Use Of Exogenous Variables Research Articles

Detecting bearing failures in wind energy parks: A main bearing early damage detection method using SCADA data and a convolutional autoencoder

AbstractWind energy maintenance and operation costs can total millions of dollars each year in an average industrial‐size wind park. Therefore, moving from preventive and corrective maintenance to predictive maintenance is imperative in the wind energy sector. This paper contributes to this challenge by providing a main bearing early damage detection technique that exclusively uses standard supervisory control and data acquisition (SCADA) data (10‐min average) and a convolutional autoencoder with the following contributions. (i) Entirely semisupervised (not requiring the labeling of data through work order logs and avoiding the problem of data imbalance between classes) based only on healthy data, thus expanding its range of application (even when the failure of interest has never occurred in the park before). (ii) Validated using real‐world SCADA data and shown to be resistant to seasonality, and operational and environmental conditions. (iii) Reliable predictions with minimum false alarms thanks to specially designed fault prognosis indicators based on the image mean square error metric. (vi) The early warning is achieved months in advance, thus providing adequate time for plant operators to plan properly. (v) The main use of exogenous variables in the model (variables that are not affected by other variables, e.g., wind speed, wind turbulence, and ambient temperature) guarantees the detection of damage directly related only to the low‐speed shaft temperature (the only nonexogenous variable used by the stated model). (vi) Finally, the proposed strategy is validated in a wind park made up of 12 wind turbines.

Anchor Regression: Heterogeneous Data Meet Causality

AbstractWe consider the problem of predicting a response variable from a set of covariates on a data set that differs in distribution from the training data. Causal parameters are optimal in terms of predictive accuracy if in the new distribution either many variables are affected by interventions or only some variables are affected, but the perturbations are strong. If the training and test distributions differ by a shift, causal parameters might be too conservative to perform well on the above task. This motivates anchor regression, a method that makes use of exogenous variables to solve a relaxation of the ‘causal’ minimax problem by considering a modification of the least-squares loss. The procedure naturally provides an interpolation between the solutions of ordinary least squares (OLS) and two-stage least squares. We prove that the estimator satisfies predictive guarantees in terms of distributional robustness against shifts in a linear class; these guarantees are valid even if the instrumental variable assumptions are violated. If anchor regression and least squares provide the same answer (‘anchor stability’), we establish that OLS parameters are invariant under certain distributional changes. Anchor regression is shown empirically to improve replicability and protect against distributional shifts.

Mitigating Effect of Additional Variables on Common Method Variance in Structural Equations Models

This study examines the effect that adding addition exogenous variables has on mitigating the effect of common method variance. In an effort that replicates and extends the work of Siemsen, Roth, and Oliveira (2010), we consider the effects of adding variables in structural equations modelling as well as regression. We also consider the implications of imperfect measurement conditions, including measurement error and the use of exogenous variables that are not orthogonal to all other variables in the study. Using two types of data simulation, the current study indicates that the use of additional variables in structural equations models can reduce the inflationary effects of CMV. We also show that the beneficial effects of adding additional variables are not mitigated by the use of imperfect marker variables. The study has implications for researchers conducting studies where CMV may be a threat, as well as for interpreting prior research that did not take steps to test for or address potential CMV concerns.

Optimizing Multireservoir System Operating Policies Using Exogenous Hydrologic Variables

AbstractStochastic dual dynamic programming (SDDP) is one of the few available algorithms to optimize the operating policies of large‐scale hydropower systems. This paper presents a variant, called SDDPX, in which exogenous hydrologic variables, such as snow water equivalent and/or sea surface temperature, are included in the state space vector together with the traditional (endogenous) variables, i.e., past inflows. A reoptimization procedure is also proposed in which SDDPX‐derived benefit‐to‐go functions are employed within a simulation carried out over the historical record of both the endogenous and exogenous hydrologic variables. In SDDPX, release policies are now a function of storages, past inflows, and relevant exogenous variables that potentially capture more complex hydrological processes than those found in traditional SDDP formulations. To illustrate the potential gain associated with the use of exogenous variables when operating a multireservoir system, the 3,137 MW hydropower system of Rio Tinto (RT) located in the Saguenay‐Lac‐St‐Jean River Basin in Quebec (Canada) is used as a case study. The performance of the system is assessed for various combinations of hydrologic state variables, ranging from the simple lag‐one autoregressive model to more complex formulations involving past inflows, snow water equivalent, and winter precipitation.

