Intelligent measuring is related to smart systems and automated measurements of weather conditions or solar irradiance, to help the manufacturer to optimize solar cell features. Technology should strive to minimize the impact of human activity upon the environment, developing greener solutions for energy consumption. In this paper, we propose a prototype for a wavelet based system, which should analyze several parameters collected by a photovoltaic system - using the Stationary Wavelet Transform. We aim to develop a general mathematical model, which should allow us to improve the manufacturing process of the solar modules. The Stationary Wavelet Transform and several mother wavelets families were tested. The most accurate results were obtained with the Daubechies mother wavelets.
Today, the emergence of new industrial values and the integration of advanced technologies have ushered in a significant evolution in industry through the implementation of Industry 4.0 strategies. As Taylor noted in 1911, a production system must be analyzed starting from its elementary processes. Building on this idea, this paper proposes a control system with artificial intelligence at a foundational level of production systems, specifically focusing on CNC machine tools. Two types of artificial intelligence control systems are proposed for manufacturing processes: one aimed at increasing processing accuracy and the other designed to reduce processing costs or enhance productivity
This paper presents the research and results obtained by authors regarding the modeling of the reactor of the fluid catalytic cracking unit (FCCU), using various machine learning techniques based on supervised learning algorithms Compared with the analytic mathematic modeling methods, machine learning techniques (ML) can solve the difficulties in FCC process modeling, such as strong correlation between variables, nonlinearities, the complexity of kinetic models, and feedstock complexity. The ML method establishes a mathematical correlation between the input and output variables of the process using industrial data. Five supervised learning machine methods are used to model the reactor: random forest regression (RFR), decision trees regression (DTR), k-nearest neighbors (KNN), support vector machines (SVM) and gradient boosting regression (GBR).
During the mastication process, the restorative materials are exposed to dynamic loading as tooth cusps and hard foods cyclically indent these materials. In this work, the indentation technique is used to investigate the mechanical behavior of G-aenial under various multi-cycling nanoindentation conditions and to study the mechanical properties, such as elastic modulus and hardness. First, static indentation tests were performed to assess the effect of the holding time on the mechanical behavior of G-aenial. Then, a series of dynamic indentation tests were conducted in which the number of indentation cycles varied from 6 to 30. The experimental results indicated that both the static indentation hardness and elastic modulus decrease with increasing holding time from 0 s to 30 s. Moreover, the elastic modulus and indentation hardness of the G-aenial composite decreased with the increasing number of cycle loadings, indicating that the restorative composite becomes more susceptible to fatigue damage.
In the project management of new product development, the cost estimates of a project are directly influenced by the scope of the project. In product development, the scope of a project is the finished product. Unless the new product is a combination of existing sub-components for which the costs are known, a decomposition of an unknown product is necessary in order to form the estimates. A quick and easy method for cost estimation will be investigated involving the use of Axiomatic Design method applied to a custom design milling machine. The results confirm the utility of the method in cost estimation of a new product.
