Flowsheet generation through hierarchical reinforcement learning and graph neural networks

Process synthesis experiences a disruptive transformation accelerated by artificial intelligence. We propose a reinforcement learning algorithm for chemical process design based on a state-of-the-art actor-critic logic. Our proposed algorithm represents chemical processes as graphs and uses graph convolutional neural networks to learn from process graphs. In particular, the graph neural networks are implemented within the agent architecture to process the states and make decisions. We implement a hierarchical and hybrid decision-making process to generate flowsheets, where unit operations are placed iteratively as discrete decisions and corresponding design variables are selected as continuous decisions. We demonstrate the potential of our method to design economically viable flowsheets in an illustrative case study comprising equilibrium reactions, azeotropic separation, and recycles. The results show quick learning in discrete, continuous, and hybrid action spaces. The method is predestined to include large action-state spaces and an interface to process simulators in future research.     

枝状管网布局优化中的3类问题

总结了枝状管网布局优化研究中普遍存在的3类问题,包括无向图枝状结构布置、枝状结构中心点的确定和有向图的枝状连接结构布置,分别论述了3类问题的数学模型和求解算法。指出无向图枝状管网布局优化和枝状结构中心点的确定,其求解算法比较稳定,是集输系统枝状结构布局普遍采用的方式,SI算法加入额外点,优化结果优于其他算法。有向图的枝状结构布局,由于问题的复制性,目前无较成熟和被广泛认可的求解方法,此问题的研究更能贴近实际,能达到更好的优化效果。     

A Graph-based Ant Colony Optimization Approach for Integrated Process Planning and Scheduling

This paper considers an ant colony optimization algorithm based on AND/OR graph for integrated process planning and scheduling (IPPS). Generally, the process planning and scheduling are studied separately. Due to the complexity of manufacturing system, IPPS combining both process planning and scheduling can depict the real situation of a manufacturing system. The IPPS is represented on AND/OR graph consisting of nodes, and undirected and directed arcs. The nodes denote operations of jobs, and undirected/directed arcs denote possible visiting path among the nodes. Ant colony goes through the necessary nodes on the graph fromthe starting node to the end node to obtain the optimal solution with the objective of minimizing makespan. In order to avoid local convergence and lowconvergence, some improved strategy is incorporated in the standard ant colony optimization algorithm. Extensive computational experiments are carried out to study the influence of various parameters on the system performance.     

A graphical approach to cause and effect analysis of chemical processing systems

A method for cause and effect analysis of chemical processing systems has been developed on the concept of a signed directed graph. The signed directed graph, called a “model graph,” is used for representing cause and effect relationships among process variables in a system, and a “Pattern” on the graph is introduced to represent a state of the system, which is specified by “0” for normal state, “ + ” and “ ? ” for larger and smaller deviation from the normal state, respectively. For a given model graph and pattern on it, the cause of state changes and the manner of their propagation are easily found out by the present approach, which is mainly based on the depth-first search algorithm developed by Tarjan. As an illustrative example, a propane vaporizer has been taken to show the usefulness of the present approach.     

复杂非线性系统的子系统分解方法

利用社区发现算法研究了一种复杂非线性化工系统的子系统分解方法,并进行了分布式模型预测控制设计。引入信息图论的节点表示系统的状态、输入和输出变量,构建非线性过程系统的加权有向图,节点通过加权边连接,加权反映了节点间连接的强度,因而能够同时反映系统内部的连通性和连接强度。利用社区结构发现算法将所有变量分成子系统的群组,使得每个组内的关联比不同组间的相互作用强,从而得到复杂化工过程系统的子系统分解。针对连续搅拌反应釜过程,实施子系统分解,并设计分布式模型预测控制算法,结果表明,所提出的子系统分解方法更能考虑子系统之间的连接权重,得到更有利于分布式模型预测控制的子系统划分,提升系统控制的性能。     

用于化工过程的SDG故障分析方法

基于符号有向图（signed directed graph,SDG）的化工过程模型可以表征过程故障与故障源之间的因果关系,SDG故障分析方法通过有向图定性推理技术实现故障诊断。本文简要介绍了SDG故障分析的基本概念,并对Iri算法^[1]中故障源的搜索步骤提出了部分改进,并进行了实例验证。最后,利用SDG故障诊断方法对某化工厂的氯气泄漏事故的故障传播进行了分析研究。     

