Stochastic Forecasts Research Articles

Designing Operationally Relevant Daily Large Fire Containment Strategies Using Risk Assessment Results

In this study, we aim to advance the optimization of daily large fire containment strategies for ground-based suppression resources by leveraging fire risk assessment results commonly used by fire managers in the western USA. We begin from an existing decision framework that spatially overlays fire risk assessment results with pre-identified potential wildland fire operational delineations (PODs), and then clusters PODs into a response POD (rPOD) using a mixed integer program (MIP) model to minimize expected loss. We improve and expand upon this decision framework through enhanced fire modeling integration and refined analysis of probabilistic and time-sensitive information. Specifically, we expand the set of data inputs to include raster layers of simulated burn probability, flame length probability, fire arrival time, and expected net value change, all calculated using a common set of stochastic weather forecasts and landscape data. Furthermore, we develop a secondary optimization model that, for a given optimal rPOD, dictates the timing of fire line construction activities to ensure completion of containment line prior to fire arrival along specific rPOD edges. The set of management decisions considered includes assignment of PODs to be included in the rPOD, assignment of suppression resources to protect susceptible structures within the rPOD, and assignment of suppression resources to construct fire lines, on specific days, along the perimeter of the rPOD. We explore how fire manager risk preferences regarding firefighter safety affect optimal rPOD characteristics, and use a simple decision tree to display multiple solutions and support rapid assessment of alternatives. We base our test cases on the FSPro simulation of the 2017 Sliderock Fire that burned on the Lolo National Forest in Montana, USA. The overarching goal of this research is to generate operationally relevant decision support that can best balance the benefits and losses from wildfire and the cost from responding to wildfire.

Evaluation of preprocessing techniques for improving the accuracy of stochastic rainfall forecast models

Accurate rainfall forecasting is one of the most important and challenging hydrological modeling tasks with significant benefits for many sectors of the economy. This study presents novel insight into how to improve the accuracy of a new generation of stochastic monthly rainfall forecast models by examining four different preprocessing techniques: (1) time series modeling without preprocessing which is the common method in stochastic modeling as the base case, (2) preprocess using differencing, spectral analysis seasonal and non-seasonal standardization techniques, (3) two-step preprocessing including stationarization and normalization of data using 8 different transformations, and (4) two-step preprocessing, unlike scenario 3, so that the main time series was normalized and transformed to be stationary. Using the autocorrelation function and partial autocorrelation function diagrams, the parameters of the stochastic model are determined. The results indicate that the proposed data preprocessing normalization and transformation techniques can lead to major improvements in the prediction accuracy of the new monthly rainfall forecast model.

A Model Framework for Stochastic Representation of Uncertainties Associated with Physical Processes in NOAA’s Next Generation Global Prediction System (NGGPS)

Abstract In this study, we propose a physical-process-based stochastic parameterization scheme using cellular automata for NOAA’s Next Generation Global Prediction System. The cellular automata, used to simulate stochastic processes such as the production and destruction of subgrid convective elements, are conditioned on unresolved vertical motion that follows a prescribed stochastically generated skewed distribution (SGS). The SGS is described by a stochastic differential equation and linked to observations by taking into account the first four moments from an observed dataset. In the proposed parameterization framework, we emphasize the need for a dynamical memory term to be included in physical-process-based stochastic parameterizations, and we illustrate the requirement for the dynamical memory using the Mori–Zwanzig formalism. Although this paper focuses on the methodology, early results indicate that if we apply our stochastic framework to deep cumulus convection, it is found that the frequency distribution of precipitation is improved in a single-member stochastic forecast, and some improved spread–skill relationship in ensemble runs can be found in state variables in the tropics, as well as in the subtropics.

