Browsing by Author "Mulholland, Michael."

Now showing 1 - 19 of 19

Adaptive dynamic matrix control for a multivariable training plant.
(2001) Guiamba, Isabel Remigio Ferrao.; Mulholland, Michael.
Dynamic Matrix Control (DMC) has proven to be a powerful tool for optimal regulation of chemical processes under constrained conditions. The internal model of this predictive controller is based on step response measurements at an average operating point. As the process moves away from this point, however, control becomes sub-optimal due to process non-linearity. If DMC is made adaptive, it can be expected to perform well even in the presence of uncertainties, non-linearities and time-vary ing process parameters. This project examines modelling and control issues for a complex multivariable industrial operator training plant, and develops and applies a method for adapting the controller on-line to account for non-linearity. A two-input/two-output sub-system of the Training Plant was considered. A special technique had to be developed to deal with the integrating nature of this system - that is, its production of ramp outputs for step inputs. The project included the commissioning of the process equipment and the addition of instrumentation and interfacing to a SCADA system which has been developed in the School of Chemical Engineering.
Applied Process Control : errata.
(Wiley-VCH., 2016) Mulholland, Michael.
“This document will be updated periodically to list errors discovered in the two books of Michael Mulholland entitled: “Applied process control : essential methods” and “Applied process control : efficient problem solving”.
Applied process control : simulations.
(2015-12-09) Mulholland, Michael.
A number of interactive simulations are provided which explore some of the techniques presented in Applied Process Control – Essential Methods, and Applied Process Control – Efficient Problem Solving. The applications arise variously from industrial research studies, undergraduate projects and student laboratory experiments. This text is a step-by-step tutorial for the use of the simulator program, and the execution of the exercises for each application. It refers to the accompanying program RTC (Real-Time Control). No support can be offered for the use of the program and the conducting of the exercises. The correspondence between the simulation studies and sections of Applied Process Control – Essential Methods, is roughly as follows: Chapter 2 Simulations – Openloop 3.1, 3.2, 3.3, 3.4 Chapter 3 Simulations – Frequency response 4.2.6.2, 8.6 Chapter 4 Simulations – SISO closedloop 4.2, 4.2.6 Chapter 5 Simulations – SISO optimisers 4.2, 4.7, 4.10, 4.11 Chapter 6 Simulations – Multi-loop strategies 4.3-4.9, 5.3.1, 5.3.2, 5.3.3 Chapter 7 Simulations – MIMO closedloop and DMC 4.2.6, 7.8, 7.8.2, 8.6, 8.7 Chapter 8 Simulations – Observers 6.2, 6.4.1, 6.5.1 Chapter 9 Simulations – Hybrid systems 7.14.3
Development amd implementation of a real-time observer model for mineral processing circuits.
(2004) Vosloo, John-Roy Ivy.; Mulholland, Michael.
Mineral processing plan ts, such as LONMIN's Eastern Platinum B-stream, typically have few on-line measurements, and key measures of performance such as grade only become available after samples have been analysed in the laboratory. More immediate feedback from a dynamic observer model promises enhanced understanding of the process, and facilitates prompt corrective actions, whether in open or closed loop . Such plant s easily enter sub-optimal modes such as large , uselessly re-circulating loads as the feed conditions change. Interpretation of such modes from key combinations of the variables deduced by an observer model , using a type of expert system, would add another level of intelligence to benefit operation. The aim of this thesis was to develop and implement a dynamic observer model of the LONMIN Eastern Platinum B-Stream into one of the existing control platforms available at the plant , known as PlantStar®, developed by MINTEK. The solution of the system of differential and algebraic equations resulting from this type of flowsheet modelling is based on an extended Kalman filter, which is able to dynamically reconcile any measurements which are presented to it, in real time. These measurement selections may also vary in real time, which provides flexibility of the model solution and the model 's uses. PlantStar passes the measurements that are available at the plant, to the dynamic observer model through a "plugin" module, which has been developed to incorporate the observer model and utilise the PlantStar control platform. In an on-line situation, the model will track the plant's behaviour and continuously update its position in real-time to ensure it follows the plant closely. This model would then be able to run simulations of the plant in parallel and could be used as a training facility for new operators, while in a real-time situation it could provide estimates of unmeasurable variables throughout the plant. An example of some of these variables are the flotation rate constants of minerals throughout the plant, which can be estimated in real time by the extended Kalman filter. The model could also be used to predict future plant conditions based on the current plant state , allowing for case scenarios to be performed without affecting the actual plant's performance. Once the dynamic observer model and "plugin" module were completed, case scenario simulations were performed using a measured data set from the plant as a starting point because real-time data were unavailable as the model was developed off-site .
