The inherent benefits of an accident prevention program are generally known only after an accident has occurred. The purpose of implementation of the program is to minimize the number of accidents and cost of damages. Allocation of resources to implement accident prevention program is vital because it is difficult to estimate the extent of damage caused by an accident. Accurate fatal accident predictions can provide a meaningful data that can be used to implement accident prevention program in order to minimize the cost of accidents. This paper forecast the fatal accidents of factories in India by using Auto-Regressive Integrating Moving Average Method (ARIMA) model. Accident data for the available period 1980 to 2013 was collected from the Labour bureau, Government of India to analyze the long term forecasts. Different diagnostic tests are applied in order to check the adequacy of the fitted models. The results show that ARIMA (0, 0, 1) is suitable model for prediction of fatal injuries. The number of fatal accidents is forecasted for the period 2014 to 2019. These results suggest that the policy makers and the Indian labour ministry must focus attention toward increasing fatal accidents and try to find out the reasons. It is also an opportunity for the policy makers to develop policies which may help in minimizing the number fatal accidents.
This article analyzes the traditional time series processing methods that are used to perform the task of short time series analysis in demand forecasting. The main aim of this paper is to scrutinize the ability of these methods to be used when analyzing short time series. The analyzed methods include exponential smoothing, exponential smoothing with the development trend and moving average method. The paper gives the description of the structure and main operating principles. The experimental studies are conducted using real demand data. The obtained results are analyzed; and the recommendations are given about the use of these methods for short time series analysis.
Selection of Hidden Layer Neurons and Best Training Method for FFNN in Application of Long Term Load Forecasting
For power industries electricity load forecast plays an important role for real-time control, security, optimal unit commitment, economic scheduling, maintenance, energy management, and plant structure planning etc. A new technique for long term load forecasting (LTLF) using optimized feed forward artificial neural network (FFNN) architecture is presented in this paper, which selects optimal number of neurons in the hidden layer as well as the best training method for the case study. The prediction performance of proposed technique is evaluated using mean absolute percentage error (MAPE) of Thailand private electricity consumption and forecasted data. The results obtained are compared with the results of classical auto-regressive (AR) and moving average (MA) methods. It is, in general, observed that the proposed method is prediction wise more accurate.
Andrzej Gałecki, Lech Grzesiak, Barłomiej Ufnalski, Arkadiusz Kaszewski and Marek Michalczuk
The paper presents a linear-quadratic current controller with damped oscillatory terms designed for three-phase grid-tie voltage source converters used in SMES systems and operated under distorted grid voltage conditions. Special emphasis is placed on a synthesis of an anti-windup mechanism to prevent wind-up derived from the oscillatory terms by the use of a new active damping loop based on a simple moving average method. As a consequence, the current feedback gain may be increased without unwanted overshoot and overruns, and performance of the system can be improved.
The article analyses the applicability of selected smoothing methods to smooth indicator diagram curves and to filter disturbances. An intermediate goal of the study was an attempt to extract disturbances recorded during pressure curve smoothing, which are believed to be a source of important diagnostic information.
Within the framework of the reported analysis, a comparison was made between the moving average method, the Savitzky-Golay filter, and the frequency filtration method. The research was performed on a marine medium-speed engine Sulzer 3Al 25/30, which has a relatively long indicator passage.
The paper explores the possibility of making investment decisions in emerging markets by using the trend analysis method on a particular example of the capital market in Serbia. The authors, starting from the common features of technical analysis, have analysed the common share index value in the capital market in Serbia, in the Belgrade Stock Exchange - Belexline from 1 March 2006 to 31 March 2009, by the usage of two moving averages method - Moving Average Convergence Divergence (MACD): an intermediate term of 50 days and a long-term one of 100 days. The above mentioned moving averages identify the establishment of a trend, the cessation of the existing one, a change and an establishment of the new one.
The capital market in Serbia had two distinctive long-term trends within the above mentioned observed period of time. The method of two moving averages in combination with the MACD indicator analysis gave quite reliable signals of weakening and change of the long-term trend direction. Analysis of the long-term trend has not been considered for the period from 2009 to date because the market during this period was illiquid with little trading volume, while some stocks that are entered in the Belexline are not more subject of trade.