Jens G. Balchen

Dynamic Optimization and Production Planning of Thermal Cracking Operation

Abstract This work addresses the dynamic optimization of the production period of thermal crackers with respect to coke formation in the cracking coil and transfer line exchanger. Optimal time-dependent trajectories to feed rate, steam to hydrocarbon ratio, and reaction severity are calculated. The net earnings based on the price of hydrocarbons, fuel, steam, decoking, and maintenance cost are maximized. All important operational constraints are included and the optimization problem is solved using parametrized free variable trajectories (piece-wise constant) and a standard SQP package. Rigorous distributed physical models are used and calculation show that dynamic optimization gives up to 2% higher net earnings (earnings–expenses) than conventional steady-state optimization performed on the same models. This is in the same range as earnings reported from steady-state optimization implementations alone.

How have we arrived at the present state of knowledge in process control? Is there a lesson to be learned?

Abstract The utilization of the results of control theory in the process control field has been lagging behind other application fields such as aerospace by many years. It is argued that the availability of high capacity computing at low price will change this situation and that new powerful control techniques can now be implemented in process control.

Elementary nonlinear decoupling control of composition in binary distillation columns

Abstract Elementary nonlinear decoupling (END) is a model based control algorithm intended to decouple and linearize a nonlinear multivariable process in order to achieve better control than can be obtained by conventional decentralized linear feedback control. The application of END to the composition control of a theoretical binary distillation column illustrates that the quality achievable is very high.

Mathematical and Numerical Modeling of Physical and Biological Processes in the Barents Sea

A short review is given of a long range program which has been started in Norway to develop mathematical and numerical models of the physical, chemical and biological processes in an ocean in order to be able to estimate past, present and future states of the system. Included in the term state are the least number of quantities which give an adequate description of the total system behaviour in terms of growth, migration and distribution of both plankton and fish. An important aspect of the program is to develop techniques for recursive updating of the states and parameters of the models against data received from measurements taken in the real system. The final goal is an operational set of models in a dedicated computer system which can be used in planning the utilization of the resources, investigating consequences of different management strategies and in planning fisheries operations. The program started in 1975 and the first system is planned to operate in 1984.

Control of a Steam Boiler by Elementary Nonlinear Decoupling (END)

The paper describes the simulated application of the END-algorithm (Balchen 1998) for model-based, nonlinear, multivariable control of a steam boiler process. The choice of the tuning parameters for the optimization criterion is discussed in some detail.

Dynamic positioning of floating vessles based on Kalman filtering and optimal control

This paper describes computer-based, dynamic positioning system for floating vessels. The system is based on a detailed mathematical model of vessel motion in response to forces from thrusters, wind, waves and water current. The system uses a Kalman filter for optimal estimation of vessel motions and environmental forces from wind, waves and current. The control system is based on feedback from the motion variables where the oscillatory, wave-induced component is removed by the estimator. Feedback from the water current estimate provides the integral action of the system and feed forward from the wind force estimates are implemented. Simulation results and recordings from actual operation of the system indicate an excellent system performance. Reference is given to installations made on actual vessels.

Application of CAD in modeling, identification and control of industrial and large scale, nontechnical processes

This paper gives a short description of the CAD facilities available at the University of Trondheim, Division of Engineering Cybernetics and presents experiences and results from an operation period of almost ten years. The CAD facilities consist of an interactive program library, CYPROS, which contains general programs for simulation, identification, parameter estimation and control system design and a specialized CAD package for handling large scale systems. The programming language is standard Fortran IV and the packages are implemented on a medium size computer system where the user-program communication is based upon numerical terminals and video color graphics.

Elementary Nonlinear Decoupling (END), A General Approach to Model Based Control of Nonlinear Multivariable Processes

The paper gives a survey of a strategy based on Elementary Nonlinear Decoupling (END) used for the control of nonlinear multivariable processes. It describes the structure of the solution and the different optimization procedures leading to the final implementable control strategy. Based on the END strategy it is possible to develop a convenient systematic method for handling constraints both in the control variables (inputs) and state variables. Application of the END strategy has been demonstrated by the simulation of a number of realistic processes such as a distillation column (Balchen and Sandrib 1995a), a fluidized catalytic cracker (FCC) (Balchen and Sandrib 1995b) and a steam boiler (Balchen and Larsen 1998).

Computer Aided Modelling of Dynamic Processes Based on Elementary Physics

Abstract This paper presents an automated approach for establishment of dynamic state space models for unit processes. Typical examples of such unit processes are separators, heat exchangers and various reactors. The emphasis is put on processes where the behaviour is dominated by phenomena like fluid flows, phase shifts and chemical reactions. A rule based computer program prototype has been developed to perform the modelling task, including selecting conservation equations, identifying possible transfer phenomena and expressing transfer phenomena mathematically. An example demonstrating the modelling of an evaporator is included.