Sorin Ciortan’s research while affiliated with "Dunarea de Jos" University of Galati and other places

The chapter presents an evaluation on long-term range of the wind energy availability in existing Romanian wind farms. The necessity of increasing the renewable energy sources in the global energy portfolio of Romania leads to the development of several wind farms around the country. Although these farms’ establishment is based on some previous assessments of wind energy potential, there are no references on the future long-term evolution of the wind speed in these locations. Taking into account that the main parameter in wind turbine efficiency is the wind speed and the permanent climate changes, wind speed changes occur too. Knowing that the estimated lifetime of a wind turbine is around 20 years, an analysis of the predicted wind speed for the 30-year period 2020–2050, based on RCP4.5/8.5 scenarios, allowed to evaluate the opportunity to maintain and develop the wind farms in the same areas or start new investigation in order to find new appropriate places for the future wind energy extraction.

The paper presents an evaluation on long-term range of the wind energy availability in existing Romanian wind farms. The necessity of increasing the renewable energy sources in the global energy portfolio of the Romania leads to the development of several wind farms around the country. Although these farms' establishment is based on some previous assessments of wind energy potential , there are no references on the future long-term evolution of the wind speed in these locations. Taking into account that the main parameter in wind turbine efficiency is the wind speed and the permanent climate changes, in the wind speed changes occurs too. Knowing that the estimated lifetime of a wind turbine is around 20 years, an analysis of the predicted wind speed for the 30-year period 2020-2050, based on RCP4.5/8.5 scenarios, allowed to evaluate the opportunity to maintain and develop the wind farms in the same areas or start new investigation in order to find new appropriate places for the future wind energy extraction.

In the frame of latest requirements to improve the renewable energy harvesting, the wind turbines get more and more used. Due to the propeller sizes, the blades� materials must comply with high resistance and weight requirements. As consequence, composite materials with dedicated structures are used. These materials are designed for long life and, as consequence, are potential pollution factors for the environment, after turbine working life is ended. Beside the methods to destroy the blades and give other use to the material, the optimization of the maintenance and repair procedures stand as valuable way to reduce both the turbines cost and the environment contamination. Taking into account the working conditions of blades such as the presence of the abrasive particles (sand or dust) and the high rotational speeds that can be reached, following the wind speed, the assessment of wear resistance of blades in abrasive conditions stands as valuable information for both designing and repairing procedures. In the present work, an industrial blade composite material is analyzed and subjected to abrasive wear with sand particles flowing in air stream, simulating the wind effect. A dedicated test rig is used, allowing to change the wear process parameters. After the abrasive tests, the damage evolution was observed by optical microscopy to identify the most affected areas of the blade, which provides important information for establishing of repairing procedures.

In order to obtain the most efficient solution in harvesting offshore wind energy, the converting devices must be located in places where the wind provides enough power, according to the used device specifications and, as possible, with constant value during turbine lifetime. On the other hand, the device must be chosen for a specific place-imposed by other reasons, like water depth, shore distance etc. The paper presents an analysis and prediction methodology for the wind speed based on the artificial neural networks modeling. The proposed approach uses as input data the previous recorded and forecasted wind speed values and the GPS coordinates for several places in Black Sea allowing the identification of the location with less wind speed modification during analyzed period of time. Also, the model provides useful information about inputs importance on output evolution. The obtained results are used for the identification of optimal placement of wind energy converting devices, leading to an improved efficiency.

