Skip to main content
Crossing and bending beams. Photo MostPhotos.

Doctoral defence of Marzieh Hosseini, MSc, 26.5.2023: Moisture control in microwave drying process using electrical capacitance tomography

The doctoral dissertation in the field of Technical Physics will be examined at the Faculty of Science, Forestry and Technology, Kuopio Campus.

What is the topic of your doctoral research? Why is it important to study the topic?

My thesis topic is designing a moisture controller for a microwave drying process while the moisture is estimated using an electrical capacitance tomography (ECT) sensor. Microwave drying has shown to be a profoundly beneficial technology compared to conventional thermal drying methods since it has the features of volumetric and selective heating, resulting in reduced drying time and energy consumption. The product quality in this process, or generally any drying process, is highly dependent on the moisture level of the dried material. However, this process is typically operated based on experience in the industry. Employing a moisture controller can ensure the desired moisture level in the product by adjusting the input microwave power. However, real-time information on the moisture distribution of the material is required for the closed-loop control. Electrical capacitance tomography is a relatively cheap, non-destructive, contactless technique that has proven to be suitable for estimating the material permittivity distribution, which is correlated with moisture.

What are the key findings or observations of your doctoral research?

During my research, an ECT sensor is designed to estimate the moisture distribution of moving polymer foams in a microwave drying process. The designed sensor has a large rectangular structure, unlike typical ECT sensors with round shapes, which imposes particular challenges in image reconstruction. Additionally, a process model is derived for the studied application with two different strategies: First principle modeling and a system identification approach. Finally, the derived process models are employed to design a moisture controller that receives the estimated moisture by ECT as feedback to control the moisture content of the dried polymer foam.

How can the results of your doctoral research be utilised in practice?

The combined process tomography and process control approach proposed in this thesis is a significant improvement in the microwave drying industry. The process no longer needs to be operated blindly, and valuable information on the moisture distribution of the material is available in real-time. Knowledge of moisture distribution is particularly important since having a non-uniform distribution can reduce the final product quality or even result in losing the whole product. Furthermore, the required microwave power to achieve the set point moisture can be automatically calculated using the moisture controller based on the current moisture content of the material. Another critical significance of the proposed tomography-based controller is that it can be further applied to other operations with some modifications and adjustments, and it is not limited to the microwave drying process. Such applications include inline fluid separation, continuous casting, and batch crystallization.

What are the key research methods and materials used in your doctoral research?

The ECT sensor consists of several electrodes on a frame around the polymer foam. The difference imaging technique is employed to use the capacitance measurements from these electrodes to reconstruct a 2D image of the material permittivity distribution correlated with moisture. In the next step, a model of the microwave drying process is derived based on a Luikov model of heat and moisture transfer, which consists of a pair of coupled parabolic PDEs. The PDE-based process model is then discretized spatially using the finite element method and converted into ODEs, which in turn are employed to design a linear quadratic regulator (LQR) to control the average moisture of the polymer foam in a simulated microwave drying process. However, as this model is computationally expensive, another process model using the system identification approach is derived using the input-output data of the process. The input-output datasets are collected by applying standard inputs to the process and recording the corresponding ECT measurements. The derived model from this approach which is a state-space model, is utilized to develop the moisture controller. A proportional-integral (PI) controller and a linear quadratic Gaussian (LQG) controller are developed independently and implemented in the actual process to receive the estimated permittivity of the foam from the ECT sensor and adjust the power level of the microwave sources accordingly.

For more information, please contact:

Marzieh Hosseini,, tel. 050 514 5317