New Publications are available for Chemical variables control
http://dl-live.theiet.org
New Publications are available now online for this publication.
Please follow the links to view the publication.Design of ZigBee-based wireless control system of circulating water concentration ratio
http://dl-live.theiet.org/content/conferences/10.1049/cp.2010.1038
This article introduced the wireless sensors and wireless actuators combined of ZigBee wireless transmission with sensor or actuator, and its application in circulating water concentration rate control system. Starting with the system topology, it introduced the CC2430 microcontroller and ZigBee features detailed design process of hardware and software in the system, including interface circuit, and each part initial setup, data acquisition and control, and a detailed discussion system design problems.Monitoring the performance of controllers on batch plants
http://dl-live.theiet.org/content/conferences/10.1049/ic_20020224
This paper investigates the benefits that the partial least squares (PLS) modelling approach offers engineers involved in the operation of fed-batch fermentation systems. It is shown that PLS can be used to provide accurate inference of quality variables that are difficult to measure online, such as biomass concentration. It is further shown that the same PLS model can be used to provide fault detection and isolation capabilities and that it can be integrated within a model predictive control framework to regulate the growth of biomass within the fermenter. This model predictive controller is shown to provide its own monitoring capabilities that can be used to identify faults within the process and also within the controller itself. (4 pages)Neural network applications in the water industry
http://dl-live.theiet.org/content/conferences/10.1049/ic_20010111
The operation of water treatment plants is significantly different from most manufacturing industrial operations because raw water sources are often subject to natural perturbations like flood and drought, both of which significantly affect the characteristics of the abstracted water. More recently, improved sensor technology has enabled the successful regulation of variables such as pH and chlorine residual. Without a precise knowledge of the characteristics of the material to be removed, most chemical dosage requirements for primary water treatment are determined from laboratory measurements which are conducted (usually) not less than once a day. This paper gives a brief explanation of water treatment plant operation, and outlines a number of case studies where system knowledge contained within artificial neural networks has been used to provide solutions to operational problems within the water industry. (6 pages)Predictive control using multiple model networks
http://dl-live.theiet.org/content/conferences/10.1049/ic_19990533
The aim of this paper is to describe a nonlinear modelling architecture, called the local model network (LMN), which introduces transparency while offering distinct advantages for nonlinear model-based control. Simulation results for a pH neutralisation process are used to illustrate the performance benefits of LMNs for two novel nonlinear dynamic matrix control schemes. (7 pages)On-line implementation of a model predictive controller on a multivariable chemical process
http://dl-live.theiet.org/content/conferences/10.1049/ic_19990539
An investigation into neural network model predictive control is described in this paper. The control strategy developed is applied to a laboratory process to control temperature, pH and dissolved oxygen. The main difficulties in control of this process are nonlinearity, coupling effects among variables and long time-delay in the heat exchanger. Parallel neural models are developed from real process data for the use with online model predictive control and off-line simulations. The online control results are demonstrated. (5 pages)A novel sensor for monitoring oilfield fouling in near-real time
http://dl-live.theiet.org/content/conferences/10.1049/ic_19990722
The goal of our research was to improve the control of process fouling by developing new monitors that could be used to measure and predict fouling in real time. Such monitors could be used to assure the application of the proper chemical treatment at the proper dosage. Operation of these monitors in real time will provide a direct condition-based assessment of fouling and a means for optimizing the response. One approach to direct fouling measurement taken by Nalco/Exxon is the use of the quartz crystal microbalance (QCM) technique. A QCM is a device that uses a piezoelectric quartz crystal wafer with electrodes bonded to the crystal as a frequency control device, or resonator. We have developed our QCM-based instrumentation, referred to as a thickness-shear mode resonator (TSMR), using an oscillator circuit that minimizes the effects of resonator loss over a wide dynamic range of liquid viscosity and density. Results demonstrate the sensitivity and versatility of the TSMR technique for monitoring both rigid and visco-elastic fouling events in both lab and field environments. The TSMR voltage response to changes in the sample's visco-elastic properties is more sensitive than a standard oscillator circuit design using only frequency measurements. This response characteristic of the TSMR has enabled us to develop a highly sensitive monitor for both rigid and visco-elastic deposits. (4 pages)Artificial intelligence approaches to model-based control
http://dl-live.theiet.org/content/conferences/10.1049/ic_19981030
