International Scientific Journals
of Scientific Technical Union of Mechanical Engineering "Industry 4.0"

The goal of the paper is estimating the normalized friction torque of a joint in an industrial robotic manipulator. For this purpose a source data, given as a figure, is digitized using a tool WebPlotDigitizer in order to obtain numeric data. The numeric data is the used within the machine learning algorithm genetic programming (GP), which performs the symbolic regression in order to obtain the equation that regresses the dataset in question. The obtained model shows a coefficient of determination equal to 0.87, which indicates that the model in question may be used for the wide approximation of the normalized friction torque using the torque load, operating temperature and joint velocity as inputs.

This paper presents an energy analysis of main and auxiliary steam turbines from conventional coal fired power plant. Main turbine is composed of three cylinders connected to the same shaft which drives an electric generator, while auxiliary steam turbine is used for the boiler feedwater pump drive. The whole analyzed main steam turbine produces mechanical power equal to 312.34 MW, while in an ideal situation, it can produce mechanical power equal to 347.28 MW. The highest part of the mechanical power in the main turbine is produced in the low pressure cylinder. Auxiliary steam turbine in exploitation develops mechanical power equal to 6768.94 kW, while in an ideal situation it can develop 8029.03 kW. Whole main turbine energy efficiency is equal to almost 90% what is in the expected range for such high power turbines. The auxiliary steam turbine has an energy efficiency equal to 84.31%, which is almost 6% lower in comparison to the main turbine. Energy flows delivered to the last two feedwater heaters (HPH2 and HPH3) in the condensate/feedwater heating system are notably higher in comparison to energy flows delivered to any other condensate/feedwater heater.

In this paper, two-cylinder steam turbine, which operates in nuclear power plant is analyzed from the energy viewpoint. Along with the whole turbine, energy analysis is performed for each turbine cylinder (High Pressure Cylinder – HPC and Low Pressure Cylinder – LPC). A comparison of both cylinders shows that the dominant mechanical power producer is LPC, which also has much higher energy loss and much lower energy efficiency. Therefore, any potential improvement of this steam turbine should be based dominantly on th e LPC, which also has a dominant influence on energy analysis parameters of the whole observed turbine. The whole turbine produces real (polytropic) mechanical power equal to 1247.69 MW, has energy loss equal to 352.70 MW and energy efficiency equal to 77.96%. According to obtained energy efficiency value it can be concluded that the whole analyzed steam turbine is comparable to main marine propulsion steam turbines, while its energy efficiency is much lower in comparison to steam turbines from conventional steam power plant s which operates by using superheated steam.

This paper presents exergy analysis of the main steam condenser, which operates in nuclear power plant. The analysis is performed in four main condenser operating regimes (loads) for a variety of the ambient temperatures. It is found that the main steam condenser has the lowest exergy destruction (equal to 72091.56 kW) and the highest exergy efficiency (equal to 66.66%) at the lowest observed ambient temperature (5 °C) and for the highest of four observed loads. Also, it is noted that an increase in the ambient temperature from 20 °C to 25 °C (two the highest observed ambient temperatures) significantly decreases main steam condenser exergy efficiency for about 21%, regardless of the observed load.

This paper presents an analysis and comparison of three steam turbines and its cylinders: from the conventional steam power plant, from nuclear power plant and from the marine propulsion plant. The best parameters for the comparison of whole turbines and its cylinders are: energy loss per unit of produced mechanical power, exergy destruction per unit of produced mechanical power, energy efficiency and exergy efficiency. Steam turbine from marine propulsion plant shows the worst performance, regardless if observing each cylinder or the whole turbine – it has the highest losses per unit of produced mechanical power and the lowest efficiencies (both energy and exergy). Such results can be explained by a fact that marine steam turbine must be much more dynamic in operation in comparison to other two turbines. Also, marine steam turbine analyzed in this paper did not possess steam reheating between the cylinders as the other two observed steam turbines, what has a dominant impact on the obtained results.

This paper presents a comparison of three methods for any pump energy analysis. Each method is used for the analysis of three different water pumps from the conventional steam thermal power plant – two feed water pumps (FWP1 and FWP2) and condensate pump (CP). For each pump three essential types of mechanical power which defines all energy analysis methods are: delivered power from power producer, real (polytropic) power and ideal (isentropic) power. Method 1 which compares delivered and real (polytropic) power show the best performances, while Method 3 which compare delivered and ideal (isentropic) power should be avoided because it results with too high energy power loss and too low energy efficiency of any pump. Method 2 which compares real (polytropic) and ideal (isentropic) pump power can be used as a good compromise for the pump energy analysis in the most of the cases – its results are similar to results of Method 1.

Previous work has determined the ability of using the Multilayer Perceptron (MLP) type of Artificial Neural Network (ANN) to estimate the power output of a Combined Cycle Power Plant (CCPP) in which optimization did not focus on the solver parameter optimization. In previous work, the solvers used the default parameters. Possibility exists that optimizing solver parameters will net better results. Two solver algorithm’s parameters are optimized: Stochastic Gradient Descent (SGD) and Adam, with 140 and 720 parameter combinations respectively. Solutions are estimated through the use of Root Mean Square Error (RMSE). Lowest RMSE achieved is 4.275 [MW] for SGD and 4.259 [MW] for Adam, achieved with parameters: = 0.05, = 0.02, and nesterov=True for SGD and with parameters = 0.001, 1 = 0.95, 2 = 0.99, and amsgrad=False for Adam. Only a slight improvement is shown in comparison to previous results (RMSE=4.305 [MW]) which points towards the fact that solver parameter optimization with the goal of improving results does not justify the extra time taken for training.