再制造视域下的注塑模具可拆卸性设计综合评价算法

根据注塑模具再制造要求,构建了模具可拆卸性设计评价指标体系,并对各单一指标进行了定量描述。研究了模具零部件可修复性评价的工作过程和各个评价指标之间的关联度和相互作用状态,采用指标关联度矩阵和数学图论中的有向图,构建了基于积和式的模具可拆卸性设计评价算法模型。以一个注塑模具设计为实例,检验了评价算法的正确性。     

基于FTP-图的时间序列研究及在设备监控中应用

采用FTP-图(频繁时序图)的时间序列方法对流程企业中的实际运行数据进行数据挖掘,首先给出基于拓扑关系的时间序列的定义,分析FTP-图连接、候选TP-图剪枝和采用哈希树的候选TP-图计数方法,然后采用模糊理论找出偏离常规运行状态的参数点并模糊化,采用时间窗对参数离散处理,得到时间序列数据库.根据Apriori算法原理,提出基于FTP-图的时间序列分析算法并编程实现,起到对设备故障预警和生产优化的决策作用.     

网络探索改进算法及其实现方法

研究了网络拓扑分析 ,在 Johnson回路探索算法的基础上提出了其改进算法 ,并介绍了其 C 程序实现方法 ,为图形组模仿真法提供了关键技术     

Modeling,simulation and structural analysis of a fluid catalytic cracking (FCC) process

Fluid catalytic cracking (FCC) is an important chemical process that is widely used to produce valuable petrochemical products by cracking heavier components. However, many difficulties exist in modeling the FCC process due to its complexity. In this study, a dynamic process model of a FCC process is suggested and its structural observability is analyzed. In the process modeling, yield function for the kinetic model of the riser reactor was applied to explain the product distribution. Hydrodynamics, mass balance and energy balance equations of the riser reactor and the regenerator were used to complete the modeling. The process model was tested in steady-state simulation and dynamic simulation, which gives dynamic responses to the change of process variables. The result was compared with the measured data from operating plaint. In the structural analysis, the system was analyzed using the process model and the process design to identify the structural observability of the system. The reactor and regenerator unit in the system were divided into six nodes based on their functions and modeling relationship equations were built based on nodes and edges of the directed graph of the system. Output-set assignment algorithm was demonstrated on the occurrence matrix to find observable nodes and variables. Optimal locations for minimal addition of measurements could be found by completing the whole output-set assignment algorithm of the system. The result of this study can help predict the state more accurately and improve observability of a complex chemical process with minimal cost.     

TEARING AND PRECEDENCE ORDERING IN FLOWSHEETING-AN EFFECTIVE METHOD

Network decomposition has been proven to be one of a large set of problems termed “NP Complete” (nondeterministic polynomial) (Gundersen and Hertzberg, 1982). NP complete problems have no known method of solution better than trying all the possibilities, nor, according to most mathematicians, is any superior method likely to be found (Wasserman, 1989). Because such an exhaustive search is generally impractical for large networks, heuristic methods have been applied to find solutions that are acceptable if not optimal. A new heuristic strategy for sequencing of calculation of the unit modules in sequential modular flowsheeting and determining the tear streams is presented. The method neither involves the use of the signal flow graph nor information on cycles in the network. The algorithm has been tested on several problems collected from chemical engineering literature and includes the very complex graph of a heavy water plant suggested as a benchmark problem by Gundersen and Hertzberg (1982). The proposed algorithm finds the optimal number of tears for twelve of the test problems, while the number of tears found for the remaining two problems is only one more than the optimum value. For large problems, which are not easily handled by optimal tearing routines, this method can serve as an upper bound algorithm in a branch and bound strategy.     

Excel在管路阻力计算中的应用

介绍了采用Excel工作表简单迭代求解柯尔布鲁克公式的计算过程,在计算方法的选择上,采用埃特金算法比直接求解法可以更有效地加快迭代计算收敛速度。结果表明,用快速收敛求法代替莫狄摩擦因数图求取摩擦因数,避免了读图误差,结果准确,计算方便、快捷。     

Sensor network design for contaminant detection and identification in water distribution networks

Water distribution networks (WDN) are vulnerable to either intentional or accidental contamination. In order to protect against such intrusions, effective and efficient online monitoring systems are needed. Due to cost and maintenance reasons, it is not possible to locate sensors at each and every potential intrusion point. In this work, we design minimal sensor networks which satisfy the two important properties of observability (ability to detect an intrusion) and identifiability (ability to identify the point of intrusion). Based on the hydraulic analysis of the network, a bipartite graph is constructed between intrusion points and the corresponding nodes that can potentially be affected by the contaminant. The problem of sensor network design is converted to a minimum set cover problem on the bipartite graph, and is solved using a greedy heuristic algorithm. The proposed method is illustrated using a medium scale urban WDN.     