While of undoubted value for nonlinear identification and control of dynamic systems, neural networks have a number of limitations for practical applications. Thus, in online training, due consideration must be given to the necessity for regularisation with noisy data and to the choice of network architecture. More fundamentally, the nontransparent black-box nature of neural models make it difficult to include a priori system information, and to interpret the final structure meaningfully in terms of physical process characteristics. Neural approaches also fail to exploit the significant body of theoretical results available for conventional model-based control, making it difficult to analyse the closed-loop behaviour in terms of stability and robustness. The aim of this paper is to describe a nonlinear modelling architecture, called the local model network (LMN), which introduces transparency while offering distinct advantages for nonlinear model-based control. Simulation results for a pH neutralisation process are used to illustrate the performance benefits of LMNs for nonlinear dynamic matrix control (DMC) and for nonlinear internal model control (IMC). (6 pages)Choosing the right control structure - examples from the petrochemical industry
http://dl-live.theiet.org/content/conferences/10.1049/ic_19980408
There is a strong move in the process industries towards more process automation and control to improve profitability through more production, energy saving, reduced losses and more consistent qualities. Automatic techniques to aid more efficient design of control schemes will be of value, but there are some requirements that need to be met. This paper describes two real industrial control examples which can illustrate some of these features for discussion. The first is a distillation column with reaction, and the second a distillation column for separating isobutane from a mixed butanes feed. (3 pages)SISO and MIMO variable structure control of fixed bed bioreactors
http://dl-live.theiet.org/content/conferences/10.1049/cp_19980232
In this paper, SISO and MIMO robust variable structure controls of fixed bed bioreactors are developed. The process considered must regulate the nitrogen content of drinkable water at desired values imposed by international norms. Micro-organisms fixed in the reactor absorb the nutrients in such a way that the substrate concentrations decrease in the outflowing water. The addition of a carbon source is needed in this operation. A SISO variable structure control is used to regulate only the total concentration of nitrates and nitrites by acting either on the influent flow rate, or on the ethanol concentration. In order to optimise the addition of the ethanol which is the carbon source and regulate the ethanol concentration of drinkable water, a MIMO variable structure control is used. The complexity of a control problem is due to nonlinear and time varying behaviour of micro-organisms used for consuming harmful substrates. The performances of the control laws are illustrated by simulations.Plant trials of a rule based control system for coal combustion
http://dl-live.theiet.org/content/conferences/10.1049/cp_19980279
A rule-based control system for coal combustion on a chain-grate stoker is presented. The objectives of the controller were to permit, automatically, continuous running of the process at different load levels while achieving high efficiency and maintaining permitted emissions levels. Its achievement of these goals has been demonstrated in plant trials on a 1 MW stoker and test results are presented.A variable transformation approach to in-line pH control
http://dl-live.theiet.org/content/conferences/10.1049/cp_19980353
Current practice and research into pH control principally entails the use of continuous stirred tank reactors (CSTR). Whilst the over-design of this process does indeed damp out disturbances, it is expensive and leads to problems with control. Industry requires systems that are: economical, both in terms of capital and operating costs; robust from a control point of view; operable and flexible from a process point of view. pH control of neutralisation is a difficult nonlinear control problem. This is largely due to the gross nonlinear behaviour of pH measurement, but also due to the variable time delays inherent in the process. Nonlinear processes have traditionally been controlled by a combination of linear control methods and gain scheduling. However in the last few years much progress has been made in the development of nonlinear control systems. This paper outlines a variable transformation approach, that is an anti-logging technique which removes the nonlinearity at source and yields an hydrogen ion concentration which is easier to handle.A real-time simulation of a 200 MW thermal power plant for optimising combustion control
http://dl-live.theiet.org/content/conferences/10.1049/cp_19980341
This paper gave an overview of the work being carried out by The Queen's University of Belfast TOPGEN project. The development implementation of a data server for monitoring and storing data from the Ballylumford power plant in Northern Ireland was presented. A computer power plant model was designed and validated using data collected by the data server. This plant model formed the basis of a virtual plant system that included an industrial standard distributed control system with multiple operator screens on a local area network. An FTP server was also incorporated for connection over the Internet with the software package CORAGE. The virtual plant system was then used as a test-bed for CORAGE to be tuned and tested remotely over the Internet before being applied to the Ballylumford power plant.Model predictive control of a chemical process using neural networks
http://dl-live.theiet.org/content/conferences/10.1049/cp_19980331