Stochastic ensemble climate forecast with an analogue model

Abstract. This paper presents a system to perform large-ensemble climate stochastic forecasts. The system is based on random analogue sampling of sea-level pressure data from the NCEP reanalysis. It is tested to forecast a North Atlantic Oscillation (NAO) index and the daily average temperature in five European stations. We simulated 100-member ensembles of averages over lead times from 5 days to 80 days in a hindcast mode, i.e., from a meteorological to a seasonal forecast. We tested the hindcast simulations with the usual forecast skill scores (CRPS or correlation) against persistence and climatology. We find significantly positive skill scores for all timescales. Although this model cannot outperform numerical weather prediction, it presents an interesting benchmark that could complement climatology or persistence forecast.

Reconfigurable Low Voltage Direct Current Charging Networks for Plug-In Electric Vehicles

An emerging theme in the development of supporting facilities for plug-in electric vehicles (EVs) is the cost-effective planning and utilization of charging networks consistent with the uptake of EVs. This paper proposes a low voltage direct current reconfigurable charging network for plug-in EVs and presents a functional energy management system (EMS) that is capable of planning and operating the charging network to minimize cost and to facilitate progressive infrastructure deployment based on EV demand. The charging network is connected to the main ac grid through one or more centralized ac/dc converters that supply a high power charge to EVs connected to the dc side of the converters. The EMS accommodates multiple parking bays, charging sources, ac constraints, non-linear EV battery loads and user charging requirements with a novel approach to managing user inconvenience. The inconvenience model is founded on the presence of user flexibility, i.e., an allowance on charging time or battery SoC, providing the capability to increase asset utilization and enable access for additional network users. Through a series of case studies and a stochastic forecasting approach, the reconfigurable network and EMS demonstrate the capacity to achieve savings over fixed ac and sequential dc systems.

A reliable linear stochastic daily soil temperature forecast model

Forecasting soil temperature profile is recognized as vital information for irrigation demand forecast in a modern/efficient agricultural water management framework in arid regions. A new linear stochastic model is proposed to more accurately forecast daily soil temperature (DST) profile at depths of 5, 10 and 20 cm below ground surface. The data used to test the proposed new method is collected from two stations in Tabriz and Jolfa, located in the East Azerbaijan Province of Iran. The proposed new method uses four preprocessing techniques, including spectral analysis, standardization, trend removing and differencing. A total of 1680 different modelling scenarios were performed in this study. The results show the superior ability of the proposed methodology in DST estimation, compared to existing nonlinear methods such as the multilayer perceptron neural network (MLPNN), with excellent performance indicators such as the coefficient of determination, mean relative error and the Nash-Sutcliffe index. Moreover, the Akaike Information Criterion (AICc) index is employed to compare the performance of the proposed method with MLPNN in terms of both accuracy and easy-of-use. The AICc of the proposed method at Jolfa at a depth of 5, 10 and 20 cm were 176, -2 and -184, respectively, in comparison with 1991, 30 and -57 for MLPNN. Similarly, the AICc index for Tabriz at 5, 10 and 20 cm are 200, 17 and -152, respectively, for the proposed method and 202, 33 and -62 for MLPNN. Consequently, the proposed new linear method is recommended for forecasting daily soil temperature profiles.

Airfield Infrastructure Management Using Network-Level Optimization and Stochastic Duration Modeling

This paper proposes a facility-specific modeling approach to plan maintenance and rehabilitation (M&R) activities on a network of airport runway pavement facilities. The objective of the modeling approach is to minimize system M&R cost while recommending M&R activities for each runway pavement facility over a planning horizon. To do so, pavement condition forecast is derived from estimating stochastic duration models which capture the inherent uncertainty and dynamics in pavement deterioration and impacts of exogenous factors. Building on the pavement condition forecast, a network optimization-based M&R planning framework is developed which accounts for the interdependence of M&R activities among facilities as reflected in (1) the requirement for aggregate pavement performance and (2) simultaneous implementation of a major M&R action on connected facilities. The budget constraint is also respected. The M&R planning framework with the stochastic duration model-based pavement condition forecast is applied to Chicago O’Hare International Airport. It is found that the proposed approach leads to much reduced M&R cost compared to the state-of-the-practice which does not consider the interdependence of M&R activities among different pavement facilities. On the other hand, accounting for the simultaneous implementation of a major M&R action on connected facilities would substantially increase M&R cost.