Inversion of sparse matrices by a method based on graph theory

Most of the published methods for the calculation of the inverse of matrix graph theory methods take advantage of the structure of the graph to reorder the matrix, for easier inversion.

This study presents a method for the inversion of a large sparse matrix by means of a signal flow graph, that is analogous to a process flowsheet, using the 1956 Mason formula [1], widely used in electrical engineering to calculate the gain of a complex electronic network.

A new data structure, called path matrix, was developed to contain the full information about the one edge paths in the graph. Information about greater edge paths is added. All the cycles of the graph are identified and their signal values are calculated. The sets of nontouching cycles are identified and the calculation formula is applied. Sparse matrices with dimensions of 10–100 were tested. The program inverted all test matrices, even those that could not be inverted by the IMSL routine LINV2F, up to the limit of the expandability of the path matrix. However, due to the complexity of the algorithm, CPU times for inversion of large matrices are unacceptable, except for ill-conditioned matrices that other methods fail to invert.

This inverse matrix of a linear process flowsheet can be calculated directly from the graph describing the process.     

污水管道系统的遗传优化设计

以城市排水系统的设计理论与设计规范为依据,建立了污水管道优化设计模型,并应用改进的遗传算法寻求管网投资最小的设计方案。该方法以管径和设计充满度为决策变量,采用二维混合编码方案进行方案表达,可直接应用水力计算基本公式进行管路计算,能有效地提高遗传算法的进化过程。研究表明,应用该方法进行污水管道系统优化设计可节约6.14%～19.74%的工程费用,在工程优化设计中有一定的实用价值。     

A new algorithm for computing distance matrix and Wiener index of zig-zag polyhex nanotubes

The Wiener index of a graph G is defined as the sum of all distances between distinct vertices of G. In this paper an algorithm for constructing distance matrix of a zig-zag polyhex nanotube is introduced. As a consequence, the Wiener index of this nanotube is computed.     

statnet: Software Tools for the Representation, Visualization, Analysis and Simulation of Network Data

statnet is a suite of software packages for statistical network analysis. The packages implement recent advances in network modeling based on exponential-family random graph models (ERGM). The components of the package provide a comprehensive framework for ERGM-based network modeling, including tools for model estimation, model evaluation, model-based network simulation, and network visualization. This broad functionality is powered by a central Markov chain Monte Carlo (MCMC) algorithm. The coding is optimized for speed and robustness.     

顺序功能图图形化组态软件的设计开发

概述SFC的功能、组成和编程规则。介绍所开发的SFC组态软件的图形系统的构成。详尽给出了SFC图形系统的设计方案及其基于VC 的方案实现。实践表明,依据这一方案设计实现的SFC组态软件组态方式灵活、易于维护。     

An algorithm for diagnosis of system failures in the chemical process

An attempt was made to apply graph theory to the diagnosis of the system failures in the chemical process. A signed digraph is used for a mathematical model representing the influences among elements of the system. The concept of a pattern on the signed digraph is introduced for representing a state of the system. In order to eliminate carrying out the complicated and inefficient quantitative simulation, the mathematical model of the system structure to represent the rpopagation of failures is simplified in a qualitative fashion. The origin of the system failure can be located at the maximal strongly-connected component in the cause-effect graph reflecting the pattern of abnormality. Even when the pattern is observed only partially, the assumption of single origin of the failure reduces, to some extent, the range of possible candidates to be the first cause of the failure.     

化工过程危险剧情分类及SDG定性识别方法

系统深入地揭示危险“剧情”是分析和解决化工过程安全问题的基础和核心内容,也是安全评价和故障诊断的核心内容。本文对危险剧情进行了定义,并在将危险剧情分为5类的基础上,采用符号定向图（SDG）定性识别方法,提出了连续系统的智能化自解释报警、单故障和多故障根原因诊断的算法步骤。研究成果对过程系统的危险与可操作性分析（HAZOP）、安全防护层分析（LOPA）、自解释报警和故障诊断系统在过程工业领域的开发具有重要的理论与实际意义。