A simulation study on the control of a multivariable chemical process by using a neural network model predictive control strategy is described in this paper. The laboratory process, in which temperature, pH and dissolved oxygen are involved, has characteristics typical of industrial processes. The main difficulties in control of this process are non-linearity, coupling effects among variables and long time-delay in heat exchange. Neural sub-system models are developed from real process data for the model predictive control strategy and also for use as a bank of parallel models to represent the process in the control simulation. The control simulations are performed before the online control to gain more insight of the process and to determine suitable controller parameters. The simulation results are demonstrated in the paper.Dissolved oxygen control strategy comparison for an activated sludge process
http://dl-live.theiet.org/content/conferences/10.1049/cp_19980383
There are increasingly important financial incentives and environmental considerations for improving the treatment of wastewater from domestic and industrial users before being released into the environment. Insufficient levels of dissolved oxygen prevent the successful degradation of organic matter present, whereas too high a level causes settling problems, in addition to the waste of energy and hence cost efficiency. Therefore, the need for controlling dissolved oxygen is of great importance. The paper presents simulation results providing comparison of controller performances for three control methods (PID, fuzzy logic and self-tuning control) for a wastewater process with typical influent flow pattern. The set-point to the sludge flow control system is optimised using an objective function including released water quality and financial running cost.Micro-sensor applications in petroleum drilling and completion
http://dl-live.theiet.org/content/conferences/10.1049/ic_19970006
This paper has reviewed some of the measurement needs of the upstream end of the petrochemical industry. the measurements are complex in a difficult environment, but nonetheless many are well suited to a micro-engineering approach, and examples have been given of where the author considers most progress could be made in a short time. (3 pages)Local model networks for nonlinear system identification
http://dl-live.theiet.org/content/conferences/10.1049/ic_19970785
Local model networks represent a nonlinear dynamical system by a set of locally valid submodels across the operating range. Training such feedforward structures involves the combined estimation of the submodel parameters and those of the interpolation functions. The paper describes a new hybrid learning approach for local model networks that uses a combination of singular value decomposition and second order gradient optimization. A new nonlinear internal model control scheme is proposed which has the important property that the controller can be derived analytically. Simulation studies of a pH neutralization process confirm the excellent modelling and control performance using the local model approach. (3 pages)A framework for the implementation of self-tuning/adaptive control at potable water treatment works
http://dl-live.theiet.org/content/conferences/10.1049/ic_19960416
This paper concerns the development of an interactive control scheme that applies self-tuning/adaptive control of simple PI and predictive PI controllers. There has been extensive interest in feedback control systems that automatically adjust their controller settings to compensate for changes in the process or the environment. Such systems are referred to as adaptive. The scheme described satisfies this basic definition but adaptation is not continuous. The algorithm consists of two basic operations. The first is based on continuous monitoring of the process performance and can be programmed to initiate a re-tune when the performance degrades to some pre-specified condition. The self-tuning strategy used, which incorporates a facility for time-delay compensation, will be outlined. The paper concludes with an example common to water treatment, namely the adjustment of the pH of the treated water. (6 pages)Chemical dosing philosophies for a water treatment plant: results of some pilot plant experimentation
http://dl-live.theiet.org/content/conferences/10.1049/cp_19960698
The tight control of pH prior to coagulation is the first phase of a longer term comprehensive research programme. Due to the complex interactions between so many variables it is a necessary precursor to other work. For example, coagulant dose control by means of a streaming current detector (SCD) or feedforward algorithm, which are other areas of on-going investigations, are heavily influenced by pH. Therefore eliminating the effect of this variable via good process regulation is an important and necessary development.μ optimal control of a laboratory plant
http://dl-live.theiet.org/content/conferences/10.1049/cp_19960681
The paper presents the design of a robust control for a real plant at laboratory scale. The process considered consists of the production of a C1H dissolution. The control system aim is to maintain the concentration at prescribed values despite variations in the plant, the presence of disturbances and the inherent nonlinearity of the process. The uncertainties in the model are studied and characterized as complex or real uncertainties. Using this information the controller was designed by optimization of the structured singular value (μ-synthesis). The obtained controller was tested in real time in the plant, fulfilling the design requirements and obtaining good performance in varying conditions.Control of pH in-line using a neural predictive strategy
