Production forecasting in shale (ultra-low permeability) gas reservoirs is of great interest due to the advent of multi-stage fracturing and horizontal drilling. The well renowned production forecasting model, Arps? Hyperbolic Decline Model, is widely used in industry to forecast shale gas wells. Left unconstrained, the model often overestimates reserves by a great deal. A minimum decline rate is imposed to prevent overestimation of reserves but with less than ten years of production history available to analyze, an accurate minimum decline rate is currently unknown; an educated guess of 5% minimum decline is often imposed. Other decline curve models have been proposed with the theoretical advantage of being able to match linear flow followed by a transition to boundary dominated flow. This thesis investigates the applicability of the Stretched Exponential Production Decline Model (SEPD) and compares it to the industry standard, Arps' with a minimum decline rate. When possible, we investigate an SEPD type curve. Simulated data is analyzed to show advantages of the SEPD model and provide a comparison to Arps' model with an imposed minimum decline rate of 5% where the full production history is known. Long-term production behavior is provided by an analytical solution for a homogenous reservoir with homogenous hydraulic fractures. Various simulations from short-term linear flow (~1 year) to long-term linear flow (~20 years) show the ability of the models to handle onset of boundary dominated flow at various times during production history. SEPD provides more accurate reserves estimates when linear flow ends at 5 years or earlier. Both models provide sufficient reserves estimates for longer-term linear flow scenarios. Barnett Shale production data demonstrates the ability of the models to forecast field data. Denton and Tarrant County wells are analyzed as groups and individually. SEPD type curves generated with 2004 well groups provide forecasts for wells drilled in subsequent years. This study suggests a type curve is most useful when 24 months or less is available to forecast. The SEPD model generally provides more conservative forecasts and EUR estimates than Arps' model with a minimum decline rate of 5%.

This master’s thesis project consists of two parts. Part I of the project compares modeling of historical resource production and forecasting of future production trends using the logit/probit transform advocated by Rutledge (2011) with conventional Hubbert curve fitting, using global coal production as a case study. The conventional Hubbert/Gaussian method fits a curve to historical production data whereas a logit/probit transform uses a linear fit to a subset of transformed production data. Within the errors and limitations inherent in this type of statistical modeling, these methods provide comparable results. That is, despite that apparent goodness-of-fit achievable using the Logit/Probit methodology, neither approach provides a significant advantage over the other in either explaining the observed data or in making future projections. For mature production regions, those that have already substantially passed peak production, results obtained by either method are closely comparable and reasonable, and estimates of ultimately recoverable resources obtained by either method are consistent with geologically estimated reserves. In contrast, for immature regions, estimates of ultimately recoverable resources generated by either of these alternative methods are unstable and thus, need to be used with caution. Although the logit/probit transform generates high quality-of-fit correspondence with historical production data, this approach provides no new information compared to conventional Gaussian or Hubbert-type models and may have the effect of masking the noise and/or instability in the data and the derived fits. In particular, production forecasts for immature or marginally mature production systems based on either method need to be regarded with considerable caution. Part II of the project investigates the utility of a novel alternative method for multicyclic Hubbert modeling tentatively termed “cycle-jumping” wherein overlap of multiple cycles is limited. The model is designed in a way that each cycle is described by the same three parameters as conventional multicyclic Hubbert model and every two cycles are connected with a transition width. Transition width indicates the shift from one cycle to the next and is described as weighted coaddition of neighboring two cycles. It is determined by three parameters: transition year, transition width, and γ parameter for weighting. The cycle-jumping method provides superior model compared to the conventional multicyclic Hubbert model and reflects historical production behavior more reasonably and practically, by better modeling of the effects of technological transitions and socioeconomic factors that affect historical resource production behavior by explicitly considering the form of the transitions between production cycles.

Current methods of production forecasting such as decline curve analysis (DCA) or numerical simulation require years of historical production data, and their accuracy is limited by the choice of model parameters. Unconventional resources have proven challenging to apply traditional methods of production forecasting because they lack long production histories and have extremely variable model parameters. This research proposes a data-driven alternative to reservoir simulation and production forecasting techniques. We create a proxy-well model for predicting cumulative oil production by selecting statistically significant well completion parameters and reservoir information as independent predictor variables in regression-based models. Then, principal component analysis (PCA) is applied to extract key features of a well’s time-rate production profile and is used to estimate cumulative oil production. The efficacy of models is examined on field data of over 400 wells in the Eagle Ford Shale in South Texas, supplied from an industry database. The results of this study can be used to help oil and gas companies determine the estimated ultimate recovery (EUR) of a well and in turn inform financial and operational decisions based on available production and well completion data.