Dynamic modelling and optimal control of sugar crystallisation in a multi-compartment continuous vacuum pan.
(2002) Love, David John.; Mulholland, Michael.
The objective of this work was to determine the operating conditions which would maximise the crystallisation performance of continuous vacuum pans used in the sugar industry. The specific application investigated in detail is crystallisation of high grade product sugar (A-sugar) in a South African raw cane sugar factory. The optimisation studies are based on a detailed dynamic mathematical model of a continuous pan. Whilst this model is based on the published work of others, the selection of variables and the formulation of the equations have been structured to produce a modular model of an individual compartment with the minimum number of independent variables. The independent variables have also been selected to meet the requirements of both a state-space control formulation and those necessary for the dynamic programming technique of optimisation. The modular compartment models are linked together to model a multi-compartment pan and the steady state model is derived as a special case of the dynamic model. For the model to simulate the conditions in South African sugar factories adequately requires appropriate descriptions of sucrose solubility and growth kinetics. Given the limited applicability of published data, experiments were undertaken to determine these parameters. Sucrose solubility in impure solutions was determined in laboratory tests designed to approach equilibrium by dissolution at conditions approximating those during pan boiling. The dependence of crystal growth rate on the concentration of impurity present in the mother liquor was investigated in both laboratory scale and pilot scale batch pan boiling experiments. The primary dependence of crystal growth rate on the super-saturation driving force was determined by fitting the steady state model to results of tests on an industrial scale continuous pan. The dynamic programming technique was used in conjunction with the mathematical model to determine the operating conditions which maximise steady state crystallisation performance. Using the crystallisation parameters determined for South African conditions, this approach has shown that the conventional wisdom of running with high crystal contents in all compartments of continuous pans boiling A-massecuite is not optimum. Pans should operate at lower crystal contents in earlier compartments, only increasing to higher crystal contents towards the final compartment. The specific values depend on seed conditions, pan design and the solubility and growth kinetics. To reap the benefits of being able to determine the optimum steady state operating condition for a continuous pan, it is necessary to be able to achieve effective steady state operation under industrial conditions. This requires both a steady loading on the pan and effective control of the crystallisation conditions within the pan. To stabilise loading, a strategy has been developed which uses buffer tanks in an optimal way to damp out flow fluctuations. This strategy accommodates multiple buffer tanks in series without the amplification of disturbances that occurs with some of the simpler published techniques. The dynamic behaviour of absolute pressure control and compartment feed control were investigated in an industrial scale pan. This work has demonstrated the importance of high quality absolute pressure control and developed techniques for effective automatic tuning of pan feed controls. As part of this research, computer control systems were developed as tools to provide the appropriate monitoring and control of the experiments undertaken.
Formulation and application of a dynamic model for atmospheric point sources.
(1977) Mulholland, Michael.; Scholtz, M. T.; Woodburn, Edward T.
No abstract available.
A high resolution model for multiple source dispersion of air pollutants under complex atmospheric structure.
(1986) Burger, Lucian Willem.; Mulholland, Michael.
No abstract available.
Inverse internal model control of an ethylene polymerisation reactor using artificial neural networks.
(2001) Dunwoodie, Ryan.; Mulholland, Michael.
An artificial neural network is a mathematical black-box modelling tool. This tool can be used to model complex non-linear multivariable processes. In attempting to create an inverse process model of an industrial linear low density polyethylene reactor, several interesting results were encountered. Both time-invariant algebraic and time-invariant dynamic models could adequately represent the process, provided an identified 50-minute time lag was taken into account. A novel variation of the traditional IMC controller was implemented which used two inverse neural network process models. This was named Inverse Internal Model Control (IIMC). This controller was initially tested on a real multivariable pump-tank system and showed promising results. The IIMC controller was adapted to an on-line version for the polymer plant control system. The controller was run in open loop mode to compare the predictions of the controller with the actual PID ratio controllers. It was hoped that by incorporating neural network models into the controller, they would take the non-linearity and coupling of the variables into account, which the present PID controllers are unable to do. The existing PID controllers operate on separate loops involving the two main feeds (co-monomer and hydrogen) to the reactor, which constitute aspects of the control system in which the scope for advanced control exists. Although the control loop was not closed, the groundwork has been laid to implement a novel controller that could the operation of the plant.