http://dl-live.theiet.org/content/conferences/10.1049/cp_19960699
Control of an experimental in-line pH process exhibiting varying nonlinearity and deadtime is described. A radial basis function (RBF) artificial neural network is used to model the nonlinear dynamics of the process. Accommodation of the varying process deadtime in the neural model is achieved by the generation of a feed-forward signal, for input to the neural network, from a downstream pH measurement. The feedforward signal is derived from a variable delay model based on process knowledge and a flow measurement. The neural model is then used to realise a predictive control scheme for the process. Development of the neural process model is described and results are presented to illustrate the performance of the neural predictive control scheme which is tested as a regulator at different setpoints.Controllability of SVD control structures for ill-conditioned plants
http://dl-live.theiet.org/content/conferences/10.1049/cp_19960605
A conventional diagonal controller can be designed from very limited process information. In the SVD control structure, a diagonal controller is applied to orthogonal sums and differences of the inputs and outputs. In this work, the process knowledge needed for the design of an SVD controller is analyzed. In contrast to conventional decentralized controllers, the SVD controller can compensate for the process directionality of ill-conditioned processes. The dual composition control problem of distillation columns is of particular interest, since reliable models for these columns are quite hard to obtain.Cumulant/bispectrum model structure identification applied to a pH neutralization process
http://dl-live.theiet.org/content/conferences/10.1049/cp_19960697
A process model structure identification criteria based on the cumulants and bispectrum of output response data is applied to a laboratory-scale pH neutralization process. The resulting model structure is appropriate and is consistent with a priori physical information.Control structure selection for energy integrated distillation column
http://dl-live.theiet.org/content/conferences/10.1049/cp_19960604
This paper concerns a case study on control structure selection for an almost binary distillation column. The column is energy integrated with a heat pump to transfer heat from the condenser to the reboiler. This integrated configuration renders the possible control structure somewhat different from what is usually seen. Further the heat pump enables disturbances to propagate faster through the system. The plant has six possible actuators of which three must be used to stabilize the system. Hereby three actuators are left for product purity control and/or pressure control. A MILP screening method based on a linear state space model is used to determine an economically optimal set of controlled and also manipulable variables. The generated set of inputs and outputs are analysed with frequency dependent RGA and singular values to determine the best pairing of the variables in terms disturbance rejection and setpoint tracking. The paring and controller design are implemented and evaluated through nonlinear simulation. The suggested control structure is also compared to a control structure applied experimentally.The use of an artificial neural network to improve precision in trace level, quantitative analysis of heavy metal pollutants
http://dl-live.theiet.org/content/conferences/10.1049/cp_19950585
The author has used various neural networks to process the response obtained from an electroanalytical technique used for the analysis of trace metal pollutants in liquids. A previous paper by H.S. Manwaring (1994), compared the capabilities of the GRNN and MLP in this respect. It is shown that using the neural network to make predictions of unknown sample concentrations shows an improvement, by a factor of about two, on the mean absolute error and the prediction confidence when compared with a traditional, calibration curve technique. In addition, the neural network method is shown to produce reliable predictions even with instrumental responses that are completely unsuitable for traditional processing.Distributed sensing of physical and chemical parameters for structural monitoring
http://dl-live.theiet.org/content/conferences/10.1049/ic_19950593
Simultaneous recovery of temperature and strain over a single sensor length has been demonstrated using the information recovered from polarimetric measurements on the LP<sub xmlns="http://pub2web.metastore.ingenta.com/ns/">01</sub> and LP<sub xmlns="http://pub2web.metastore.ingenta.com/ns/">11</sub> modes of polarisation maintaining fibre. Temperature information was recovered to within 2°C and strain recovery was better than 10 μɛ. The measurement method is entirely compatible with distributed measurement methods. A novel generic form of sensor capable of performing distributed measurements on a chemical species has been designed. This sensor has been shown to be an efficient means of detecting water ingress. With the use of suitable OTDR instrumentation, wetted sensor lengths of less than one metre, can be detected and located. The sensor has given meaningful information on the location of voided regions during a cement grouting process. (6 pages)Chlorination control in a large water treatment works
http://dl-live.theiet.org/content/conferences/10.1049/ic_19950738
