Electromagnetic radar has been used for detecting internal defect in concrete structure, since the presence of a defect, such as a crack, causes changes in the scattered waves. However, visually identifying scattering difference due to a defect requires a high level of skill, and it takes a lot of effort when there is a large amount of observation data. Therefore, the development of automatic identification technology using machine learning (ML) is underway. However, the training data collection method and the performance of air-gap detection using ML has not been explored in detail. In this paper, firstly, we investigate the detection of an infinite air-gap parallel to the surface of a concrete slab using artificial neural network (ANN) identification technology in layered medium models, and discuss how to train high performance ANN with less training data. Secondly, we study the characteristic of ANN detection for an oblique air-gap using 2D homogeneous medium models.

Solving the inverse scattering problem of determining the position, shape, and electrical parameter distributions of a target object requires information not only on the scattered waves but also on the incident waves. Almost all inverse scattering methods assume that the incident wave is known on the observation curve and in the estimation region that includes the object. However, it is sometimes difficult to remove the object and measure the incident wave. To address this situation, a method to estimate the incident field from the total electromagnetic field data (the sum of incident and scattered wave fields) measured on the observation circle has been proposed in recent years. In measurements outside an anechoic chamber, the incident wave to an object is a composite of the electromagnetic wave to probe the object and the unwanted electromagnetic wave from mobile phones and other sources. In this paper, we apply this method to the extraction of incident wave information from total electromagnetic field data measured on the observation circle in the presence of unwanted electromagnetic waves, and also propose a method to extract incident wave information from total electric field data alone. The influence of noise will also be discussed.

This paper describes an effect of transient electromagnetic fields on the germination of seed plants. In our experiment, transient electromagnetic fields by spark discharges were applied to seeds of asparagus, parsley, carrot, lettuce and Komatsuna. In this experiment, the pulse widths of discharge current (800 A) were set to 3 µs, and the number of applications were set to 5 times, 10 times, 20 times and 30 times, respectively. As a result, we found that 1) the germination rate tended to increase within a certain range of the number of applications, 2) the germination rate decreased when the number of applications exceeded a certain number.

The electrical tree is degradation traces of insulating materials. The mechanism of this electrical tree progress is not yet understood. Therefore, we wondered if machine learning could be used to elucidate the progress mechanism. First, we investigated the applicability of machine learning to electrical tree.