Some aspects of the implementation of power plant management technologies taking into account the concept of autonomous vessels

Обеспечение удаленного управления судовой энергетической установкой автономного судна, контроль параметров процессов и оценка их технического состояния с использованием искусственных нейро-нечетких сетей для морского автономного надводного судна (МАНС). Создание цифровых двойников и прогнозирование технического состояния, моделирование сценариев изменения технического состояния судовых технических средств. Современные тенденции и перспективы научных изысканий в области управления судовыми энергетическими установками и системами, их обслуживающими, применительно к концепции автономного судна и безэкипажного (беспилотного) судоходства, а также выполнения ужесточающихся требований международного морского законодательства по предотвращению загрязнения атмосферы с судов. Использование альтернативных топлив и переход на электроснабжение с берега при стоянке таких судов в портах, на оборудованных рейдовых стоянках, а также использование «чистой электроэнергии» для электроснабжения объектов водного транспорта и шельфовой инфраструктуры. Providing remote control of the ship's power plant of an autonomous ship, monitoring process parameters and assessing their technical condition using artificial neuro-fuzzy networks for a marine autonomous surface vessel (MANS). Creation of digital twins and forecasting of the technical condition, modeling of scenarios of changes in the technical condition of ship technical equipment. Modern trends and prospects of ship power plants and their elements in relation to the concept of an autonomous vessel and crewless (pilotless) navigation, incorporating meeting the potential newer requirements / restrictions of international maritime legislation regarding airborne pollutions from ships. The use of alternative fuels and the alternative to marine power technology from the shore when such vessels are moored in ports, at equipped roadsteads, as well as the use of “clean electricity” to supply power to marine transport facilities and offshore infrastructure.

BOUNDARY CONDITION MODELING EFFECT ON THE SPENT FUEL CHARACTERIZATION AND FINAL DECAY HEAT PREDICTION FROM A PWR ASSEMBLY

In this paper, two main exercises have been carried out to describe the effect that varying an albedo boundary condition has in the computation of observables such as decay heat, neutron emission rate and nuclide inventory from a PWR fuel assembly (or a configuration of assemblies) during a depletion scenario. The SERPENT2 code was then employed to emphasize the importance of modeling a proper boundary condition for such purposes. Moreover, the effect of taking into account more than a single fuel-pin region for depletion studies while varying the type of boundary condition, was also accounted for. The first exercise has the main objective of comparing in a single fuel assembly the albedo variations ranging from 1.1 up to full vacuum conditions. By comparing to the reference assembly (considered to be the case of full reflective conditions), relative differences up to +17% were observed in decay heat and up to almost -30% in neutron emissions. Also, a clear dependence on the albedo was detected if more than one depletable zone was considered while computing the integral value of observables of interest. Regarding the second exercise, where a 3 × 3 configuration of fuel assemblies is being now considered with a reflector section in the middle, a negligible effect on the observables was observed for the single fuel pin zone case; instead, an effect in the 244Cm computation when analyzing two fuel pin-zones produced a change in the neutron emission rate during cooling time up to 2.5% (while comparing it to the reference single assembly case).

Development of Insert Condition Classification System for CNC Lathes Using Power Spectral Density Distribution of Accelerometer Vibration Signals

Insert conditions significantly influence the product quality and manufacturing efficiency of lathe machining. This study used the power spectral density distribution of the vibration signals of a lathe machining accelerometer to design an insert condition classification system applicable to different machining conditions. For four common lathe machining insert conditions (i.e., built-up edge, flank wear, normal, and fracture), herein, the insert condition classification system was established with two stages—insert condition modeling and machining model fusion. In the insert condition modeling stage, the magnitude features of the segmented frequencies were captured according to the power spectral density distributions of the accelerometer vibration signals. Principal component analysis and backpropagation neural networks were used to develop insert condition models for different machining conditions. In the machining model fusion stage, a backpropagation neural network was employed to establish the weight function between the machining conditions and insert condition models. Subsequently, the insert conditions were classified based on the calculated weight values of all the insert condition models. Cutting tests were performed on a computer numerical control (CNC) lathe and utilized to validate the feasibility of the designed insert condition classification system. The results of the cutting tests showed that the designed system could perform insert condition classification under different machining conditions, with a classification rate exceeding 80%. Using a triaxial accelerometer, the designed insert condition classification system could perform identification and classification online for four common insert conditions under different machining conditions, ensuring that CNC lathes could further improve manufacturing quality and efficiency in practice.

EQUIPMENT CONDITION MODELING BASED ON RANDOM FOREST AND ARIMA MACHINE LEARNING ALGORITHM STACKING

The article discusses algorithms for constructing predicative models of the industrial equipment condition using data analysis and machine learning. The model is based on Random Forest (RF) and ARIMA (AR) algorithms. The authors consider approaches to learning algorithms and optimizing parameters. A block diagram of a time series predictive model applying stacking is presented, as well as an assessment of the simulation results.