The introduction of tighter drinking water quality standards and economic considerations has focused attention on ways to improve control systems for water treatment, and for the disinfection process in particular. This control problem has, however, proved difficult to tackle with automatic control systems, due to the long contact time between the water and the disinfectant needed to ensure reliable disinfection, and hence the long delay between a control action and its measurement. The variable quality of raw water, together with the large effect that seasonal changes in ambient temperature and pH have on the dissociation, and thus the disinfecting capability, of dosed chlorine-based disinfectants, compound the problem. This paper describes a simple and cost-effective approach to the control of water chlorination that is being implemented for a large UK water treatment works. (16 pages)Experience with nonlinear control and identification strategies
http://dl-live.theiet.org/content/conferences/10.1049/cp_19940251
Many common and commercially important industrial processes such as distillation columns, chemical reactions, and pH neutralizations exhibit inherently nonlinear behavior. For these control problems, conventional PID controllers must be conservatively tuned in order to ensure closed-loop stability over the full range of operating conditions. Consequently, there are considerable incentives to develop practical control techniques based on nonlinear process models. This paper provides an overview of applications to the pH neutralization process with emphasis placed on the comparison of alternative modeling and control techniques. The author looks at radial basis function models and adaptive control.Study of pH control process using fuzzy modelling
http://dl-live.theiet.org/content/conferences/10.1049/cp_19940144
The inherent nonlinearity of the pH process often renders conventional control difficult. This nonlinearity does however suggests that pH control would be a suitable application area for a fuzzy control system, whose ability to handle nonlinearities is well known. Many fuzzy controllers are of the rule based type where the controllers output response is described by a series of control rules. However, the controller described in this paper is of a fuzzy-model based type which means that a fuzzy model of the process itself is embedded into the predictive control structure. The paper presents the performance of a fuzzy-model based pH controller for a strong base-weak acid neutralisation carried out in a constant volume CSTR. The performance of the fuzzy-model based controller is tested for various set points on the neutralisation curve.A neural network air-fuel ratio estimator
http://dl-live.theiet.org/content/conferences/10.1049/cp_19940127
The paper suggests that a cheap, reliable method of measuring or estimating engine Air-Fuel Ratio (AFR) is needed for effective control. The behaviour of the intake manifold, which is the main cause of the control problem, is discussed, and the use of neural networks for estimating AFR is suggested. The main features of such networks in system modelling are given and the training of two different networks using a simulator is described. The results of tests carried out on the trained networks are given and discussed, and it is concluded that such work deserves further research.On-line control of dissolved oxygen concentration using an automatic tuning PID controller
http://dl-live.theiet.org/content/conferences/10.1049/cp_19940279
With improvements in sensor reliability and computer systems, digital process control has been widely accepted as an effective way of improving the performance of fermentation processes, and of reducing operating costs. Despite the availability of methodical techniques for manually tuning PID controllers, the problem of finding an optimum set of controller parameters for the full operating range of the process can be arduous, time-consuming and prone to mistakes. In general, controllers which have fixed tuning parameter values, such as conventional PID controllers, are considered to be insufficient to cover a wide range of dynamic changes in fermentation processes. Improved performance can be achieved by adaptive techniques which monitor the time-variant dynamics of the process and adjust the controller tuning parameters to continuously optimise the performance of the closed-loop system.LiNbO<sub xmlns="http://pub2web.metastore.ingenta.com/ns/">3</sub> acoustic plate mode sensor for dilute ionic solutions
http://dl-live.theiet.org/content/journals/10.1049/el_19900547
Experimental studies of ZX-LiNbO<sub xmlns="http://pub2web.metastore.ingenta.com/ns/">3</sub> acoustic plate modes as detectors for dilute electrolyte or metal ion solutions are reported which indicate a strong acousto-ionic interaction. The results show an increase in sensitivity and resolution by more than two orders of magnitude compared with previously reported acoustic wave sensors.Optical-fibre network system for air-pollution monitoring over a wide area by optical absorption method
http://dl-live.theiet.org/content/journals/10.1049/el_19790536
A low-loss optical-fibre network system for monitoring air pollution and species concentrations in various environments utilising the differential absorption method is proposed and discussed as a sensitive and economical as well as nonhazardous method. Its feasibility and advantages are also described on the basis of the present state of the art of the relevant technology.Passing the pollution test. Preparing for environment regulations
http://dl-live.theiet.org/content/journals/10.1049/pe_20040106