Mathematical modelling and experimental study of the kinetics of the acid sulphite pulping of eucalyptus wood.
(1992) Watson, Edward.; Wright, Dave.; Mulholland, Michael.
The chemistry of the batch cooking process at Sappi Saiccor, relating to both the pulp and liquor, was investigated with the aim of using kinetic expressions to develop an improved process control model. The mill produces dissolving pulps using the acid sulphite method. Three process reactions were identified as important: cellulose hydrolysis, delignification and hemicellulose dissolution. Of these, cellulose hydrolysis is the most important since the primary aim is to achieve a targeted cellulose degree of polymerisation (DP) or viscosity (DP is commonly expressed in terms of this measurement). This is directly determined by the rate of this reaction during the cook, and the acidity of the cooking liquor was found to be the key factor. As existing equipment was not suitable for obtaining the data required to perform a kinetic analysis, a pilot plant was constructed. A commercially available probe was used for the first time to measure pH directly. The measured acidity is not directly equivalent to hydrogen ion activity at these temperatures and pressures; however, since the conditions of each cook are similar the errors incurred were found to be constant from cook to cook. The probe was found to be prone to drift due to ageing and this was accounted for by using an 'on line' calibration based on a liquor analysis. The kinetics of the cellulose hydrolysis reaction were determined using the on-line measurement of acidity and the concept of degradation increase (DI) which relates the reduction in DP value to the rate at which the polymeric chains are split. Delignification and hemicellulose dissolution were examined, since it is beneficial to maximise these reactions to reduce the quantities of chemicals consumed during the bleaching process. A model for controlling cooks to a set target cellulose DP value within a set time was developed based on the reaction kinetics. This was capable of predicting cooking conditions required with sufficient accuracy to control the cellulose DP value to within ±6 cp SNIA on the viscosity scale.
Model predictive control of hybrid systems.
(2002) Ramlal, Jasmeer.; Mulholland, Michael.
Hybrid systems combine the continuous behavior evolution specified by differential equations with discontinuous changes specified by discrete event logic. Usually these systems in the processing industry can be identified as having to depend on discrete decisions regarding their operation. In process control there therefore is a challenge to automate these decisions. A model predictive control (MPC) strategy was proposed and verified for the control of hybrid systems. More specifically, the dynamic matrix control (DMC) framework commonly used in industry for the control of continuous variables was modified to deal with mixed integer variables, which are necessary for the modelling and control of hybrid systems. The algorithm was designed and commissioned in a closed control loop comprising a SCADA system and an optimiser (GAMS). GAMS (General Algebraic Modelling System) is an optimisation package that is able to solve for integer/continuous variables given a model of the system and an appropriate objective function. Online and offline closed loop tests were undertaken on a benchmark interacting tank system and a heating/cooling circuit. The algorithm was also applied to an industrial problem requiring the optimal sequencing of coal locks in real time. To complete the research concerning controller design for hybrid behavior, an investigation was undertaken regarding systems that have different modes of operation due to physicochemical (inherent) discontinuities e.g. a tank with discontinuous cross sectional area, fitted with an overflow. The findings from the online tests and offline simulations reveal that the proposed algorithm, with some system specific modification, was able to control each of the four hybrid systems under investigation. Based on which hybrid system was being controlled, by modifying the DMC algorithm to include integer variables, the mixed integer predictive controller (MIPC) was employed to initiate selections, switchings and determine sequences. Control of the interacting tank system was focused on an optimum selection in terms of operating positions for process inputs. The algorithm was shown to retain the usual features of DMC (i.e. tuning and dealing with multivariable interaction). For a system with multiple modes of operation i.e. the heating/cooling circuit, the algorithm was able to switch the mode of operation in order to meet operating objectives. The MPC strategy was used to good effect when getting the algorithm to sequence the operation of several coal locks. In this instance, the controller maintained system variables within certain operating constraints. Furthermore, soft constraints were proposed and used to promote operation close to operating constraints without the danger of computational failure due to constraint violations. For systems with inherent discontinuities, a MPC strategy was proposed that predicted trajectories which crossed discontinuities. Convolution models were found to be inappropriate in this instance and state space equations describing the dynamics of the system were used instead.
Modelling and control of a co-current sugar dryer.
