【摘 要】
：
This paper introduces a practical optimization model of multiple cracking furnaces system with consideration of feedstock allocation.The goal of our work is to model this problem and to find the best
【机 构】
：
Key Laboratory of Advanced Control and Optimization for Chemical Processes,Ministry of Education
论文部分内容阅读
This paper introduces a practical optimization model of multiple cracking furnaces system with consideration of feedstock allocation.The goal of our work is to model this problem and to find the best allocation combination of feedstock to the cracking furnaces to maximize the profit of high-added-value production.Using artificial neural network(ANN)technique,the model can be formulated as a binary-integer nonlinear programming model.However,owing to the existence of integer variables and equality constraints,its hard to be solved by traditional evolutionary algorithms.Thus,an asynchronous updating differential algorithm(AUDE)is proposed in this paper to better handle discrete variables.Finally,to test the effectiveness and robustness of the proposed algorithm,simulation results on benchmark problems and practical application case shows that the proposed algorithm is robust and capable of getting satisfactory optimization results when compared with the state-of-the art evolution algorithm.
其他文献
Control of the fed-batch ethanol fermentation processes to produce maximum product ethanol is one of the key issues in the bioreactor system.However,ethanol fermentation processes exhibit complex beha
This paper investigates the problem of discrete-time(DT)model identification of industrial processes with time delay from sampled data.An iterative method is proposed to solve this problem,that is,the
This paper introduces a novel approach to learn occupants personalized thermal comfort profile of HVAC system based on the relationship between users thermal preference and ambient environmental condi
The task of learning the value function under a fixed policy in continuous Markov decision processes(MDPs)is considered.Although ELM based approximator can achieve better performance in learning predi
LED 由于其高效、节能、环保等优点，在照明、显示和路灯等领域得到了越来越广泛的应用和研究。然而，由于LED本身具有高热敏感性和非线性，给提高LED恒流控制系统的抗干扰能力带来了一定的困难。本文运用Stm32 微处理器，设计了一种LED鲁棒变增益恒流控制方案。该方案的控制器根据实时检测到的系统变参，可动态的调整状态反馈增益。本文提出的鲁棒变增益控制算法具有动态响应好，鲁棒性强等优点。基于Matla
The minimum squared error(MSE)method is a widely used linear discriminant function-based method.Nevertheless,MSE may not deal well with the uneven distributed data sets.To address this problem,we impr
Along with such dangerous sources as big fire,explosion and toxic matter leak in the chemical plants,the visual tracking technology is a simple yet effective solution.As an effective real-time visual
This paper aims at providing a mapping schema for mission and devicein distributed sensor network(DSN)domain from the semantic perspective,which is very helpful to machine-interpreting and autonomous
In order to characterize the transient dynamics of steam turbine subsections,this paper presents a complete approach to develop an application model and the process parameter estimation for a general
In this paper,a new modeling and monitoring approach in multi-mode processes based on subspace separation is proposed.