Over the past few years, pollution legislation has placed tighter and tighter demands on the way operators produce power. In 1996 we saw the Integrated Pollution Prevention and Control (IPPC) Directive. Enforced by the European Commission this includes measures to prevent or reduce emissions in the air, water and land. The year 2000 saw the UK's Environment Agency launching the Pollution Prevent and Control (PPC) regulations. These were created to implement the IPPC directive across the UK, and replaced existing Integrated Pollution Control (IPC) regulations The main thrust of these regulations is to move operators towards greater environmental sustainability and the deadline is March 2006. By this time, operators of existing combustion plants with a rated thermal input of 50MWth or more, as well as some waste-fired plants of at least 3MWth, have to apply for the PPC licence.Building in reliability [factory automation]
http://dl-live.theiet.org/content/journals/10.1049/me_20070208
The prime strategic drivers of factory automation are increased productivity, greater visibility and enhanced control. Gilead Sciences, an international biopharmaceutical company, assures consistent product quality and maximises batch yields, by improving manufacturing process efficiency with reactive, real-time pH measurements using the Foxboro 871PH Series sensor. On the other hand, for the complex task of a litho plate packaging, Klikok-Woodman built a packaging machine around the Rockwell Automation Integrated Architecture concept using Allen-Bradley control systems. Its ControlLogix controller is fitted with a DeviceNet module which receives the size information of an arriving plate and then adjusts the production system to automatically cope with the arriving plate, elegantly solving the problem of dealing with varying product sizes.On controlling pH
http://dl-live.theiet.org/content/journals/10.1049/ip-d.1987.0029
The paper summarises a recent investigation to find out what contribution could be made to pH regulation by modern algorithms for online estimation and/or adaptive control. Topics discussed include a mathematical model to account for nonlinearity; full-scale trials using selftuning controllers; a specially developed nonlinear estimator; use of resulting estimates to drive a suboptimal ‘cautious’ control law; and a specialpurpose CAD package. It is concluded that pH regulation is amenable to good engineering practice, and that whether or not this should include modern algorithms is a problem-specific question which can be answered using the CAD package.Online adaptive control of a fermentation process
http://dl-live.theiet.org/content/journals/10.1049/ip-d.1984.0021
The feasibility of an adaptive controller for the fermentation of baker's yeast is investigated. A simplified model of the process was developed to aid the design of the parameter estimator and multivariable controller. Experiments were performed on a small-scale, fully instrumented fermenter under the control of the online real-time adaptive package. Results are presented to show the successful performance of the adaptive technique achieved by controlling various process variables, in both SIMO and MIMO configurations. The work indicates that the application of the adaptive technique compares favourably with semiautomated control techniques while providing great opportunities for enhancement and thereby superior control. It has been found necessary to omit a considerable amount of the biological detail surrounding the fermentation process so that attention can be focussed upon the adaptive-control applications.Online computer control of pH in an industrial process
http://dl-live.theiet.org/content/journals/10.1049/ip-d.1980.0025
The paper reports on a collaborative, university/industry project to investigate the contribution which modern control technology can make to a difficult industrial problem: pH control. A digital computer was interfaced for online neutralisation of effluent from a full-scale production process. The investigation consisted of trials, some of them extended, using the computer to control neutralisation according to various algorithms and to monitor performance, and of simulation studies. Algorithms investigated include p.i.d., self-tuning control, and optimal-<i xmlns="http://pub2web.metastore.ingenta.com/ns/">k</i>-step-ahead (o.k.) adaptive control. The paper summarises work to date which leads to the conclusion that an online digital computer can contribute substantially to pH control and that o.k. control is a promising candidate for general-purpose feedback within a configuration of combined feedbadk, feeciforwardancilineansationSelf-tuning control of hypotension during ENT surgery using a volatile anaesthetic
http://dl-live.theiet.org/content/journals/10.1049/ip-d.1988.0014
The generalised minimum-variance self-tuning controller of Clarke and Gawthrop has been used to regulate induced hypotension (reduced blood pressure) in 34 patients undergoing ENT surgery, by automatic adjustment of the inhaled concentration of the volatile anaesthetic agent isoflurane. The purpose of the investigation was: to assess the use of self-tuning control (STC) as an aid to the anaesthetist in normal clinical situations; to obtain information in an acceptable manner to illuminate the clinical situation and any signal conditioning, sources of outliers, and other possible problems to be met therein; to see to what extent STC compares with manual adjustments by an anaesthetist (10 patients); to provide a basis for the design of a stand alone controller for use in anaesthesia which would enable comparison with other drugs in the same situations; and to provide controlled conditions for measurement of various physiological variables of interest to anaesthetists. In summarising our investigations and experiences from an engineering standpoint the potential of STC methods in clinical applications is indicated, and some ways of improving the performance of our implementation are discussed.Binary encoded 2nd-differential spectrometry using UV-Vis spectral data and neural networks in the estimation of species type and concentration