(2001) Lacave, Benoit.; Mulholland, Michael.
The drying of sugar is the last step in the recovery of solid sugar from sugar-cane. To ensure that the sugar can be transported and stored, the final moisture content leaving the sugar mill must be carefully controlled. Data spanning periods of normal plant operation were collected at the Tongaat-Hulett Ltd Darnall sugar mill. These measurements were reconciled to achieve instantaneous mass and energy balances across the sugar dryer. Using these measurements, a general model has been developed to simulate the sugar drying. It includes ten compartments through which the sugar and drying air flow, with a mass and energy balance in each compartment. It was assumed that a "film" around the sugar crystal is supersaturated, and that crystallisation is still occurring. A sorption isotherm determining the equilibrium moisture content of the sugar, at which point mass transfer ceases, was included. The model has been matched to process measurements by adjusting the heat and mass transfer coefficients. A Dynamic Matrix Controller was developed and tested off-line on the model, using the reconciled measurement sequences. The controller manipulated the inlet air temperature in order to control the exit sugar moisture content. The model predictive control format successfully dealt with the large process dead-time (5 minutes).
Modelling and control of potable water chlorination.
(2003) Pastre, Amelie.; Mulholland, Michael.; Brouckaert, Christopher John.; Buckley, Christopher Andrew.; Le Lann, Marie Veronique.
In potable water preparation, chlorination is the last step before the potable water enters the distribution network. Umgeni Water Wiggins Waterworks feeds the Southern areas of Durban. A reservoir at this facility holds treated water before it enters the distribution network. To ensure an adequate disinfection potential within the network, the free chlorine concentration in the water leaving the reservoir at the Umgeni Water Wiggins Waterworks should be between 0.8 and 1.2 mg/L. The aim of this study was to develop an effective strategy to predict and control the chlorine concentration at the exit of the reservoir. This control problem is made difficult by the wide variations in flow and level in the reservoirs, together with reactive decay of the chlorine concentration. A Computational Fluid Dynamic study was undertaken to gain understanding of the physical processes operating in the reservoir (FLUENT software). As this kind of modelling is not yet applicable for real-time control, compartment models have been created to simulate the behaviour of the reservoir as closely as possible, using the results of the fluid dynamic simulation. These compartment models were initially used in an extended Kalman filter (MATLAB software). In a first step, they were used to estimate the kinetic factor for chlorine consumption and in a second step, they predicted the chlorine concentration at the outlet of the reservoir. The comparison between predictions and data, allowed the validation of the compartment models. A predictive control strategy was developed using a Dynamic Matrix Controller, and tested offline on the compartment models. The controller manipulated the chlorine concentration in the inlet of the reservoir in order to control the chlorine concentration in the outlet of the reservoir. Finally, the simplest compartment model was implemented on-line, using the Adroit SCADA system of the plant, in the form of a Kalman filter to estimate the chlorine decay constant, as well as a predictive model, using this continuously-updated decay parameter. The adaptive Dynamic Matrix Controller using this model was able to control the outlet chlorine concentration quite acceptably, and further improvements of the control performance are expected from ongoing tuning.
Optimal operation of a water distribution network by predictive control using MINLP.
(2004) Biscos, Cedric P. G.; Mulholland, Michael.; Buckley, Christopher Andrew.; Brouckaert, Christopher John.; Le Lann, Marie Veronique.
The objective of this research project is to develop new software tools capable of operational optimisation of existing, large-scale water distribution networks. Since pumping operations represent the main operating cost of any water supply scheme, the optimisation problem is equivalent to providing a new sequence for pumping operations that makes better use of the different electricity tariff structures available to the operators of distribution systems. The minimisation of pumping costs can be achieved by using an optimal schedule that will allow best use of gravitational flows, and restriction of pumping to low-cost power periods as far as possible. A secondary objective of the operational optimisation is to maintain the desired level of disinfectant chlorine at the point of delivery to consumers. There is a steady loss of chlorine with residence time in the system. If the level drops too low there is a risk of bacterial activity. Re-dosage points are sometimes provided in the network. Conversely, too high a level produces an unacceptable odour. The combinatation of dynamic elements (reservoir volumes and chlorine concentration responses) and discrete elements (pump stati and valve positions) makes this a challenging Model Predictive Control (MPC) and constrained optimisation problem, which was solved using MINLP (Mixed Integer Non-linear Programming). The MINLP algorithm was selected for its ability to handle a large number of integer choices (valves open or shut / pumps on or off in this particular case). A model is defined on the basis of a standard element, viz. a vessel containing a variable volume, capable of receiving multiple inputs and delivering just two outputs. The physical properties of an element can be defined in such a way as to allow representation of any item in the actual network: pipes (including junctions and splits), reservoirs, and of course, valves or pumps. The overall network is defined by the inter-linking of a number of standard elements. Once the network has been created within the model, the model predictive control algorithm minimises a penalty function on each time-step, over a defined time horizon from the present, with all variables also obeying defined constraints in this horizon. This constrained non-linear optimization requires an estimate of expected consumer demand profile, which is obtained from historical data stored by the SCADA system monitoring the network. Electricity cost patterns, valve positions, pump characteristics, and reservoir properties (volumes, emergency levels, setpoints) are some of the parameters required for the operational optimisation of the system.
