The Ship’s Digital Twin (SDT) is a digital record of a ship’s behaviour or a software clone, which can be used to simulate scenarios that are expensive or hardly feasible to perform on a real object and especially in real time. The purpose of the SDT is to achieve cost reduction, obtain timely warnings of irregularities, and optimise individual ship system performances or the operation of the whole ship and to assist ship management. The aim of this paper is to describe the concept of the SDT and clarify some perplexities that may occur from initial introduction to concept. To that end, the paper identifies the steps in the SDT formulation process and methods used in each step of the process. Furthermore, a four-step iterative procedure for the SDT development is proposed. The applications of the concept are numerous, and some of them are presented in a review analysis in this paper. The presented analysis leads to a conclusion that should give some direction to future research in this area.

Artificial neural networks (ANN) are a data driven tool that has been used for modeling, prediction, optimization, classification, diagnostics, decision-making, etc., in various systems where measurements are available to produce significant amount of data. Ship processes are constantly monitored in order to control the operation of the ship and to ensure efficient and safe environment, generating large amount of data. Those data are increasingly being exploited by ANNs and the number of applications is growing. The aim of this paper is to analyze the applications of ANNs in maritime industry, and especially on ships. Based on the review analysis of the sixty-nine papers found published on this topic over the last 10 years in relevant databases, applications have been classified into eight categories in this paper. ANN types, training algorithms, activation functions, as well as measures used to evaluate the performance of the ANN models, have been analyzed for each application category. ANNs rely on data, therefore data acquisition, data processing, organization of the data for training ANN models, their validation and testing have also been addressed in this paper. The conclusions from the review analysis presented should be useful for future work in the area of ANN applications on ships and in maritime industry.

The paper deals with sea wave modelling based on available data acquired from a satellite-calibrated numerical model. The idea is to use an artificial neural network, as a flexible tool capable of modelling nonlinear processes, for significant wave height (SWH) modelling at a single point in the Adriatic Sea. The focus of the paper was not to develop a new type of ANN, but rather to use it as a modelling tool and identify the most significant input variables for SWH modelling in the Adriatic Sea, among the available data. Linear and nonlinear regression models were also developed for purposes of comparison of neural network performances with those of traditional data modelling methods. A total of 22 years of data were used - 20 years of data with a 6 h sampling step time, i.e. 30684 data samples were used to calibrate the models, while 2 years of data, i.e. 2920 data samples were used to test the models’ performances. Simulation results proved the ability of an artificial neural network to model SWH with high ccuracy based on available data. Furthermore, the artificial neural network model proved to be more accurate than traditional statistical models, especially when multiple input variables were used.

Artificial neural networks have been successfully used to model and predict water flows for a few decades. Different network types have proven to work better in different cases and additional tools and algorithms have been implemented to improve thos...

Accurate controller tuning is important for ensuring optimal performance in flow control processes, particularly onboard ships, where precise control of fluid systems such as oil, gas, and water flows determines the operation of the whole ship. Models can be used to avoid experimentation on a real system, which is essential if systems are unavailable for controller tuning experiments. This paper presents the results of controller tuning using the ANN model. To evaluate the effectiveness of the proposed method, it is compared to the transfer function model-based tuning and to the controller tuning performed on a real system. The controller uses an ideal PID structure, and controller parameters are calculated by four well-known tuning methods for each case. The experiments are conducted on a flow control unit in a laboratory, while modelling and simulations are performed in Matlab. Graphical evaluation and numerical RMSE metrics are used to determine the quality of the tuning method. By evaluating controller tuning performance based on the ANN model, this research aims to validate the effectiveness of ANN models for practical control applications.

A simple neural model for electromagnetic field prediction in indoor environment was created based on field strength measurements at 2.4 GHz, conducted at University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture (FESB Split). Vertical rod antenna (omnidirectional in the horizontal plane) was placed in a faculty hallway and used as the electromagnetic field source. Electromagnetic field distribution was defined by commonly used rectangular grid of uniformly distributed measurement points. However, instead of commonly used Cartesian coordinates for measurement points location description, we used polar coordinates of distance and azimuth angle measured from the field source. These coordinates are found to be more suitable for the organization of input data, as the physical distribution of the field strength around the antenna depends on the same variables. This resulted with predictive ability improvement of the neural model, as confirmed by simulation results. Three-Layer Perceptron, trained with Levenberg-Marquardt (LM) algorithm, produced the best results.