Stochastic forecasting of the state of the soil under the roadbed

The stochastic forecasting method was used to simulate the state of the soil under the roadbed for a period of four years. It is shown that the probability of finding soil in the thawed state with a thermal insulator is 0% and without a thermal insulator is 70%.

Стохастическое прогнозирование состояния грунта с учетом радиационного теплообмена

Стохастическое прогнозирование состояния грунта с учетом радиационного теплообмена

Impacts of stochastic forecast errors of renewable energy generation and load demands on microgrid operation

Abstract Evaluating the impacts related to stochastic forecast errors (SFE) of renewable energy generation and load demands on the operation of micro grid is an important issue. In previous researches, the negative effects of SFE on micro grid are mainly focusing on power quality, system control, and operation, etc. The objective of this study is to investigate how SFE affects the operation stability of energy storage system (ESS) from the perspective of the state of charge (SOC). To this end, novel SFE propagation and accumulation models are introduced. Efficient quantification models of impacts of SFE on the variance of SOC are presented. These models elucidate in detail the mechanism of ESS working from stable to unstable state due to SOC deviation. Then, some metrics are introduced to evaluate the impacts of SFE on service lifetime of units, the cycle life of ESS and operation economy of micro grid. Finally, illustrative numerical results are provided and compared using a self-developed new simulation program named MG-ROS, which can simulate the actual operation of micro grid under SFE.

On the Simulation of Aggregated Solar PV Forecast Errors

The uncertainty arising from high levels of solar photovoltaic (PV) penetration can have a substantial impact on power system operation. Therefore, there is a need to develop models capable of representing PV generation in a rigorous manner. This paper introduces a novel transformation-based methodology to generate stochastic solar area power forecast scenarios; easy to apply to new locations. We present a simulation study comparing day-ahead solar forecast errors covering regions with different geographical sizes, total installed capacities, and climatic characteristics. The results show that our model can capture the spatio-temporal properties and match the long-term statistical properties of actual data. Hence, it can be used to characterize the PV input uncertainty in power system studies.

Stochastic Forecasting of Demographic Components Based on Principal Component

Stochastic Forecasting of Demographic Components Based on Principal Component

Stochastic modelling of the monthly average maximum and minimum temperature patterns in India 1981–2015

The paper investigates the stochastic modelling and forecasting of monthly average maximum and minimum temperature patterns through suitable seasonal auto regressive integrated moving average (SARIMA) model for the period 1981–2015 in India. The variations and distributions of monthly maximum and minimum temperatures are analyzed through Box plots and cumulative distribution functions. The time series plot indicates that the maximum temperature series contain sharp peaks in almost all the years, while it is not true for the minimum temperature series, so both the series are modelled separately. The possible SARIMA model has been chosen based on observing autocorrelation function (ACF), partial autocorrelation function (PACF), and inverse autocorrelation function (IACF) of the logarithmic transformed temperature series. The SARIMA (1, 0, 0) × (0, 1, 1)12 model is selected for monthly average maximum and minimum temperature series based on minimum Bayesian information criteria. The model parameters are obtained using maximum-likelihood method with the help of standard error of residuals. The adequacy of the selected model is determined using correlation diagnostic checking through ACF, PACF, IACF, and p values of Ljung–Box test statistic of residuals and using normal diagnostic checking through the kernel and normal density curves of histogram and Q–Q plot. Finally, the forecasting of monthly maximum and minimum temperature patterns of India for the next 3 years has been noticed with the help of selected model.

Efficient River Management using Stochastic MPC and Ensemble Forecast of Uncertain In-flows

Efficient river management is essential in improving water resource utilisation. However, river flows and water-levels are affected by unregulated in- and out-flows. Therefore, it is important to consider the forecasts of these unregulated flows and the uncertainties in the forecasts. The paper describes control and modelling tools from the literature that suit the river management problem. Specifically, a scenario-based Stochastic Model Predictive Control (MPC) strategy, that makes use of ensemble forecast of unregulated flows, is proposed, where the ensemble forecast contains multiple flow scenarios to characterise future flows and their uncertainties, and are obtained from catchment hydrological models.

