Yongheng Jiang

A New Strategy of Integrated Control and On-line Optimization on High-purity Distillation Process

Abstract For high-purity distillation processes, it is difficult to achieve a good direct product quality control using traditional proportional-integral-differential (PID) control or multivariable predictive control technique due to some difficulties, such as long response time, many un-measurable disturbances, and the reliability and precision issues of product quality soft-sensors. In this paper, based on the first principle analysis and dynamic simulation of a distillation process, a new predictive control scheme is proposed by using the split ratio of distillate flow rate to that of bottoms as an essential controlled variable. Correspondingly, a new strategy with integrated control and on-line optimization is developed, which consists of model predictive control of the split ratio, surrogate model based on radial basis function neural network for optimization, and modified differential evolution optimization algorithm. With the strategy, the process achieves its steady state quickly, so more profit can be obtained. The proposed strategy has been successfully applied to a gas separation plant for more than three years, which shows that the strategy is feasible and effective.

A Novel Scheduling Strategy for Crude Oil Blending

For those refineries which have to deal with different types of crude oil, blending is an attractive solution to obtain a quality feedstock. In this paper, a novel scheduling strategy is proposed for a practical crude oil blending process. The objective is to keep the property of feedstock, mainly described by the true boiling point (TBP) data, consistent and suitable. Firstly, the mathematical model is established. Then, a heuristically initialized hybrid iterative (HIHI) algorithm based on a two-level optimization structure, in which tabu search (TS) and differential evolution (DE) are used for upper-level and lower-level optimization, respectively, is proposed to get the model solution. Finally, the effectiveness and efficiency of the scheduling strategy is validated via real data from a certain refinery.

Improved Hybrid Differential Evolution-Estimation of Distribution Algorithm with Feasibility Rules for NLP/MINLP Engineering Optimization Problems

Abstract In this paper, an improved hybrid differential evolution-estimation of distribution algorithm (IHDE-EDA) is proposed for nonlinear programming (NLP) and mixed integer nonlinear programming (MINLP) models in engineering optimization fields. In order to improve the global searching ability and convergence speed, IHDE-EDA takes full advantage of differential information and global statistical information extracted respectively from differential evolution algorithm and annealing mechanism-embedded estimation of distribution algorithm. Moreover, the feasibility rules are used to handle constraints, which do not require additional parameters and can guide the population to the feasible region quickly. The effectiveness of hybridization mechanism of IHDE-EDA is first discussed, and then simulation and comparison based on three benchmark problems demonstrate the efficiency, accuracy and robustness of IHDE-EDA. Finally, optimization on an industrial-size scheduling of two-pipeline crude oil blending problem shows the practical applicability of IHDE-EDA.

Bottleneck analysis for network flow model

It is very important to develop effective strategies for process industry to implement feasible scheduling while the process bottlenecks work optimally. However, the bottlenecks adopted by the existed scheduling strategies are often partial bottlenecks, local bottlenecks, or even deceptive bottlenecks, which are basically not key constraints to achieve optimal objective, so that the corresponding scheduling strategies are not perfect, and the further improvement and evaluation can hardly be proposed. So, it is a valuable issue to analyze the bottlenecks both in academic and engineering fields. This paper aims at the minimum cost problem of generalized network flow model to define and analyse three classes of bottlenecks based on generalized network simplex algorithm, and the corresponding search algorithms are proposed in this paper. The obtained bottlenecks cannot only be used to determine whether the object cost will be increased or decreased, but also be used to propose the corresponding strategy to evaluate the improvement on the network flow model for scheduling. Finally, a typical example is discussed.

An Extended AUDI Algorithm for Simultaneous Identification of Forward and Backward Paths in Closed-Loop Systems

Abstract In closed-loop system identification, most of the existing methods only focus on the forward path, yet few on simultaneous identification of the forward and backward paths. Meanwhile, an augmented UD identification (AUDI) algorithm has been proved effective in open-loop system identification, but it only extracts the forward path information, while not including the backward path information provided in an augmented information matrix, which is helpful for the closed-loop system identification. In this paper, an extended AUDI (EAUDI) is proposed to simultaneously identify the model orders and parameters of both forward and backward paths of a closed-loop system. The conditions of identifiability and uniform convergence for closed-loop systems using the EAUDI algorithm are also given. The effectiveness of this algorithm is demonstrated by a numerical example.

