Neural model for forecasting temperature in a distribution network of cooling water supplied to systems producing petroleum products

This article presents the application of a neural model of heat transfer for the purpose of forecasting temperature at selected points of a circulating water ring network. The purpose of a circulating water system is to lower the temperature of petroleum products manufactured on numerous petrochemical lines at a Polish petrochemical plant. Temperature forecasting at 96 nodes of the circulating water system, significant from the point of view of system operation, is carried out using SVM neural networks. Neural networks learn based on archival data recorded in the process parameter monitoring system. Thermal, hydraulic and control parameters of the cooling process, as well as weather variables, constitute crucial input data for the neural model. The temperature forecasting algorithm has been implemented in a computer program that was then applied and remains in use for temperature forecasting in a maintenance department of an industrial plant.     

A contaminant detection system for early warning in water distribution networks

Water distribution systems are spatially diverse. As such, they are inherently vulnerable to accidental or deliberate physical, chemical, or biological threats. Efficient water quality monitoring is one of the most important tools to guarantee a reliable potable water supply. A methodology and two example applications for finding the optimal layout of a detection system, taking explicitly into account the dilution and decay properties of the water quality constituents as distributed with flow, as well as the ability of the monitoring equipment to detect contaminant concentrations, are formulated and demonstrated. The detection system outcome is aimed at capturing contaminant entries within a pre-specified level of service (LOS) defined as the maximum volume of polluted water exposure to the public at a concentration higher than a minimum hazard level. The proposed methodology couples hydraulic simulations with graph theory techniques to identify a minimum set of monitoring stations that ‘covers’ the entire network for a given LOS, at a maximum degree of system invulnerability. The model developed extends a previous work of the authors through explicitly considering the deterioration and dilution of water quality as distributed with flow, and by taking into account the monitoring equipment capabilities to detect pollutant concentrations. The methodology is demonstrated using two example applications.     

Self-adaptive multi-objective harmony search for optimal design of water distribution networks

In multi-objective optimization computing, it is important to assign suitable parameters to each optimization problem to obtain better solutions. In this study, a self-adaptive multi-objective harmony search (SaMOHS) algorithm is developed to apply the parameter-setting-free technique, which is an example of a self-adaptive methodology. The SaMOHS algorithm attempts to remove some of the inconvenience from parameter setting and selects the most adaptive parameters during the iterative solution search process. To verify the proposed algorithm, an optimal least cost water distribution network design problem is applied to three different target networks. The results are compared with other well-known algorithms such as multi-objective harmony search and the non-dominated sorting genetic algorithm-II. The efficiency of the proposed algorithm is quantified by suitable performance indices. The results indicate that SaMOHS can be efficiently applied to the search for Pareto-optimal solutions in a multi-objective solution space.     

Evolutionary Neural Network Modeling of Blast Furnace Burden Distribution

A neural network-based model of the burden layer thickness in the blast furnace is presented. The model is based on layer thicknesses estimates from a single radar measurement of the burden (stock) level in the furnace and describes the dependence between the layer thickness and key charging variables. An evolutionary algorithm is applied to train the network weights and connectivity by optimizing the model structure and parameters simultaneously, tackling part of the parameter estimation by linear least squares. This enhances convergence and results in parsimonious and transparent network models with actions that can be explained. Finally, the networks are used in a hybrid model for analyzing novel charging programs and for studying the limits of the charging process.     

Leakage detection in water pipe networks using a Bayesian probabilistic framework

A Bayesian system identification methodology is proposed for leakage detection in water pipe networks. The methodology properly handles the unavoidable uncertainties in measurement and modeling errors. Based on information from flow test data, it provides estimates of the most probable leakage events (magnitude and location of leakage) and the uncertainties in such estimates. The effectiveness of the proposed framework is illustrated by applying the leakage detection approach to a specific water pipe network. Several important issues are addressed, including the role of modeling error, measurement noise, leakage severity and sensor configuration (location and type of sensors) on the reliability of the leakage detection methodology. The present algorithm may be incorporated into an integrated maintenance network strategy plan based on computer-aided decision-making tools.     