Les marchés de l’électricité sont aujourd’hui en forte transformation, notamment du fait des efforts de libéralisation pour étendre la compétence de gestion du système électrique par le marché. C’est par exemple le cas avec la mise en place de prix nodaux pour gérer les congestions sur le réseau. Par ailleurs, le développement des moyens de production d’électricité d’origine renouvelable met en cause le fonctionnement du système électrique. Dans ce cadre, la question d’investissement pour un producteur éolien se complexifie. Sa rémunération est susceptible à court terme de passer d’un système de subvention, à une rémunération basée sur le produit des ventes sur le marché, fluctuante dans le temps et l’espace (dans le cadre de marchés à prix nodal). Dans ce contexte, ce travail de thèse propose une analyse de l’impact des caractéristiques éoliennes de sites potentiels d’installation, le facteur de charge et la prédictibilité (capacité d’un site à fournir de bonnes prévisions), sur la décision d’investissement. Nous commençons par une analyse statistique pour plusieurs marchés, puis proposons un estimateur du revenu des producteurs éoliens, afin de réaliser le même travail d’une manière moins coûteuse qu’avec un calcul exhaustif. Ensuite, afin de mener ce type d’analyse avec un mix énergétique paramétrable, nous développons un outil de résolution du problème d’optimisation de l’implantation de centrales éoliennes dans un cadre de marché à prix nodal. Il prend en compte une participation au marché de l’électricité la veille pour le lendemain, ainsi que les pénalités versées pour les déviations introduites entre les productions prévues et injectées en temps réel (dues aux erreurs de prévision). Nous faisons l’hypothèse que les productions renouvelables sont suffisamment importantes pour impacter les prix de marché (qui sont également générés avec l’outil), et nous prenons en compte des scénarios pour les productions éoliennes et la demande. Il s’agit donc d’un problème d’optimisation stochastique résolu à l’aide d’une décomposition de Benders. Enfin, nous analysons l’impact du facteur de charge et de la prédictibilité sur l’investissement optimal, selon la configuration pour le coût de la régulation, la capacité des lignes et la corrélation des données éoliennes. / Electricity markets are in a period of intense change. This is notably due to liberalization efforts to increase the extent of electricity system’s management carried out through market operations. One such example is the implementation of nodal prices for network constraints. Moreover, the surge for electricity from renewable sources questions the operation of the electricity system. In this framework, the investment issue for wind producers is becoming more complex. Its income may go from a subsidy-based scheme to a full market participation in the short term, and more volatile according to time and location (in a nodal-pricing scheme). Bearing all this in mind, this PhD work first analyzes the impact of potential installation sites’ characteristics: load factor, and predictability (a site’s ability to enable reliable predictions), on investment. To this end, we carry out a statistical analysis on historical data from several markets, then we suggest an estimator of wind producers revenue, to carry out the same work with a less costly approach than exhaustive calculation. Then, in order to carry out the same kind of analysis, this time in a customizable framework, we build an algorithm to solve the problem of Optimal investment planning of wind turbines within a nodal price market environment. It takes into account the participation in the Day-ahead market as well as penalties paid for imbalances between the energy contracted and injected in real-time (due to forecasting errors). We assume renewable production is important enough to influence market prices which are also generated with our model, and we integrate scenarios for wind production and demand. Therefore we have a stochastic problem which we solve using Benders decomposition. Ultimately we analyze the impact of load factor and predictability on optimal investment according to the chosen setting for regulation cost, line capacities and wind data correlation.

A sudden extreme change in the weather can result in significant impact onthe life system in the present-day scenario. A well-planned prediction for damage during extreme weather conditions can have minimal impact on the grid components and efficient response and recovery models. With technology advancements and innovation in smart grid technologies we can now have accesses to uninterrupted power supply with smart utilization of energy and reduce CO2 emissions. Artificial Intelligence plays a vital role insolving present day power issues. Large amounts of data and rapid usage of computational power has accelerated to use machine learning models topredict and forecast the energy demand. Hence this study aims to determine how machine learning will improve the microgrid operation at Tezpur University. The main application areas studied in this thesis are identified as demand and load forecasting, simulating Photovoltaic (PV)production in a Microgrid and power outages. This thesis is aimed to develop and compare different ML algorithms to test validate and predict the PV production, energy demand and power outages.

