SYSTEM OF LOAD DISTRIBUTION OF SHIP DIESEL GENERATORS BASED ON NEURAL NETWORKS
Journal Title: Современные информационные технологии и ИТ-образование - Year 2018, Vol 14, Issue 3
Abstract
The dependence of the specific fuel consumption of the ship diesel-generator set on the generated electric power and the power factor is given. An algorithm for load distribution between parallel marine diesel generator sets is given, taking into account the dependencies of the specific fuel consumption on the generated power and the power factor of each marine diesel generator set. The possibility of using neural networks to solve the problem of minimizing the specific fuel consumption for parallel operation of ship diesel-generator sets in the steady state mode is considered. A neural network model for reducing fuel consumption is proposed by distributing the total electric load of a ship power plant between diesel generators operating in parallel. Such a model, based on the database obtained in the training process, which is implemented by microprocessor controllers with the algorithms of searching for a combination of capacities inherent in them for each of the parallel running generator, at which the minimum flow is reached, independently determines how it is necessary to load the units operating in parallel, to achieve a minimization of fuel consumption; and periodically checks the correspondence between the estimated (calculated) and the actual fuel costs, which ensures the control of the technical condition of the mechanisms.
Authors and Affiliations
Sergey Kuznetsov, Nikolay Alekseev, Andrey Vinogradov
Keywords
Related Articles
- EP ID EP520775
- DOI 10.25559/SITITO.14.201803.603-608
- Views 96
- Downloads 0
METHODS OF CREATING DIGITAL TWINS BASED ON NEURAL NETWORK MODELING
It is assumed that by 2021, about half of the companies will use digital counterparts of different levels. The simplest digital twin models may not use machine learning, but the models using machine learning algorithms w...
CLUSTERING MODEL OF LOW-STRUCTURED TEXT DATA
The article proposes a clustering model for collections of news text messages, as well as the corresponding bubble trap clustering algorithm. The essence of the proposed approach is to divide the entire vector space of t...
BENCHMARKING BIG SPATIAL DATA PROCESSING FRAMEWORKS
Today, the processing of large amounts of spatial data in distributed systems plays a crucial role in many areas of our life. Large data are often unstructured, and special algorithms are required for its processing. One...
ABOUT PERSONIFICATION OF TEACHING SCHOOLCHILDREN PROGRAMMING
Information education of the personality is one of the most mobile types of education depending on the dominating paradigm of development of society, degree of development and the prospects of further development of econ...
SIGN LANGUAGE NOTATION SYSTEMS AND THEIR COMPARATIVE ANALYSIS
In this paper we describe gestures and sign language as means of non-verbal communication of people with hearing loss. We give a brief overview of existing sign language notations on currently, designed to depict gesture...