Non linear models and the load of an electricity distributor

Purpose – The purpose of this study is to ascertain whether nonlinearities could be present in electricity loads observed in subtropical environments, where none or little heating is required, and whether threshold autoregressive (TAR)-type regime switching models could be advantageous in the modeling of those loads. Design/methodology/approach – The actual observed load of a Brazilian regional electricity distributor from January 2013 to August 2012 was modeled using a popularly employed ARMA model for reference, and smooth and non-smooth TAR transition (non-linear) models were used as non-linear regime switching models. Findings – Evidence of nonlinearities were found in the load series, and evidence was also found on the intrinsic resistance of this type of models to structural breaks in the data. Additionally, to reacting well to asymmetries in the data, these models avoid the use of exogenous variables. Altogether, this could prove to be a definite advantage of the use of such model alternatives. Research limitations/implications – However, even if the present work may have been limited by the observation frequency of the available data, it appears TAR models appear to be a viable alternative to forecasting short-term electricity loads. Nonetheless, additional research is required to achieve a higher accuracy of forecast data. Practical implications – If such models can be successfully used, it will be a great advantage for electricity generators, as the computational effort involved in the use of such models is not significantly larger than regular linear ones. Originality/value – To our knowledge, this type of research has not yet been made with subtropical/tropical electricity load data.

Relative Income and Female Autonomy

This study attempts to explore how relative income contributes to the autonomy of women at both personal and household levels. The potential problem of endogeneity, or reverse causality, in examining the link between female autonomy and relative income is addressed using the Instrumental Variable (IV) estimation approach. The years of education of a woman and her husband are used as instruments for the woman's relative income. Along with the use of exogenous variables as instruments, the heteroskedasticity present in the data is also exploited to form instruments for the relative income of women. The results of this study suggest that relative income unambiguously improves autonomy of women in household decision making but its beneficial effect on autonomy in personal decision making was not found to hold true in all the model specifications (with alternative identification schemes) as employed in this study.

Electricity Distributor Load Forecasting with ARMA Models

The present work attempts to evaluate the advantages inherent to the use of exogenous variables highly correlated to the electric load, for the forecast of future demand. Here we utilize time series models of the auto-regressive moving average types incorporating seasonal treatment and exogenous variables. The details relevant to good modeling are presented and various different models are adjusted, including different combinations of load against the exogenous regressors IBC-Br and temperature. Considerations regarding non-stationary data such as different data generating processes (DGP) using trend or difference integration alternatives are investigated. In the end we demonstrate that predictions made using ARMA type models do not benefit from further sophistication in the form of exogenous variables due to the error inherent to the method itself, when applied to long term forecasts.

Optimization of prediction methods for patents and trademarks in Spain through the use of exogenous variables

An accurate forecast of patent and trademark application filings is strategic for resource planning at the Spanish Patents and Trademarks Office and other patent offices, national and supranational. The need for reliable forecasts of patents and trademarks application filings has been accentuated by the current situation of budgeting rationalization imposed by the economic crisis. In this study we have evaluated the suitability and effectiveness of different methodologies for advanced data analysis to predict the number of national patent and trademark applications in the short and medium terms (2011–2014), including the use of exogenous variables or predictors which help to understand the changes in these variables. The inclusion of exogenous variables which explain the behavior of patent and trademark application filings, in particular the investment in R&D and GDP, and the use of advanced predictive analysis techniques, amongst which the most notable are Polynomial Distributed Lags and Intelligent Transfer Function models, have all achieved an improvement upon the prediction and modeling power possessed by the models formerly used to predict trademark and patent series based only on the analysis of time series.

A study on the medium-term forecasting using exogenous variable selection of the extra-virgin olive oil with soft computing methods

Time series forecasting is an important task for the business sector. Agents involved in the olive oil sector consider that, for the olive oil price, medium-term predictions are more important than short-term predictions. In collaboration with these agents the forecasting of the price of extra-virgin olive oil six months ahead has been established as the aim of this work. According to expert opinion, the use of exogenous variables and technical indicators can help in this task and must be included in the forecasting process. The amount of variables that can be considered makes necessary the use of feature selection algorithms in order to reduce the number of variables and to increase the interpretability and usefulness of the obtained forecasting system. Thus, in this paper CO2RBFN, a cooperative-competitive algorithm for Radial Basis Function Network design, and other soft computing methods have been applied to the data sets with the whole set of input variables and to the data sets with the selected set of input variables. The experimentation carried out shows that CO2RBFN obtains the best results in medium term forecasting for olive oil prices with the whole and with the selected set of input variables. Moreover, the feature selection methods applied to the data sets highlighted some influential variables which could be considered not only for the prediction but also for the description of the complex process involved in the medium-term forecasting of the olive oil price.

