8 December 2020 in IEEE Access
Object recognition has been widely investigated in computer vision for many years. Currently, this process is carried out through neural networks, but there are very few public datasets available with mask and class labels of the objects for the training process in usual applications. In this paper, we address the problem of fast generation of synthetic datasets to train neural models because creating a handcraft labeled dataset with object segmentation is a very tedious and time-consuming task. We propose an efficient method to generate a synthetic labeled dataset that adequately combines background images with foreground segmented objects. The synthetic images can be created automatically with random positioning of the objects or, alternatively, the method can produce realistic images by keeping the realism in the scales and positions of the objects. Then, we employ Mask-RCNN deep learning model, to detect and segment classes of kitchen objects using images. In the experimental evaluation, we study both synthetic datasets, automatic or realistic, and we compare the results. We analyze the performance with the most widely used indexes and check that the realistic synthetic dataset, quickly created through our method, can provide competitive results and accurately classify the different objects.
25 July 2023 in IEEE Access
The performance of induction cooktops (inductor and electronic power converter) is closely related to the inductive properties of the cookware used. Traditionally, they are designed to work optimally with fully ferromagnetic cookware. However, multimaterial cookware with a bottom surface composed of a ferromagnetic material and small pieces of aluminum are nowadays more commonly used. Designing cooktops to also work optimally with this cookwares requires identifying the equivalent parameters, R eq and L eq , obtained from the electromagnetic models of the inductor-load system. Modeling multimaterial cookware, with a large number of inserts, complicates the model design and increases its computational cost. This work presents an equivalent model for multimaterial loads that allows the evaluation of the inductor performance and obtaining the equivalent parameters quickly and accurately. The multimaterial cookware is modeled as a disk of uniform material with equivalent electromagnetic properties, μ r,eq and σ eq , which depend on the percentage occupied and the properties of each material. The equivalent model has been validated by electromagnetic simulation using a FEA tool and by experimental results. Finally, based on the results obtained, an analysis has been carried out to evaluate the importance of design factors such as the choice of the cookware base material and the size, arrangement, and number of inserts.
23 May 2023 in IEEE Access
Induction cooktops are a key element in home electrification. This paper presents a new use of induction cooktops, which allows the preparation of infusions directly in a ceramic mug. It is proposed the design of a water heater with an immersible metallic load. The dimensions of the load allow its embedding in small ceramic vessels. Moreover, thanks to its specific shape it can efficiently receive the energy from induction cooktops and heat the liquid quickly despite being smaller than conventional inductors. The geometry of the immersible load consists of a flat disk with sidewalls on its edge, which provides low reluctance paths for the magnetic field, resulting in improved transmitted power and efficiency. The load is arranged in a multilayer structure consisting of a ferromagnetic alloy with low Curie temperature at the bottom and aluminum on the top, which provides self-detection capabilities against dangerous overheating. The study of the proposal has been carried out with a finite element analysis model, and the feasibility of this proposal has been demonstrated through experimental results of its performance. Finally, the immersible water heater, powered by an induction cooktop, has been compared to other liquid heating systems, such as the microwave and the kettle, with promising results.
7 August 2025 in IEEE Industrial Electronics Magazine
Creating a sustainable future is a transdisciplinary challenge where industrial electronics (IE) plays a key role by providing environmentally friendly solutions to daily processes. Domestic device electrification is a powerful tool for sustainable energy consumption and indoor air quality improvement. In this context, this article covers a relevant example of household appliance electrification with improved energetic performance: domestic induction heating. Domestic IH has now been on the market for 50 years. This article reviews the main historical milestones in electrical cooking appliances, and presents the enabling technologies and advances behind the success of current IH technology. Additionally, recent technological development advancements and future trends are introduced showing the potential of IH to help pave the path to create a sustainable future.
22 December 2022 in IEEE Access
Security aspects in the household sphere have become a major concern in modern societies. In particular, regardless of the technology used, users increasingly appreciate a protection system to prevent material damage in the case of human errors or distractions during the cooking process. This paper presents a sensorless method for detecting and limiting overtemperature, unique to induction cooktops, based on their specific features, such as automatic pot detection and load power factor estimation. The protection system exploits the change in the load material properties at certain temperatures, the effect of which may be enhanced by arranging a multilayer structure comprising a low Curie temperature alloy and an aluminum layer. The proposed multilayer load exhibits two differentiated states: a normal state, where the cookware is efficiently heated, and a protection state, above the safety temperature, where the power factor abruptly decreases, limiting the overheating and making the state easily detectable by the cooktop. This method of overtemperature self-protection uses the electronics of conventional induction cooktops; therefore, no other sensors or systems are required, reducing its complexity and costs. Simulation and experimental results are provided for several cookware designs, thereby proving the feasibility of this proposal.
22 November 2024 in IEEE Access
Food cooking process are complex dynamical systems to model. In the state of the art we find that a good solution consists of physics-based finite element models (FEM). FEM models, although being very accurate, have a high computational cost making them unfeasible for real-time applications. To solve this problem, we consider neural networks (NN) trained from FEM simulations. Specifically, we propose a Nonlinear AutoRegressive with eXogenous inputs Neural Network (NARX-NN). The main novelty is that we define a novel training algorithm adapted to the modeling of real-time dynamical systems, allowing a NARX-NN with a simple structure to obtain a negligible error compared to the results of the original FEM model. The NARX-NN trained with the proposed training algorithm obtains an R-squared of more than 0.99 in the test simulations, while the same NARX-NN trained with the standard training algorithm obtains an R-squared of 0.78 in the same tests. The proposed NARX-NN achieves a speedup of 8 orders of magnitude compared to the original FEM model. Moreover, the developed NN is able to predict the complete cooking of the food in a few milliseconds without the need of external sensors. Alternatively, our approach can also be used in real time with information captured with sensors. The presented methodology is highly scalable and could be adapted to different types of food and cooking processes, as well as to other dynamical systems in general.