http://dl-live.theiet.org/content/journals/10.1049/ip-smt_19970713
An approach to determining the type and concentration of a range of representative contaminants, chlorine, nitrate and ammonia in waste water, based on a three-stage scheme for processing data from ultraviolet and visible (UV-Vis) spectra, is described. In simulation in the laboratory, data for the study are derived from laboratory-based measurements of such spectra from mixtures of common chemical pollutants in water at levels around their legal limits and from mathematical models based on these measurements. Through the work, it is concluded that mathematical procedures alone, i.e. self-learning, are not currently effective, while classification based on a model for absorption spectra with prior knowledge of the expected chemistry in a particular water system under study, is more likely to be successful.Data fusion and artificial neural networks for biomass estimation
http://dl-live.theiet.org/content/journals/10.1049/ip-smt_19970887
The ability of artificial neural networks (ANNs) to learn from experience rather than from mechanistic descriptions makes them the preferred choice to model processes with intricate variable interrelations. Some of these processes can be found in the area of biotechnology. The authors aim to use ANNs and data fusion to provide better instrumentation for a fermentation process and eventually optimise its performance. Of particular interest is the robust estimation of biomass in the production of an antibiotic. Several feed-forward backpropagation neural networks (BPNs) have been chosen for the experiments using the Levenberg–Marquardt learning algorithm. Work has been carried out to test the generalisation capabilities and performance in the presence of noise and sensor failure. It has been observed that, given the appropriate training, data fusion and ANN methodology lead to estimation of these parameters with an accuracy comparable to instrumentation errors.Near-optimal productivity control of a continuous bioreactor
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_19952166
Near-optimal productivity control of a continuous bioreactor by a conventional proportional-integral controller poses stability problems, and certain nonlinear controllers yield excessive variation of the manipulated variable. A nonlinear controller based on the Hammerstein model is proposed to overcome such problems. The performance of the proposed nonlinear control is compared with that of the nonlinear controllers proposed by Henson and Seborg.Design and analysis of neural/fuzzy variable structural PID control systems
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_19960261
The paper describes the design method of a neural/fuzzy variable structural proportional-integral-derivative (neural/fuzzy VSPID) control system. The neural/fuzzy VSPID controller has a structure similar to that of the conventional PID. In this controller, the PD mode is used in the case of large errors to speed up response, whereas the PI mode is applied for small error conditions to eliminate the steady-state offset. A sigmoidal-like neuron is employed as a preassigned algorithm of the law of structural change. Meanwhile, the controller parameters would be changed according to local conditions. Bounded neural networks or bounded fuzzy logic systems are used for constructing the nonlinear relationship between the PID controller parameters and local operating control conditions. Flexible changes of controller modes and resilient controller parameters of the neural/fuzzy VSPID during the transient could thereby solve the typical conflict in nature between steady-state error and dynamic responsiveness. A neutralisation process is used to demonstrate the applicability of such a controller for controlling highly nonlinear processes.pH control: handling nonlinearity and deadtime with fuzzy relational model-based control
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_19971139
The application of fuzzy logic to the design of nonlinear controllers has become increasingly popular in recent years. Most of the developments have been in controllers of the rule-based type. An alternative approach, and one which reflects trends in conventional control, is to use fuzzy logic to build a process model, and then to incorporate this into a standard model-based controller scheme. The paper proposes the application of fuzzy relational models (FRMs) for the non-linear control of a pH process. The pH in both a simulated and a laboratory continuously stirred tank reactor (CSTR) was controlled by a model predictive controller (MPC), incorporating a fuzzy model created using a recently developed method of FRM identification. The controller performance is compared with that of a fuzzy rule-based controller, that of a PID controller and that of a linear MPC. The comparison shows the superiority of fuzzy relational model-based control (FRMBC) for highly nonlinear processes. The suitability of the FRMBC for real-world applications is demonstrated by its control performance on a laboratory-scale plant.Fuzzy neural networks for nonlinear systems modelling
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_19952255