Hybrid Stochastic Forecasting Model for Management of Large Open Water Reservoir with Storage Function

The main advantage of stochastic forecasting is fan of possible value whose deterministic method of forecasting could not give us. Future development of random process is described better by stochastic then deterministic forecasting. Discharge in measurement profile could be categorized as random process. Content of article is construction and application of forecasting model for managed large open water reservoir with supply function. Model is based on neural networks (NS) and zone models, which forecasting values of average monthly flow from inputs values of average monthly flow, learned neural network and random numbers. Part of data was sorted to one moving zone. The zone is created around last measurement average monthly flow. Matrix of correlation was assembled only from data belonging to zone. The model was compiled for forecast of 1 to 12 month with using backward month flows (NS inputs) from 2 to 11 months for model construction. Data was got ridded of asymmetry with help of Box-Cox rule (Box, Cox, 1964), value r was found by optimization. In next step were data transform to standard normal distribution. The data were with monthly step and forecast is not recurring. 90 years long real flow series was used for compile of the model. First 75 years were used for calibration of model (matrix input-output relationship), last 15 years were used only for validation. Outputs of model were compared with real flow series. For comparison between real flow series (100% successfully of forecast) and forecasts, was used application to management of artificially made reservoir. Course of water reservoir management using Genetic algorithm (GE) + real flow series was compared with Fuzzy model (Fuzzy) + forecast made by Moving zone model. During evaluation process was founding the best size of zone. Results show that the highest number of input did not give the best results and ideal size of zone is in interval from 25 to 35, when course of management was almost same for all numbers from interval. Resulted course of management was compared with course, which was obtained from using GE + real flow series. Comparing results showed that fuzzy model with forecasted values has been able to manage main malfunction and artificially disorders made by model were founded essential, after values of water volume during management were evaluated. Forecasting model in combination with fuzzy model provide very good results in management of water reservoir with storage function and can be recommended for this purpose.

Stochastic Forecasting of Labor Supply and Population: An Integrated Model

This paper presents a stochastic model to forecast the German population and labor supply until 2060. Within a cohort-component approach, our population forecast applies principal components analysis to birth, mortality, emigration, and immigration rates, which allows for the reduction of dimensionality and accounts for correlation of the rates. Labor force participation rates are estimated by means of an econometric time series approach. All time series are forecast by stochastic simulation using the bootstrap method. As our model also distinguishes between German and foreign nationals, different developments in fertility, migration, and labor participation could be predicted. The results show that even rising birth rates and high levels of immigration cannot break the basic demographic trend in the long run. An important finding from an endogenous modeling of emigration rates is that high net migration in the long run will be difficult to achieve. Our stochastic perspective suggests therefore a high probability of substantially decreasing the labor supply in Germany.

Combined deterministic – stochastic forecasting of monthly river flows for water management

Monthly discharges can be modelled and predicted by the decomposition of the runoff process model into two components – deterministic and stochastic. For such approach the term hybrid was often adopted. In this study a hybrid (deterministic-stochastic) modelling approach for one step ahead forecasting of mean monthly discharges at the gauging stations Banská Bystrica and Brehy on the Hron River in Slovakia was developed. The aim was to join the conceptual monthly rainfall runoff model KVHK and several time series models of the forecasting error time series of the conceptual model into a hybrid framework. Since these rainfall-runoff model error series may exhibit nonstationarity and heteroscedasticity, beside traditional ARMA models, GARCH type nonlinear time series models were also considered.

