This article highlights the industry experience of the development and practical implementation of a short-term photovoltaic forecasting system based on machine learning methods for a real industry-scale photovoltaic power plant implemented in a Russian power system using remote data acquisition. One of the goals of the study is to improve photovoltaic power plants generation forecasting accuracy based on open-source meteorological data, which is provided in regular weather forecasts. In order to improve the robustness of the system in terms of the forecasting accuracy, we apply newly derived feature introduction, a factor obtained as a result of feature engineering procedure, characterizing the relationship between photovoltaic power plant energy production and solar irradiation on a horizontal surface, thus taking into account the impacts of atmospheric and electrical nature. The article scrutinizes the application of different machine learning algorithms, including Random Forest regressor, Gradient Boosting Regressor, Linear Regression and Decision Trees regression, to the remotely obtained data. As a result of the application of the aforementioned approaches together with hyperparameters, tuning and pipelining of the algorithms, the optimal structure, parameters and the application sphere of different regressors were identified for various testing samples. The mathematical model developed within the framework of the study gave us the opportunity to provide robust photovoltaic energy forecasting results with mean accuracy over 92% for mostly-sunny sample days and over 83% for mostly cloudy days with different types of precipitation.

Secure communication has a wide range of interesting techniques.Here, a wavelet decomposed data communication packet is used, whose advantages are described. The problems of wide spectrum noise-based jamming, by hostile elements is counteracted. For this, several wavelet coefficients of a data packet are transmitted in non-harmonically related spread frequencies. The reconstruction of the wavelets, despite noisy detection at the received end, is shown to be better, typically for test data packets. The number of frequencies and the coefficients can be varied to suit the detection error limits. Several wavelets have been tired for this purpose and testing has been done even with image data. The results are found to be useful and interesting.

Without building new technology and involved hardware circuit boards, how to develop industrial Artificial Neural Network (ANN) based robotic projects for enterprises is explained with special reference to GSM cell and robotic models. In the process of training to develop enterprise robots, one has to make use of all possible resources. Without the need for development of new circuits or methodology, existing devices, components and services can be used to develop a standard robotic project. There are many examples like traffic control, agricultural plantation and adaptive surveillance. The groundwork of technology for the above robot data collection and control is elucidated and schemes using simple readily available hardware for easy and quick rigging up of robotic systems is the theme of this paper.

No abstract

Unattended locations today require detection of any mal-intentional intrusion. Sensors of the Pyroelectric (PIR) orMicrowave reflecting (MW) types are commonly available for detecting intrusion by a human. In this paper, a digital video camera is interfaced to a computer, say a simple laptop, which has provision for wireless USB based Ethernet connection. The program developed monitors a site in view of the camera and positively detects an intrusion by video frame based evaluation. A novel and sensitive algorithm for detection of motion from video frames is based on integrating the pixel intensities in successive finite difference method over a set of circular and way contour paths in the field of view of the camera. The proposed method is very fast compared to area pixel methods thus far available. The detected intrusion and its location within the scene are instantly messaged through the wireless cellular network adaptor, such as the "Tata Photon" using a simple "Compaq" laptop. The image itself follows the detection. Thus, this method is cheaper than the expensive sensing devices based on MW or PIR.