Optimization of a multi-level steam distribution system by mixed integer non-linear programming.
(2001) Saunion, Roland.; Mulholland, Michael.
The objective of this project is to optimize the SAPREF oil refinery steam distribution in which imbalances between the various levels presently require the venting of steam from the lowest level. The overall steam balance shows that the problem originates from an excess of high·pressure (HP) steam production for too few medium pressure steam users and turbines. We proposed to solve this problem by considering the replacement of selected steam turbines with electrical drives. Given a set of demands of electricity, mechanical power and steam at various pressure levels, the objective is to recommend configuration changes to minimize overall cost. This is not a trivial problem, as steam not passed down through turbines to lower levels can create a shortage there, so a combination of replacements is required. The variables of the problem are both decision variables on every steam turbine and continuous variables, such as flows and enthalpies. These decision variables are integer variables, 0 or 1 for every steam turbine. Depending on whether it is kept on steam use or replaced with an electrical drive, these variables are as follows: E = 0: keep the existing steam turbine E - 1: switch it to an electrical drive. A complete and realistic model of this utility section must be constructed in order to represent the actual distribution accurately. This model will include an objective function to minimize, some equality and inequality constraints, and some cost functions. If we want this model to be accurate, we shall have to deal with nonlinearities to avoid simplifications, and these non-linearities could lead to infeasabilities or sub-optimal solutions. So we are facing a typical MTNLP (Mixed Integer Non-Linear Programming) problem to find optimal configuration changes which will maximize the return on investment, meeting the electrical, mechanical and steam demands of the refinery. In order to solve this difficult optimization problem we shall use the user-friendly package GAMS (General Algebraic Modeling System).
Practical on-line model validation for model predictive controllers (MPC)
(2010) Naidoo, Yubanthren Tyrin.; Mulholland, Michael.
A typical petro-chemical or oil-refining plant is known to operate with hundreds if not thousands of control loops. All critical loops are primarily required to operate at their respective optimal levels in order for the plant to run efficiently. With such a large number of vital loops, it is difficult for engineers to monitor and maintain these loops with the intention that they are operating under optimum conditions at all times. Parts of processes are interactive, more so nowadays with increasing integration, requiring the use of a more advanced protocol of control systems. The most widely applied advanced process control system is the Model Predictive Controller (MPC). The success of these controllers is noted in the large number of applications worldwide. These controllers rely on a process model in order to predict future plant responses. Naturally, the performance of model-based controllers is intimately linked to the quality of the process models. Industrial project experience has shown that the most difficult and time-consuming work in an MPC project is modeling and identification. With time, the performance of these controllers degrades due to changes in feed, working regime as well as plant configuration. One of the causes of controller degradation is this degradation of process models. If a discrepancy between the controller’s plant model and the plant itself exists, controller performance may be adversely affected. It is important to detect these changes and re-identify the plant model to maintain control performance over time. In order to avoid the time-consuming process of complete model identification, a model validation tool is developed which provides a model quality indication based on real-time plant data. The focus has been on developing a method that is simple to implement but still robust. The techniques and algorithms presented are developed as far as possible to resemble an on-line software environment and are capable of running parallel to the process in real time. These techniques are based on parametric (regression) and nonparametric (correlation) analyses which complement each other in identifying problems -iiwithin on-line models. These methods pinpoint the precise location of a mismatch. This implies that only a few inputs have to be perturbed in the re-identification process and only the degraded portion of the model is to be updated. This work is carried out for the benefit of SASOL, exclusively focused on the Secunda plant which has a large number of model predictive controllers that are required to be maintained for optimal economic benefit. The efficacy of the methodology developed is illustrated in several simulation studies with the key intention to mirror occurrences present in industrial processes. The methods were also tested on an industrial application. The key results and shortfalls of the methodology are documented.