Optimal cost design of water distribution networks using harmony search

This study presents a cost minimization model for the design of water distribution networks. The model uses a recently developed harmony search optimization algorithm while satisfying all the design constraints. The harmony search algorithm mimics a jazz improvisation process in order to find better design solutions, in this case pipe diameters in a water distribution network. The model also interfaces with a popular hydraulic simulator, EPANET, to check the hydraulic constraints. If the design solution vector violates the hydraulic constraints, the amount of violation is considered in the cost function as a penalty. The model was applied to five water distribution networks, and obtained designs that were either the same or cost 0.28–10.26% less than those of competitive meta-heuristic algorithms, such as the genetic algorithm, simulated annealing and tabu search under similar or less favorable conditions. The results show that the harmony search-based model is suitable for water network design.     

The berth allocation optimisation with the consideration of time-varying water depths

Confronting the challenges provided by jumbo container vessels, managers of world-class seaports must take into consideration water depths-related constraints when making the vessel berthing decisions. That is, when the water depths satisfy the requirement of jumbo vessels’ drafts, these vessels are allowed to sail through the navigation channel. In this paper, we study a tactical berth allocation problem under the consideration of time-varying water depths. For the resolution, we propose an integer programming model, a set partitioning formulation and an improved differential evolutionary algorithm. Computational experiments demonstrate that the set partitioning formulation outperforms the integer programming model and can be applied to solve small- and medium-sized problems. The improved differential evolutionary algorithm is recommended for large-sized problems.     

Hierarchical parameter estimation of GRN based on topological analysis

Reverse engineering of gene regulatory network (GRN) is an important and challenging task in systems biology. Existing parameter estimation approaches that compute model parameters with the same importance are usually computationally expensive or infeasible, especially in dealing with complex biological networks.In order to improve the efficiency of computational modeling, the paper applies a hierarchical estimation methodology in computational modeling of GRN based on topological analysis. This paper divides nodes in a network into various priority levels using the graph‐based measure and genetic algorithm. The nodes in the first level, that correspond to root strongly connected components(SCC) in the digraph of GRN, are given top priority in parameter estimation. The estimated parameters of vertices in the previous priority level ARE used to infer the parameters for nodes in the next priority level. The proposed hierarchical estimation methodology obtains lower error indexes while consuming less computational resources compared with single estimation methodology. Experimental outcomes with insilico networks and a realistic network show that gene networks are decomposed into no more than four levels, which is consistent with the properties of inherent modularity for GRN. In addition, the proposed hierarchical parameter estimation achieves a balance between computational efficiency and accuracy.Inspec keywords: biology computing, network theory (graphs), reverse engineering, graph theory, genetics, genetic algorithms, directed graphs, parameter estimationOther keywords: hierarchical parameter estimation, GRN, topological analysis, gene regulatory network, important task, computational systems biology, compute model parameters, complex biological networks, efficient information, model quality, parameter reliability, computational modelling, study divides nodes, priority levels, graph‐based measure, previous priority level, hierarchical estimation methodology obtains, computational resources, single time estimation, insilico network, realistic network show, computational efficiency     

Redundancy model for water distribution systems

This paper presents a model, based on head driven simulation, for assessing the redundancy of water distribution systems (WDS). The formulation recognises the pressure dependency of water consumption in the solution procedure. A new algorithm for pressure dependent modelling of WDS has, therefore, been developed. Notable features of the proposed network analysis technique include the introduction of a new subcategory of nodes called key partial-flow nodes and the use of a joint head-flow system of equations. The algorithm is reliable, quick and easy-to-implement.The redundancy assessment methodology addresses the randomness of component failure or unavailability. Results are presented which demonstrate the suitability and meaning of the redundancy measure. In particular, it is recommended that redundancy be evaluated along with reliability when assessing system performance. The computer program developed can seamlessly calculate several performance indicators including reliability.     

Neural and fuzzy model performance evaluation of a dynamic production system

Although mathematical modelling techniques are very well developed, some production processes are difficult to be modelled by these modelling techniques or their math-models are too complex to be used for real-time control due to uncertain, imprecise and vague parameters’ relations. Spray dryers are complex, dynamic and ill-defined production processes. Their product (powder) must have a controllable size distribution consisting of spherical shapes and free-flowing characteristic of particles, which is required for an ideal pressing operation to overcome the product sticking in the dies. The relations of production process' parameters are highly non-linear. In this study, these non-linear parameters were studied and three different soft-computing intelligent models were developed and used to predict uncertain parameter relations. The first is the fuzzy model of the production process; the others are the artificial neural network (ANN) architectures; the back-propagation multilayer perceptron (BPMLP) algorithm and the radial basis function network (RBF). To deal with uncertainty and vagueness of the production system, a method (methodology) based on a fuzzy hierarchical analytic process modelling approach and two ANN approaches was applied. The performance of the BPMLP algorithm was found most vigorous than the RBF and fuzzy modelling approach.     