Analysis of prediction performance of training-based models using real network traffic

Traffic prediction constitutes a new hot research topic of network metrology. Thus, tuning the prediction model parameters is very crucial to achieve accurate prediction. This work focuses on the design, the empirical evaluation and the analysis of the behavior of training-based models for predicting the throughput of a single link i.e. the incoming input data in Megabit per time interval called granularity. In this work, a neurofuzzy model (alpha_SNF) and the autoregressive moving average (ARMA) model are used for predicting. Via experimentation on real network traffic of different links, we study the effect of some parameters on the prediction performance in terms of error. These parameters are the amount of data needed to identify the model (i.e. training set), the number of last observations of the throughput (i.e. lag) needed as inputs for the model, the data granularity, variance and packet size distribution. We also investigate the use of exogenous variables as inputs for the model. Exogenous variables are variables which are different from the lags such as the number of packets or sampled data. Experimental results show that training-based models, identified with small training set and using only one lag, can provide accurate prediction. We show that counts of packets and especially large packets can be used to efficiently predict the throughput.

비정상성 Markov Chain Model을 이용한 통계학적 Downscaling 기법 개발

기존의 정상성 Markov Chain 모형은 자료 자체의 Markov 특성만을 고려하여 모의하는 기법으로서 수자원 설계에서 여러 가지 목적으로 이용되어 지고 있다. 그러나 일강수량의 천이확률 및 매개변수 등이 과거와 일정하다는 정상성을 기본 가정으로 하기 때문에 평균의 변동성 등과 같은 외부충격을 모형에 적용할 수 없다. 이러한 관점에서 본 연구의 가장 큰 목적은 기존일강수량 모형을 외부인자를 받아들일 수 있는 모형으로 개발하는 것이다. 즉, Markov Chain 모형의 매개변수인 천이확률과 확률분포형의 매개변수 등을 연결함수(link function)를 통해 외부인자와 연동하도록 하였으며 정준상관분석을 통해 매개변수를 추정하였다. 개발된 모형을 서울지방 1961-2006년까지의 일강수량 자료를 대상으로 검증하는 절차를 가졌다. 추정된 결과를 보면 계절강수량의 특성뿐만 아니라 일강수량의 특성 또한 적절하게 모의되는 것을 확인할 수 있다. 따라서 본 연구에서 개발된 모형은 GCM 예측결과를 입력자료로 활용한다면 일강수계열의 장단기 모의를 위한 downscaling 기법으로 사용될 수 있다. 또한, 기후변화 시나리오가 입력자료로 이용된다면 기후변화에 따른 수자원 영향 평가를 위한 downscaling 기법으로 활용이 가능할 것으로 판단된다. A stationary Markov chain model is a stochastic process with the Markov property. Having the Markov property means that, given the present state, future states are independent of the past states. The Markov chain model has been widely used for water resources design as a main tool. A main assumption of the stationary Markov model is that statistical properties remain the same for all times. Hence, the stationary Markov chain model basically can not consider the changes of mean or variance. In this regard, a primary objective of this study is to develop a model which is able to make use of exogenous variables. The regression based link functions are employed to dynamically update model parameters given the exogenous variables, and the model parameters are estimated by canonical correlation analysis. The proposed model is applied to daily rainfall series at Seoul station having 46 years data from 1961 to 2006. The model shows a capability to reproduce daily and seasonal characteristics simultaneously. Therefore, the proposed model can be used as a short or mid-term prediction tool if elaborate GCM forecasts are used as a predictor. Also, the nonstationary Markov chain model can be applied to climate change studies if GCM based climate change scenarios are provided as inputs.

Structural Analysis of Cointegrating VARs

This survey uses a number of recent developments in the analysis of cointegrating Vector Autoregressions (VARs) to examine their links to the older structural modelling traditions using Autoregressive Distributed Lag (ARDL), and Simultaneous Equations Models (SEMs). In particular, it emphasizes the importance of using judgement and economic theory to supplement the statistical information. After a brief historical review it sets out the statistical framework, discusses the identification of impulse responses using the Generalized Impulse Response functions, reviews the analysis of cointegrating VARs and highlights the large number of choices applied workers have to make in determining a specification. In particular, it considers the problem of specification of intercepts and trends and the size of the VAR in more detail, and examines the advantages of the use of exogenous variables in cointegration analysis. The issues are illustrated with a small U.S. Macroeconomic model.

Exploration and Comparison of New Methods for Electric Demand Forecasting

A variety of time series forecasting methods were tested for the prediction of state-wide monthly residential electrical sales per customer, in a formal experimental design across ten states in the U.S. Models were fitted to historical data from 1962 through 1977, and evaluated for their performance over the following three-year period from January, 1978 through December, 1980. Rolling forecasts were generated on a monthly basis through the evaluation period to obtain 36 one-month forecasts, 35 two-month forecasts, etc. and were reduced to a set of summary statistics for different horizons and averaging methods. The forecasting models included several econometric techniques and several pure time series techniques. The conclusions were that there were gains in accuracy for structural models (use of exogenous variables), for dynamic models (involving lagged dependent variables), and for adaptive models. It was further concluded that the systematic use of model selection criteria such as the Akaike Information Criterion, and of diagnostic tests, led to improved accuracy in the accepted models.