,this paper proposes a dynamic following strategy,and based on this strategy,proposes a new and efficient dynamic optimization strategy based on dynamic following strategy Self adaptive composite DE-Dynamically following PSO(SCDE-DFPSO).The SCDE algorithm and the Particle Swarm Optimizer(PSO)algorithm are integrated.Then the performance of the proposed SCDE-DFPSO algorithm is evaluated using the CEC2013 parameter optimization benchmark problems and compare with DE-DFPSO and other state-of-the-art optimization.The results show that the proposed algorithm is superior to the comparison algorithm in both the convergence performance and the convergence accuracy of the 28 test functions of CEC2013.Finally,the results of simulated experimental data of cement rotary kiln materials is used to establish temperature prediction model for the rotary kiln firing zone.Then based on the actual production situation of the rotary kiln and the temperature prediction result of the firing zone,prediction model of cement clinker fCaO based on the multi-core least squares support vector machine(MKLSSVM)is established,and the proposed SCDEDFPSO is utilized.The DFPSO algorithm optimizes the parameters of the prediction model,and based on the actual production data,the simulation model of the cCao content prediction model is carried out to verify the validity and robustness of the proposed prediction model.