Process integration design methods for water conservation and wastewater reduction in industry

This three part series of papers addresses operational techniques for applying mass integration principles to design in industry with special focus on water conservation and wastewater reduction. Part 1 covers design techniques for any number of wastewater streams containing a single contaminant. The technique comprises a two stage graphical approach. In the first stage, the water pinch diagram is used to identify key design targets (such as the minimum freshwater requirement of the studied system, the amount of achievable water recycling and reuse and the water quality concentration bottleneck) for the industrial process of interest. Key practical insights provided by the water pinch diagram are discussed. In the second stage, source–sink mapping diagrams are used to identify the water recycling and reuse network, and any alternative networks, that achieve the identified targets. A case study is included to illustrate the proposed methodology. Electronic Publication     

An estimation of distribution algorithm for hybrid flow shop scheduling under stochastic processing times

The estimation of distribution algorithm (EDA) has recently emerged as a promising alternative to traditional evolutionary algorithms for solving combinatorial optimisation problems. This paper presents a novel two-phase simulation-based EDA (TPSB-EDA) for minimising the makespan of a hybrid flow shop under stochastic processing times. To address the stochastic scheduling problem efficiently, the proposed TPSB-EDA incorporates a two-phase simulation model to estimate the performance of candidate solutions. In this model, an optimal back propagation network is firstly applied to identify a set of roughly good solutions, and then the selected solutions are further evaluated by a discrete-event simulation algorithm. Moreover, an annealing selection mechanism (ASM) is adopted to preserve the population diversity of EDA. Different from the selection operators of common EDAs, the ASM uses Boltzmann probability in the annealing algorithm to select part of population to establish the probabilistic model. Computation results indicate that the TPSB-EDA provides good solutions in the aspects of solution quality and computational efficiency.     

Designing eco-industrial water exchange networks using fuzzy mathematical programming

This paper presents a model for optimizing water and wastewater reuse among several plants contained in an eco-industrial park. Although these plants operate independently of each other, the establishment of an eco-industrial network requires mutual cooperation among the different firms. To encourage cooperation it is necessary to look at how the benefits of networking relate to the individual goals and targets of the participants. This aspect is incorporated in the model through fuzzy mathematical programming, which is utilized to identify the optimal network that satisfies the fuzzy goals of the participating plants. Two case studies from literature are presented to illustrate the methodology. Results show that the solution satisfying individual targets does not necessarily coincide with the optimum for the entire eco-industrial park.     

Optimal long-term design,rehabilitation and upgrading of water distribution networks

Given a limited budget, the choice of the best water distribution network upgrading strategy is a complex optimization problem. A model for the optimal long-term design and upgrading of new and existing water distribution networks is presented. A key strength of the methodology is the use of maximum entropy flows, which reduces the size of the problem and enables the application of linear programming for pipe size optimization. It also ensures the reliability level is high. The capital and maintenance costs and hydraulic performance are considered simultaneously for a predefined design horizon. The timing of upgrading over the entire planning horizon is obtained by dynamic programming. The deterioration over time of the structural integrity and hydraulic capacity of every pipe are explicitly considered. The upgrading options considered include pipe paralleling and replacement. The effectiveness of the model is demonstrated using the water supply network of Wobulenzi town in Uganda.     

Dealing with design options in the optimization of manufacturing systems: An evolutionary approach

When designing a manufacturing system, several features have often to be determined to optimize a performance criterion. We propose an optimization simulation method that can be applied to optimize problems taking into account design options and any type of parameters. This method is based on the association of an evolutionary algorithm using a specific data structure and a simulation model. New mutation and recombination operators are proposed, which are adapted to this specific data structure. This approach is implemented on a network of Sun workstations, each station being responsible for running its own simulation model and its own evolutionary algorithm. This method is illustrated through the problem of designing a manufacturing system.     

Bi-directional evolutionary structural optimization for design-dependent fluid pressure loading problems

This article presents an evolutionary topology optimization method for compliance minimization of structures under design-dependent pressure loads. In traditional density based topology optimization methods, intermediate values of densities for the solid elements arise along the iterations. Extra boundary parametrization schemes are demanded when these methods are applied to pressure loading problems. An alternative methodology is suggested in this article for handling this type of load. With an extended bi-directional evolutionary structural optimization method associated with a partially coupled fluid–structure formulation, pressure loads are modelled with hydrostatic fluid finite elements. Due to the discrete nature of the method, the problem is solved without any need of pressure load surfaces parametrization. Furthermore, the introduction of a separate fluid domain allows the algorithm to model non-constant pressure fields with Laplace's equation. Three benchmark examples are explored in order to show the achievements of the proposed method.     