In the marine environment several renewable energy sources can be found as wind, waves, tide, thermal and osmotic. Taking into account that the oceans' estimated energy resource is around 83340TWh/year, representing up to 400% of world electricity demand, it is obvious that its exploitation might represent an important issue in the actual frame of looking for sustainable, green energy. Among the mentioned resources the most targeted are wind, waves and tidal. In order to obtain the most efficient energy conversion two criteria must be followed: the appropriate pairing of converting device type and placement location and an analysis of the specific energy resource evolution in time. If the first criterion is linked mostly by the technological evolution, which is day-by-day more rapid, the second one is linked to the climate changes, requiring forecasting procedures in order to optimize the energy extraction. Usually, these predictions relay on numerical methods, using meteorological parameters linkages and correlation coefficients as inputs. But, during climate changes also the weights of these parameters into the linkage are changing. So, in order to obtain efficient forecast two possibilities occur: the continuously updating of correlation coefficients used or, using a different approach, which can offer prediction and optimization without mathematical formulas-Artificial Neural Networks (ANN). The present work highlights the possibility and the opportunity of using ANN for the analysis, forecasting and optimization of the renewable energy sources evolution in marine environment. ANNs are parallel calculus systems working in a different way comparing to the systems based on von Neumann machine, i.e. trivial computers. As a consequence, several benefits can be obtained: there is no need for a mathematical algorithm, they can process directly experimental acquired values, less sensitivity to errors and missing data etc. ANNs are working in two stages: first, the network is trained with known input-output pairs of data, second the network can provide predicted outputs for input data which have no known output pairs. ANN based models are similar to biological brains, composed by several so called "artificial neurons", distributed in layers and interconnected through links. These links are modified during the training stage, defining this way the correspondence between input and output data. In the second stage, the ANN model can be used for prediction, optimization or importance analysis. In the case of the waves, with a power potential roughly estimated at around 1 TW, ANN based models, trained with month, day and hour and wind speed as inputs and wave height as output, offers the benefit to predict waves' height starting from future date, the establishing the most influencing input or finding an optimal date when the waves' height is extreme (minimum or maximum). The training data are obtained from historical records, available in international databases.: www.dream.ugal.ro 2/2 Other important renewable energy resource is the wind, the oldest one used by humanity. Today, almost 600 GW of power are obtained all over the world from the wind farms. Lately, due to advances in wind turbine technologies the amount of offshore wind energy conversion is continuously increasing. As consequence, the evolution analysis and forecasting of the available wind power is a top subject for researchers. The ANN based modelling, using as inputs the wind turbine height, the water surface roughness and the wind speed can provide as output the wind power density and, at consequence, the available power provided. The inputs data are provided by existing international data bases. The model can be used for prediction, importance analysis or optimization of the wind farms placement.

The energy of the waves is estimated around 1 TW, worldwide. Harvesting of this enormous potential is influenced by several factors, mainly by: high values of the initial investment, changes of the weather conditions, time distribution of the wave height and the conversion devices. Several wave energy conversion devices types are available, each one working based on different principles and offering different facilities. Choosing a particular type of converter relay on an analysis involving the initial investment value, meaning the device price, the installation and maintenance price and other unexpected expenses; all these can reach important amounts. Also, the waves present very often changes in frequency and direction, following the weather changes. As a consequence, there is no universal optimal converter, for each location being necessary studies for establishing the optimal one. This analysis is more important for closed seas, where the waves' energy is less than in the oceans. Following the dynamic systems theory, a system is a collection of components linked together with relationships. The system behaviour is influenced both by components properties and by the links between. As a consequence, modelling and analyse of a system must be performed taking into account not just the components but the interactions between these too. Based on this theory, any real system structure can be modelled with few elements: level (acting as reservoirs), rates (acting as valves), flows (acting as links between the components) and constants (containing constants values). Considering that the waves, with their properties (significant height and period), the energy converter device, with specific working conditions and efficiency and environmental conditions (water density and temperature, gravitational acceleration and water depth) can be considered together as a dynamic system, a corresponding model can be built and simulated. The connections between the models' components follow the correlations between the real components: the wave power is the result of interactions between the wave height, wave period, water density and gravitational acceleration. But the wave height and wave period are influenced by the time, day, hour etc. following the chosen timestep for the simulation. All these factors are influencing also the converting device efficiency. From device side of view, the technical specifications and performances must be taken into account. In order to include into the model the forecasting of wave characteristics over the simulation time, an artificial neural network can be used. If several devices are included into the model, the analysis of the simulation results can provide valuable information, allowing in this way choosing of the appropriate device for a precise location.

The paper presents an analysis of the wind energy availability on long-term basis in the Republic of Moldova. This is motivated by the necessity of increasing the renewable energy sources in the global energy portfolio of the Republic of Moldova. Although there are some studies on this subject, concluding that there is a satisfactory wind energy potential, there are no references on the future evolution of the wind energy. Starting from the analysis of the historical wind speed data corresponding to the 31-year period 1990-2020 and taking into account the most significant geographic and civil requirements, several locations were chosen as possible targets for the future wind farms placement. An analysis of the predicted wind data for the 31-year period 2020-2050 based on RCP4.5 scenario, allowed to identify the optimal areas for the wind energy extraction in the republic of Moldova.

The paper presents an analysis of renewable energy available in the Republic of Moldova. Because of geographic localization, there is no possibility to use wave energy so, the wind, solar, rivers and biomass sources are taken into account. After a short presentation of benefits and drawbacks of each energy source, an estimation for the Republic of Moldova is performed. The analysis results pointed out that the wind energy is the most used type and there is a lot of information about the wind speed values over the whole country area.