ehsan bahrami

The challenge of excessive energy consumption, related environmental issues, and optimal utilization of restructured power systems require proper management on the demand side. In this context, demand response programs, try to encourage, persuade, or force consumers to adjust their consumption patterns within that established by the system operator. In this article, to improve accuracy, nonlinear models for demand response programs (power, exponential, and logarithmic models) for incentive-based and time-based programs, have been developed based on price elasticity and customer profit function. Then, the behavior of the proposed models against changes in elasticity, encouragement/penalty rate, and the effect of consumer participation have been investigated in several scenarios. Since consumer participation in these programs depends on various economic, cultural, and social factors, it cannot be accurately predicted. In addition, consumer participation has a significant effect on the program results. So, the consumer participation percentage has been considered an uncertain parameter based on a normal probability distribution function. The results show that the non-linear models are more accurate and more conservative than the linear model. Moreover, considering consumer participation as an uncertain normal parameter results in more reliable responses with more peak reduction and a greater reduction in total energy consumption.

Nonrespose is a source of error in the survey results and national statistical organizations are always looking for ways to control and reduce it. Predicting nonrespons sampling units in the survey before conducting the survey is one of the solutions that can help a lot in reducing and treating the survey nonresponse. Recent advances in technology and the facilitation of complex calculations have made it possible to apply statistical learning methods, such as regression and classification trees or support vector machines, to many issues, including predicting the nonresponse of sampling units in statistics. In this article, while reviewing the above methods, the nonresponse sampling units are predicted using them in an establishment survey and it is shown that a combination of the above methods is more accurate in predicting the correct nonresponse than any of these methods.

In this paper, we propose a new parametric distribution which called as the Beta Modified Weibull Power Series (BMWPS) distribution. This distribution is obtained by compounding Beta Modified Weibull (BMW) and power series distributions. BMWPS distribution contains, as special sub-models, such as Beta Modified Weibull Poisson (BMWP) distribution, Beta Modified Weibull Geometric (BMWG) distribution, Beta Modified Weibull Logarithmic (BMWL) distribution, among others. We obtain closed-form expressions for the cumulative distribution, density, survival function, failure rate function, the r-th raw moment and the moments of order statistics. A full likelihood-based approach that allows yielding maximum likelihood estimates of the BMWPS  arameters is used. Finally, application to the Aarset data are given.