Optimal operation of multiquality water distribution systems: unsteady conditions

This paper describes the methodology and application of a genetic algorithm scheme tailor-made to EPANET, for optimizing the operation of a water distribution system under unsteady water quality conditions. The water distribution system consists of sources of different qualities, treatment facilities, tanks, pipes, control valves, and pumping stations. The objective is to minimize the total cost of pumping and treating the water for a selected operational time horizon, while delivering the consumers the required quantities at acceptable qualities and pressures. The decision variables for each of the time steps that encompass the total operational time horizon include: the scheduling of the pumping units, settings of the control valves, and treatment removal ratios at the treatment facilities. The constraints are: head and concentrations at the consumer nodes, maximum removal ratios at the treatment facilities, maximum allowable amounts of water withdrawals at the sources, and returning at the end of the operational time horizon to a prescribed total volume in the tanks. The model is explored through two example applications.     

Application of a genetic algorithm: near optimal estimation of the rate and equilibrium constants of complex reaction mechanisms

In iterative non-linear least-squares fitting, the reliable estimation of initial parameters that lead to convergence to the global optimum can be difficult. Irrespective of the algorithm used, poor parameter estimates can lead to abortive divergence if initial guesses are far from the true values or in rare cases convergence to a local optimum. For determination of the parameters of complex reaction mechanisms, where often little is known about what value these parameters should take, the task of determining good initial estimates can be time consuming and unreliable. In this contribution, the methodology of applying a genetic algorithm (GA) to the task of determining initial parameter estimates that lie near the global optimum is explained. A generalised genetic algorithm was implemented according to the methodology and the results of its application are also given. The parameter estimates obtained were then used as the starting parameters for a gradient search method, which quickly converged to the global optimum. The genetic algorithm was successfully applied to both simulated kinetic measurements where the reaction mechanism contained one equilibrium constant and two rate constants to be fitted, and to kinetic measurements of the complexation of Cu²⁺ by 1,4,8,11-tetraazacyclotetradecane where two equilibrium and two rate constants were fitted. The implementation of the algorithm is such that it can be generally applied to any reaction mechanism that can be expressed by standard chemistry notation. The control parameters of the algorithm can be varied through a simple user interface to account for parameter range and the number of parameters involved.     

Optimal cost design of water distribution networks using a decomposition approach

Water distribution network decomposition, which is an engineering approach, is adopted to increase the efficiency of obtaining the optimal cost design of a water distribution network using an optimization algorithm. This study applied the source tracing tool in EPANET, which is a hydraulic and water quality analysis model, to the decomposition of a network to improve the efficiency of the optimal design process. The proposed approach was tested by carrying out the optimal cost design of two water distribution networks, and the results were compared with other optimal cost designs derived from previously proposed optimization algorithms. The proposed decomposition approach using the source tracing technique enables the efficient decomposition of an actual large-scale network, and the results can be combined with the optimal cost design process using an optimization algorithm. This proves that the final design in this study is better than those obtained with other previously proposed optimization algorithms.     

Customized evolutionary optimization procedure for generating minimum weight compliant mechanisms

In this article, a customized evolutionary optimization procedure is developed for generating minimum weight compliant mechanisms. A previously-suggested concept of multi-objectivization in which a helper objective is introduced in addition to the primary objective of the original single-objective optimization problem (SOOP) is used here. The helper objective is chosen in a way such that it is in conflict with the primary objective, thereby causing an evolutionary multi-objective optimization algorithm to maintain diversity in its population from one generation to another. The elitist non-dominated sorting genetic algorithm (NSGA-II) is customized with a domain-specific initialization strategy, a domain-specific crossover operator, and a domain-specific solution repairing strategy. To make the search process computationally tractable, the proposed methodology is made suitable for parallel computing. A local search methodology is applied on the evolved non-dominated solutions found by the above-mentioned modified NSGA-II to refine the solutions further. Two case studies for tracing curvilinear and straight-line paths are performed. Results demonstrate that solutions having smaller weight than the reference design solution obtained by SOOP are found by the proposed procedure. Interesting facts and observations brought out by the study are also narrated and conclusions of the study are made.