A System Using Artificial Intelligence to Detect and Scare Bird Flocks in the Protection of Ripening Fruit.

Flocks of birds may cause major damage to fruit crops in the ripening phase. This problem is addressed by various methods for bird scaring; in many cases, however, the birds become accustomed to the distraction, and the applied scaring procedure loses its purpose. To help eliminate the difficulty, we present a system to detect flocks and to trigger an actuator that will scare the objects only when a flock passes through the monitored space. The actual detection is performed with artificial intelligence utilizing a convolutional neural network. Before teaching the network, we employed videocameras and a differential algorithm to detect all items moving in the vineyard. Such objects revealed in the images were labeled and then used in training, testing, and validating the network. The assessment of the detection algorithm required evaluating the parameters precision, recall, and F1 score. In terms of function, the algorithm is implemented in a module consisting of a microcomputer and a connected videocamera. When a flock is detected, the microcontroller will generate a signal to be wirelessly transmitted to the module, whose task is to trigger the scaring actuator.

A Sensor System for Detecting and Localizing Partial Discharges in Power Transformers with Improved Immunity to Interferences.

The paper reports on the solution, principles, and application results related to a system for diagnosing main transformers in power plants via the radiofrequency method. The subject of the diagnostics is the occurrence of partial discharge activity in transformers. The technical solution of the system is characterized in the introductory section of the article. There then follows a description of the operating principle and the implemented novel advanced methods for signal detection and source localization. The results obtained from practical application of the system within the diagnostics of high-power transformers are presented as well. Because ambient electromagnetic disturbance was recognized as a major issue during the system development, novel detection methods were proposed, implemented, and verified. The principal approach utilizes an external radiofrequency sensor to detect outer impulse disturbance and to eliminate disturbance-triggered acquisitions, and it also ensures direct real-time visualization of the desired impulse signals. The ability of weak signal detection was verified via artificial impulse signal injection into the transformer. The developed detection methods were completed with localization techniques for signal source estimation. The desired impulse signal was detected and localized during full operation of the main transformer, despite the presence of strong electromagnetic interference.