Probabilistic Trajectory Prediction in Intelligent Driving

For intelligent vehicles, the prediction of trajectories of surrounding traffic participants is the basis for the ego trajectory planning processes. The sole deterministic results provided by traditional methods are insufficient because of the uncertainties from the real world. Therefore, probabilistic trajectory prediction based on real traffic data is proposed. The prediction model in this paper can provide probability distributions of future positions of the concerned objects, which describe the possibilities of different trajectory executions. To improve the prediction results, the interaction between vehicles is also considered. The model is built based on actual trajectory data offline so that only iterative computations are needed when it is applied online. In the end of this paper, the performance of the proposed method is demonstrated and verified on real data sets. The simulation show that the obtained prediction results accord with the actual stochastic position distribution.

Optimization on distillation via piecewise linear approximation

Distillation is very important in process industry, and the optimization aiming to energy saving and yield improvement is still important and difficult. Piecewise linear approximation is firstly introduced into traditional Chemical Engineering field. And therefore, a novel surrogate optimization strategy based on piecewise linear approximation for distillation, is developed in this paper. The nonlinear character has been approximated by a surrogate model based on continuous piecewise linear function. Then the original nonlinear optimization has been transformed into piecewise linear programming (PLP). The PLP problem is solved to get the new operating point for the control scheme. Some properties of PLP are discussed and a condition of local optimum is given. An algorithm for PLP is also proposed, with a distillation process model based on adaptive hinging hyperplane. The statistics show that the new strategy is feasible and effective.

Identification of FIR models using basis models of first-order plus time delays

In this article, a new methodology is presented for identification of finite-impulse-response (FIR) models. The central idea is to use a series of stable models of first-order plus time delays (FOPTD) to approximate the dynamics of the process. A quantitative analysis of the modelling error is provided. Compared to the existing basis model approaches, the advantages of the proposed method lie in the following aspects: (i) a simple model structure with a small number of basis models is sufficient to obtain satisfactory approximations, thereby reducing the model structural risks; and (ii) basis models could be determined with limited prior process information, which is beneficial for practical implementations. Representative simulation examples are provided to illustrate the superiority of our method over the existing identification approaches based on basis models.

Trajectory Planning for Automated Driving Based on Ordinal Optimization

This paper proposes an approach based on Ordinal Optimization (OO) to solve trajectory planning for automated driving. As most planning approaches based on candidate curves optimize the trajectory curve and the velocity profile separately, this paper formulates the problem as an unified Non-Linear Programming (NLP) model, optimizing the trajectory curve and the acceleration profile (acceleration is the derivative of velocity) simultaneously. Then a hybrid optimization algorithm named OODE, developed by combining the idea of OO and Differential Evolution (DE), is proposed to solve the NLP model. With the acceleration profile optimized “roughly”, OODE computes and compares “rough” (biased but computationally-easier) curve evaluations to select the best curve from candidates, so that a good enough curve can be obtained very efficiently. Then the acceleration profile is optimized again “accurately” with the selected curve. Simulation results show that good enough solutions are ensured with a high probability and our method is capable of working in real time.

Intelligent computing budget allocation for on-road trajectory planning based on candidate curves

In this paper, on-road trajectory planning is solved by introducing intelligent computing budget allocation (ICBA) into a candidate-curve-based planning algorithm, namely, ordinal-optimization-based differential evolution (OODE). The proposed algorithm is named IOODE with ‘I’ representing ICBA. OODE plans the trajectory in two parts: trajectory curve and acceleration profile. The best trajectory curve is picked from a set of candidate curves, where each curve is evaluated by solving a subproblem with the differential evolution (DE) algorithm. The more iterations DE performs, the more accurate the evaluation will become. Thus, we intelligently allocate the iterations to individual curves so as to reduce the total number of iterations performed. Meanwhile, the selected best curve is ensured to be one of the truly top curves with a high enough probability. Simulation results show that IOODE is 20% faster than OODE while maintaining the same performance in terms of solution quality. The computing budget allocation framework presented in this paper can also be used to enhance the efficiency of other candidate-curve-based planning methods.