[pt] A produção de petróleo possui alta relevância em âmbito brasileiro e mundial. Por outro lado, a incerteza do setor presume alta variabilidade nas previsões de produção de óleo, e exerce um impacto significativo nas decisões. O estudo contempla analisar o cenário da bacia geográfica de Campos, em estudo de caso aplicado em empresa nacional de óleo e gás, com objetivo de aprimorar a previsão de produção de óleo. Para isso, são empregados métodos de simulação, clusterização e previsão, sendo integrados com julgamento humano. Busca-se inferir as incertezas inerentes às atividades, analisar os principais riscos envolvidos e subsidiar a definição das metas de produção. Com esse intuito, foi desenvolvida uma modelagem orientada a dados, por meio da criação de um simulador com linguagem de programação em R. Os dados compreendem os anos de 2017 a 2021, e a projeção é realizada para o ano de 2022. O modelo incorpora julgamento humano durante o processo, permitindo que os especialistas realizem modificações no resultado das previsões, agregando sua experiência e informações exclusivas. A análise de série temporal avalia oito métodos de previsão, seu resultado mostra que a entidade do potencial produtivo apresenta menor erro do que na eficiência, e o método TBATs obteve o menor erro na predição. A análise do planejamento das paradas e entrada dos novos poços é realizada por meio de análise gráfica, observando os principais riscos relacionados. Por fim, o simulador apresenta proposta para auxiliar na definição das metas de produção, ele verifica a probabilidade para alcançar a meta com base nos resultados das simulações. / [en] Oil production has Brazilian and World importance. However, the randomness of the sector results a high variability in oil production forecasts. This variability has a significant impact on decisions. The study analyzes the challenging scenario at geographic Campos basin, in a case applied in a national energy company. The objective is to improve the risk analysis associated with the achievement of oil production targets. Simulation, clustering, and time series forecasting methods are employed, integrating into human judgment. It tries to infer the uncertainties inherent of the activities to increase the accuracy of oil production forecasts, analyze the main risks involved, and subsidize the definition of production targets. A data-driven model is developed, creating a simulator with R language. The data used the years 2017 to 2021, and the projection is made for the year 2022. Human judgment is incorporated into the model during the process, specifying the input parameters to enable experts to make modifications based on the predictions, adding their unique experience and information. The time series analysis eight prediction methods, the results show that the oil potential presents less error than in the production efficiency, and TBATS was the prediction method that obtained the lowest prediction error. The main risks related to the maintenance planning and the entry of new wells are identified through graphical analysis. Finally, the simulator presents a possible solution to help define production goals, it verifies the probability of reaching the goal based on the simulation results.

New decline curve models have been developed to overcome the boundary-dominated flow assumption of the basic Arps’ models, which restricts their application in ultra-low permeability reservoirs exhibiting long-duration transient flow regimes. However, these new decline curve analysis (DCA) methods are still based only on production rate data, relying on the assumption of stable flowing pressure. Since this stabilized state is not reached rapidly in most cases, the applicability of these methods and the reliability of their solutions may be compromised. In addition, production performance predictions cannot be disassociated from the existing operation constraints under which production history was developed. On the other hand, DCA is often carried out without a proper identification of flow regimes. The arbitrary application of DCA models regardless of existing flow regimes may produce unrealistic production forecasts, because these models have been designed assuming specific flow regimes. The main purpose of this study was to evaluate the possible benefits provided by including flowing pressures in production decline analysis. As a result, it have been demonstrated that decline curve analysis based on pressure-normalized rates can be used as a reliable production forecasting technique suited to interpret unconventional wells in specific situations such as unstable operating conditions, limited availability of production data (short production history) and high-pressure, rate-restricted wells. In addition, pressure-normalized DCA techniques proved to have the special ability of dissociating the estimation of future production performance from the existing operation constraints under which production history was developed. On the other hand, it was also observed than more consistent and representative flow regime interpretations may be obtained as diagnostic plots are improved by including MBT, pseudovariables (for gas wells) and pressure-normalized rates. This means that misinterpretations may occur if diagnostic plots are not applied correctly. In general, an improved forecasting ability implies greater accuracy in the production performance forecasts and more reliable reserve estimations. The petroleum industry may become more confident in reserves estimates, which are the basis for the design of development plans, investment decisions, and valuation of companies’ assets.

Even though the theory of storage is historically a popular view to explain commodity futures prices, many authors focus on the oil price link. Past studies have shown an increased futures price volatility on Mondays and days when natural gas storage levels are released, which could both implicate that storage levels and temperature data are incorporated in the prices. In this thesis, the U.S. natural gas storage level change is studied as a function of the consumption and production. Consumption and production are furthered segmented and separately forecasted by modelling inverse problems that are solved by least squares regression using temperature data and timeseries analysis. The results indicate that each consumer consumption segment is highly dependent of the temperature with R2-values of above 90%. However, modelling each segment completely by time-series analysis proved to be more efficient due to lack of flexibility in the polynomials, lack of used weather stations and seasonal patterns in addition to the temperatures. Although the forecasting models could not beat analysts’ consensus estimates, these present natural gas storage level drivers and can thus be used to incorporate temperature forecasts when estimating futures prices.