Real-time observer model for Kraft wood digester.
(2005) Mollereau, Antoine.; Mulholland, Michael.
At SAPPI-Tugela a continuous Kraft wood chip digester operates in EMCC mode (extended modified continuous cooking). Chips are initially exposed to a NaOH / Na2S liquor at high temperature in the top section. The chips move downward in plug flow passing circumferential screens used to draw liquor for various circulations. About midway down the spent black liquor is removed and the chips enter the cooler bottom section where some further reaction and washing occurs. Liquor level and chip level are maintained close to each other near the top. Chips require 8-12 hours to pass through the digester, depending on the chip feed rate. The key parameter of interest at the digester exit is the Kappa number, which is a measure of the extent of delignification which has occurred. Different board and paper products require different Kappa number pulp feed. (Final properties such as tensile, tear and bursting strengths will also depend on the way fibres have been modified in the digestion). The objective of this investigation is to predict the Kappa number of the product pulp in real-time, thus facilitating quicker reaction than the present dependence on laboratory analysis permits, possibly even allowing closed-loop control. The extent of delignification depends on liquor strength, temperature and exposure time, with final Kappa number also depending on the properties of the chip feed (wood type and moisture content). Compensation to maintain a steady Kappa number is made difficult by the long and varying residence time, and the fact that any changes apply to the whole profile held up in the digester. A number of static models for Kappa number prediction have been developed by previous workers, but these do not compare well with plant measurements. The collection of data from the Sappi-Tugela reactor, and the pulp quality reports, have been used to determine an efficient model. This step required a considerable data collection exercise, and similar results to the quality reports have been obtained using a simple linear model based on this data. The problem of model error is being reduced by arrangement as a Smith Predictor, in which the model is intermittently corrected by available laboratory analyses. At the same time, an interface was created, in order to synchronise measurement data for the chips presently leaving the reactor. In order to deal with the dead time, each parcel of chips entering the reactor is effectively tracked, and the changes in Kappa number integrated for reaction time under the varying conditions in transit. Knowing the present inventory of the reactor, this model can also be run forward in time as a predictive controller, to determine optimal control actions for maintenance of the target Kappa number.
Real-time observer modelling of a gas-phase ethylene polymerisation reactor.
(2000) Thomason, Richard.; Mulholland, Michael.
The desire for precise polymer property control, minimum wastage through grade transitions, and early instrument fault detection, has led to a significant effort in the modelling and control of ethylene polymerisation world-wide. Control is difficult due to complex inter-relationships between variables and long response times from gas to solid phase. The approach in this study involves modelling using the kinetic equations. This forms the basis of a scheme for real-time kinetic parameter identification and Kalman filtering of the reactor gas composition. The scheme was constructed off-line and tested on several industrial polymer grades using historical plant data. The scheme was also converted into a form for use on the linear low-density polyethylene plant, Poly 2, at POLlFIN Limited. There proved to be no difficulty in the identification step, but the Kalman filter requires more tuning for reliable fault detection. The software has been commissioned on-line and results from the POLlFIN plant match the off-line model exactly.
Simulation and optimisation of the controls of the stock preparation area of a paper machine.
(2004) Lacour, Sebastien.; Mulholland, Michael.