A Novel Hybrid BND-FOA-LSSVM Model for Electricity Price Forecasting

Accurate electricity price forecasting plays an important role in the profits of electricity market participants and the healthy development of electricity market. However, the electricity price time series hold the characteristics of volatility and randomness, which make it quite hard to forecast electricity price accurately. In this paper, a novel hybrid model for electricity price forecasting was proposed combining Beveridge-Nelson decomposition (BND) method, fruit fly optimization algorithm (FOA), and least square support vector machine (LSSVM) model, namely BND-FOA-LSSVM model. Firstly, the original electricity price time series were decomposed into deterministic term, periodic term, and stochastic term by using BND model. Then, these three decomposed terms were forecasted by employing LSSVM model, respectively. Meanwhile, to improve the forecasting performance, a new swarm intelligence optimization algorithm FOA was used to automatically determine the optimal parameters of LSSVM model for deterministic term forecasting, periodic term forecasting, and stochastic term forecasting. Finally, the forecasting result of electricity price can be obtained by multiplying the forecasting values of these three terms. The results show the mean absolute percentage error (MAPE), root mean square error (RMSE) and mean absolute error (MAE) of the proposed BND-FOA-LSSVM model are respectively 3.48%, 11.18 Yuan/MWh and 9.95 Yuan/MWh, which are much smaller than that of LSSVM, BND-LSSVM, FOA-LSSVM, auto-regressive integrated moving average (ARIMA), and empirical mode decomposition (EMD)-FOA-LSSVM models. The proposed BND-FOA-LSSVM model is effective and practical for electricity price forecasting, which can improve the electricity price forecasting accuracy.

A Probabilistic Short-Term Water Demand Forecasting Model Based on the Markov Chain

This paper proposes a short-term water demand forecasting method based on the use of the Markov chain. This method provides estimates of future demands by calculating probabilities that the future demand value will fall within pre-assigned intervals covering the expected total variability. More specifically, two models based on homogeneous and non-homogeneous Markov chains were developed and presented. These models, together with two benchmark models (based on artificial neural network and naïve methods), were applied to three real-life case studies for the purpose of forecasting the respective water demands from 1 to 24 h ahead. The results obtained show that the model based on a homogeneous Markov chain provides more accurate short-term forecasts than the one based on a non-homogeneous Markov chain, which is in line with the artificial neural network model. Both Markov chain models enable probabilistic information regarding the stochastic demand forecast to be easily obtained.

Stochastic forecasting of project streams for construction project portfolio management

BackgroundConstruction companies typically work on many projects simultaneously each with its own objectives and resource demands. Consequently, a key managerial function is to allocate financial, equipment, and human resources between these concurrent projects in a way that satisfies the individual project constraints while optimizing the company’s overall objectives.Project portfolio management is concerned with managing multiple projects to accomplish strategic goals. The main concentrations of research so far have been project selection, prioritization and alignment of a portfolio with strategic goals among a pool of awarded projects. The objective of this research is to develop a model capable of generating representative future streams of projects that can be used to assist in strategic planning and portfolio management.MethodsThis paper reports on an on-going research project aimed at developing and validating a stochastic model of streams of uncertain and unknown future projects and to demonstrate the significance and implications of such uncertainties on project portfolio and strategic planning. The scope of the research is Florida Department of Transportation (FDOT) projects. Records of the past 12 years letting information combined with a pool of candidate variables is analyzed to capture characteristics of the time series data and to determine any project characteristic correlations with macroeconomic factors.ResultsThe performances of the various model components presented indicate the viability of an integrated project stream forecaster that predicts, within a simulation environment, the frequencies of projects and empirical distributions of project duration and cost.ConclusionThe gap in the current body of knowledge is a lack of consideration of the effects of uncertainties associated with future projects, both known (but yet to be awarded to a contractor) and unknown (although statistically quantifiable). It is evident that companies should not just focus on current known projects but also uncertain and unknown future projects. Such a capability, looking into the future, is key to the effective medium and long-term strategic planning for the company. Contractors can, for example, use these stochastic streams of data to test different bidding strategies and to see how sensitive the performance of their portfolio is to changes in different market factors.