A technique for the modelling of nonlinear systems using a fuzzy neural network topology is described. The input space of a nonlinear system is initially divided into a number of fuzzy operating regions within which reduced order models are able to represent the system. The complete system model output, the global model, is obtained through the conjunction of the outputs of the local models. The fuzzy neural network approach to nonlinear process modelling provides a way of opening up the purely 'black box' approach normally seen in neural network applications. Process knowledge is used to identify appropriate local operating regions and as an aid to initialising the network structure. Fuzzy neural network models are also easier to interpret than conventional neural network models. The weights in a trained fuzzy network model can be interpreted in terms of process information. This technique has been applied to model the nonlinear dynamic behaviour of a pH reactor and two static nonlinear systems. Correlation based tests are used to assess the fuzzy network model validity for nonlinear dynamic systems.Adaptive stabilisation of ethanol production during the continuous fermentation of <i xmlns="http://pub2web.metastore.ingenta.com/ns/">Saccharomyces cerevisiae</i>
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_20030872
Two different approaches for adaptive control design are investigated and compared from a user's point of view. The first one is an application of the Matlab toolbox routine Designer, and the second approach is based on the commonly used adaptive linearising control (ALC) design method. The continuous aerobic fermentation of <i xmlns="http://pub2web.metastore.ingenta.com/ns/">Saccharomyces cerevisiae</i> is taken as an object for control. The control task is the stabilisation of ethanol production at a previously set low value of concentration. Simulations using a biochemical model of the process verify both algorithms. In general the obtained results for both algorithms satisfy the microbiologists' requirements. From a user's point of view, Designer shows more advantages then the ALC. The use of Designer seems to be promising for application in the bio-industry.Nonlinear and direction-dependent dynamic process modelling using neural networks
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_19960061
The paper discusses several methods of modelling complex nonlinear dynamics using neural networks. Particular reference is made to the problem of modelling direction-dependent relationships. A typical example of this would be top product composition control in a distillation column, where it is easier (i.e. faster) to make the product less pure than it is to make it more pure by an equivalent amount. Recurrent neural networks are identified as a potential method of modelling this type of relationship. The particular architecture chosen for this example is referred to as ‘semirecurrent’, since only past values of the predictions of the network are fed back to the input layer. This architecture is successfully used to model direction-dependent relationships in both simulated and actual industrial process data.Wiener model identification and predictive control of a pH neutralisation process
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_20040438
Wiener model identification and predictive control of a pH neutralisation process is presented. Input-output data from a nonlinear, first principles simulation model of the pH neutralisation process are used for subspace-based identification of a black-box Wiener-type model. The proposed nonlinear subspace identification method has the advantage of delivering a Wiener model in a format which is suitable for its use in a standard linear-model-based predictive control scheme. The identified Wiener model is used as the internal model in a model predictive controller (MPC) which is used to control the nonlinear white-box simulation model. To account for the unmeasurable disturbance, a nonlinear observer is proposed. The performance of the Wiener model predictive control (WMPC) is compared with that of a linear MPC, and with a more traditional feedback control, namely a PID control. Simulation results show that the WMPC outperforms the linear MPC and the PID controllers.Fuzzy neural networks and application to the FBC process
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_19960258
Implementation of fuzzy systems using multilayer perceptron artificial neural networks is considered, and detailed equations for training the network are derived. Models for the fluidised bed nitrogen emissions are created using linear regression, multilayer perceptron neural networks, and fuzzy neural networks to illustrate and compare the behaviour of the algorithms.Guaranteed tracking and regulatory performance of nonlinear dynamic systems using fuzzy neural networks
http://dl-live.theiet.org/content/journals/10.1049/ip-cta_19990499
A new technique for the design of stable tracking and regulatory control systems for nonlinear systems using fuzzy neural networks is described. A class of nonlinear systems is considered where a few of the input variables can be manipulated (controlled) while the rest are considered as disturbances. System dynamics are modelled using a fuzzy neural network where each fuzzy operating region is associated with a series–parallel linear model. Two new control strategies are proposed using the Lyapunov synthesis approach. In one, a disturbance invariant control scheme is proposed where the sensitivity of the system response with respect to the control variable is estimated using fuzzy neural model. In the other a model predictive scheme is developed in which disturbances are predicted online and the fuzzy neural model is used to predict the control action for a desired setpoint. The proposed controllers have been implemented for a nonlinear pH reactor through simulation. In a pH control problem the acid and buffer flow rate are considered to be disturbances, while the base flow rate is controlled to maintain a constant pH. Simulation results show that the proposed scheme provides guaranteed tracking and regulatory performance.