At Mondi Paper Ltd, Merebank, South of Durban, Paper Machine 2 has recently been transferred onto a Distributed Control System (DCS). This was seen as a good opportunity to enhance the control of the pulp feed to the machine. A prime concern in operating a paper machine is to ensure consistent set-point paper properties in the Cross-Direction (CD: ie. across the paper width) and in the Machine-Direction (MD: ie. along the paper length). Sophisticated adjustments are available to ensure an even feed of the stock (consistencies around 2% m/m wood fibres in water) from the head-box across the receiving width of the paper machine. The properties of prime interest as the pulp is pumped through the head-box distributor onto the receiving belt of the machine are the basis weight (fibre mass per unit area) and moisture content (per unit area). However, the distribution system is highly dependent on the properties of the stock as it arrives at the head-box. Variations in upstream chest levels, the supplied pressure, flow-rate and fibre/water ratio, all cause MD and even CD variations. The problems of maintaining steady conditions at the head-box are well known, and are understood to arise from sub-optimal control in the preceding section involving a blend chest and machine chest, amongst other items, where several pulp streams and dilution water are combined. A number of control loops are involved, but appear to require different tuning for different paper grades. Often individual loops are taken off-line. In this study, an understanding of the controller interactions in the stock preparation section has been developed by detailed dynamic modelling, including all of the existing control loops. The model is built up in a modular fashion using a basic element, having one input (which can collect multiple streams originating elsewhere) and four outputs, linked through a vessel of variable volume. Several basic elements are linked together to form the overall system. All of the necessary properties can be defined so that the model allows the simulation of all features of the network: vessels, pipes, junctions, valves, levels and consistencies. A set of first order differential equations is solved which includes total water balance, species mass balances, derivatives of flow controller action, and derivatives of supervisory controller action. Supervisory controllers for consistency or level cascade onto flow controllers. Flow controllers manipulate valves which give a first-order dynamic response of actual flow. Where valves are manipulated directly by the supervisory level, the flow controller is effectively bypassed. This study involves a constraint problem around the blend chest, resulting in a loss of specification at the paper machine. This was solved by the implementation of a static optimiser. Its objective function penalizes deviations from setpoint of five parameters (ratios, consistency and level) using respective weight factors. Both the model and its optimiser were included in a simulator designed with the graphical user interface (GUI) of Matlab. The simulator has then been used to explore control performance over the operating range, by means of a set of scenarios.
Simulation tests for the operation of a water main with break pressure tanks.
(2016) Ally, Ismaeel Haroon Tar.; Mulholland, Michael.; Buckley, Christopher Andrew.; McLeod, Neil.
The Ashley Drive break pressure tank (BPT-20 𝑀ℓ) has been installed on Durban’s Western Aqueduct. Its purpose is to release the 20 bar gravity head of the 1.4m trunk main supply from Umgeni Water at Umlaas Road. The expected peak conditions (400 𝑀ℓ/𝑑𝑎𝑦) will only allow 14 minutes for valves to close, yet they must be moved slowly in order to avoid dynamic shock. The high pressure upstream supply is admitted to the BPT through a set of thee parallel sleeve valves, which are in a control loop to maintain level in the BPT against the downstream draw. These cavitation-resistant valves cannot be operated without electrical power, so an added complication of the design is a set of 3 hydraulically-operated globe valves which switch in at extreme tank levels. Though the commissioning of the Ashley Drive BPT is already in progress, it is important to simulate the overall operation of the system for projected future flows, in order to detect possible operational problems, and to build in solutions if necessary. Optimisations include such issues as the valve closing sequence and speeds, settling level variations, and smoothness of the draw from Umgeni Water. The simulation study involved the modelling of the trunk main, the Ashley Drive BPT, the downstream Wyebank BPT and the reservoirs drawing from the trunk main before and after these two BPTs. Data handling techniques were developed in order to formulate the daily demand profiles for each of the reservoirs. Design information was used to calculate the hydraulic parameters that featured in the simulation, and to determine the residual pressures at the inlet valve sets of the BPTs. Implicit calculations with the Newton-Raphson iterative method were employed in order to obtain a pressure distribution across the BPT valves. Simple mechanisms were built into the MATLAB® program in order to accommodate the complexities of the system, e.g. the possibility of power loss, valve or BPT chamber maintenance, or the deliberately slowed movement of the valves to avoid pressure surges within the pipeline. The analysis of the results of the simulation study involved examining the efficacy of the control set-points and valve sequencing, and determining whether these settings satisfy the design specifications. Random and anticipated scenario testing was carried out within the study in order to accommodate for situations such as electricity outages or unusual consumer demands. The BPT control system was analysed to assess its adequacy and the risks associated with the proposed staggered sleeve valve control scheme. The results of this investigation are presented as multiple time-sequence graphs depicting the results of the different scenario tests. Support for the design concept, additional recommendations and indications of adverse scenarios, have emerged from this study. The original design is found to be capable of duty within the ranges of expected normal operation in 2036, and the system was observed to be capable of conveying a throughput greater than that of the design. The normal operating level was also found to be higher than intended, and valve oscillations were deemed a significant concern. It was established that operation with just two sleeve valves active within each BPT would achieve better correspondence to the design specifications. The revised control system (Control 2.0) was found to be better suited to the application, but was also diagnosed to be too slow to react under certain circumstances.