Johannes Jäschke

Dynamic Model and Control of Heat Exchanger Networks for District Heating

The various governmental policies aimed at reducing the dependence on fossil fuels for space heating and the reduction in its associated emission of greenhouse gases such as CO2 demands innovative measures. District heating systems using residual industrial waste heats could provide such an efficient method for house and space heating. In such systems, heat is produced and/or thermally upgraded in a central plant and then distributed to the final consumers through a pipeline network. In this work two main objectives will be considered: the first is to create a dynamic model which can represent the main characteristics of a district heating network and the second one is to design a non-linear model predictive controller (NLMPC) to satisfy the heat demands of the consumers in the heat exchanger network. As the model predictive controller is based on minimizing an objective function, it is totally perfect to find the way to reduce the superfluous energy consumption and make the best of using the freely applicable industrial waste heats. Beside this environmental aspect, reducing the invested energy consumption can reduce the operational costs.

Self-optimizing control – A survey

Abstract Self-optimizing control is a strategy for selecting controlled variables. It is distinguished by the fact that an economic objective function is adopted as a selection criterion. The aim is to systematically select the controlled variables such that by controlling them at constant setpoints, the impact of uncertain and varying disturbances on the economic optimality is minimized. If a selection leads to an acceptable economic loss compared to perfectly optimal operation then the chosen control structure is referred to as “self-optimizing”. In this comprehensive survey on methods for finding self-optimizing controlled variables we summarize the progress made during the last fifteen years. In particular, we present brute-force methods, local methods based on linearization, data and regression based methods, and methods for finding nonlinear controlled variables for polynomial systems. We also discuss important related topics such as handling changing active constraints. Finally, we point out open problems and directions for future research.

Control structure selection for optimal operation of a heat exchanger network

We consider the control structure design for a heat exchanger network (HEN), where a stream is split into parallel lines which are heated individually before they are merged together again. The objective is to find a control structure which maximizes the final temperature. We consider two scenarios, where (Scenario 1) the flow rates and the heat transfer coefficients are considered as disturbances, and (Scenario 2) where the hot stream temperatures are treated as additional disturbances. In both scenarios it is found that controlling linear measurement combinations gives very good performance, and that including flow measurements in the combinations gives little advantage over using only combinations of temperature measurements.

Self-optimizing invariants in dynamic optimization

In optimal control, the input trajectories are often solved numerically or analytically. This requires that all variables which enter the optimality conditions are known or measured. We use techniques from polynomial elimination theory to eliminate variables which are not known from the optimality conditions. The result is an expression of the optimality conditions in known variables only, which can easily be evaluated and controlled by feedback.

Using Process Data for Finding Self-Optimizing Controlled Variables

Abstract In the process industry it is often not known how well a process is operated, and without a good model it is difficult to tell if operation can be further improved. We present a data-based method for finding a combination of measurements which can be used for obtaining an estimate of how well the process is operated, and which can be used in feedback as a controlled variable. To find the variable combination, we use past measurement data and fit a quadratic cost function to the data. Using the parameters of this cost function, we then calculate a linear combination of measurements, which when held constant, gives near-optimal operation. Unlike previously published methods for finding self-optimizing controlled variables, this method relies only on past plant measurements and a few plant experiments to obtain the process gain. It does not require a model which is optimized off-line to find the controlled variable.

Optimal Operation by Controlling the Gradient to Zero

Abstract From an optimization point of view, the gradient is the key variable which gives information about the optimality of a process. In this paper we present how the gradient is related to the loss from optimality, and show how determining a good set of controlled variables can be considered as weighted approximation of the gradient. We show that even if there are setpoint changes for the controlled variables, this can still be considered as approximating the gradient.

Self-optimizing Control and NCO tracking in the Context of Real-Time Optimization

Abstract This paper reviews the role of self-optimizing control (SOC) and necessary conditions of optimality tracking (NCO tracking). We argue that self-optimizing control is not an alternative to real-time optimization (RTO), NCO tracking or model predictive control (MPC), but is to be seen as complementary. In self-optimizing control we determine controlled variables (CV), that keep the process close to the optimum when a disturbance enters the process. These CVs are controlled at their setpoints using PID or model predictive controllers. Preferably, the setpoints are kept constant, but they may also be adjusted using RTO or NCO tracking. In any case, a good choice of CVs will reduce the frequency of setpoint changes by RTO or NCO tracking. When selecting self-optimizing CVs, a set of disturbances has to be assumed, as unexpected disturbances are not rejected in SOC. On the other hand, RTO and NCO tracking adapt the inputs at given sample times without any assumptions on what disturbances occur. It is only assumed that they occur on a slower time scale than the sampling. Disturbances with high frequencies or which which do not lead to a steady state are not rejected optimally. By using NCO tracking in the optimization layer and SOC in the control layer below, we demonstrate that the advantages of both methods complement each other. This combination allows fast optimal action for the expected disturbances, while other disturbances are compensated by NCO tracking on a slower time scale.

Optimal operation of a waste incineration plant for district heating

A case study of a waste incineration plant operating close to optimality by using simple feed-back control schemes is presented. Using off-line optimization the structure of the optimization problem is exploited and a set of variables is found, such that if the process is controlled with those variables are at their setpoints, operation is near-optimal.

Dynamic online optimization of a house heating system in a fluctuating energy price scenario

Abstract We consider dynamic optimization of the energy consumption in a building with energy storage capabilities. The goal is to find optimal policies which minimize the cost of heating and respect operational constraints. The main complication in this problem is the time-varying nature of the main disturbances, which are the energy price and outdoor temperature. To find the optimal operable policies, we solve a moving horizon optimal control problem assuming known disturbances. Next, we proposed simple implementation based on feedback control, which gives a near-optimal operation for a range of disturbances. The methods were successfully tested in simulations, which show that there is a great economical gain in using dynamic optimization for the case of variable energy price.

A Self-Optimizing Strategy for Optimal Operation of a Preheating Train for a Crude Oil Unit

Abstract We apply a recently developed approach for optimizing heat exchanger networks with stream splits to the case study of a preheating train of the crude oil unit at the Mongstad Refinery in Norway. To maximize heat transfer, we adjust the split such that the “Jaschke Temperatures” assume equal values for each branch. For a branch with one heat exchanger, the Jaschke Temperature is calculated as T J = ( T − T 0 ) 2 T h − T 0 where T is the temperature of the split stream at the heat exchanger exit, and T h and T 0 are inlet temperatures of the hot and cold stream, respectively. Controlling the Jaschke Temperatures to equal values gives near-optimal operation despite varying flow rates, stream temperatures, and heat transfer coefficients. We fitted a model to plant data obtained from the refinery, and consider two cases with decentralized PI control, and one case where the Jaschke Temperatures are controlled by a model predictive controller. Our paper demonstrates that controlling the Jaschke Temperatures of each branch to equal values is a simple alternative to online real-time optimization methods. Moreover, it is significantly cheaper to implement as an online optimizer, and it is easier to maintain.