A base case model is always required. Higher probability intervals are often so large that they are difficult to use in risk qualified decision making. There are some statistical methods to establish P10 and P90 reserve figures.
The conventional approaches to estimate the reserves are divided into deterministic and probabilistic methods. The deterministic approach consists of volumetric, material balance and decline curve analysis and they use a single value for each parameter for estimating the reserves, there are no P10, P50 and P90 values in this method. The probabilistic approach uses a full range of values for each parameter in the reserve calculation.
For example, the volumetric method could use a distribution of values for porosity, initial water saturation, formation volume factor and so on to get a range of values for the reserve. P90 refers to proved reserves, P50 refers to proved and probable reserves and finally P10 refers to proved, probable and possible reserves.
In this paper, P10 refers to a p-value of 0. The problem with conventional statistical methods is that there are no specific realizations. It is not possible to run a flow simulator and assess the dynamic performance of the models under different conditions. The traditional geostatistical approach to finding P10 and P90 models is based on ranking procedures.
Multiple realizations often are generated, and then some quick-to-calculate static reservoir attribute such as connected pore volume is chosen to rank the realizations. Realizations with specific position in the distribution of the static response are selected. Sign In or Register. Advanced Search. Sign In. Skip Nav Destination Proceeding Navigation. Close mobile search navigation. Controllable verses non controllable forecast factors. Discounting and risking in production forecasting.
Documentation and reporting in production forecasting. Empirical methods in production forecasting. Establishing input for production forecasting. Integrated asset modelling in production forecasting. Long term verses short term production forecast.
Look backs and forecast verification. Material balance models in production forecasting. Probabilistic verses deterministic in production forecasting. Production forecasting activity scheduling. Production forecasting analog methods. Production forecasting building blocks.
Production forecasting decline curve analysis. Production forecasting expectations. Production forecasting flowchart. Production forecasting frequently asked questions and examples.
Production forecasting in the financial markets. Production forecasting principles and definition. Production forecasting purpose. The standard data deliveries include information about the model uncertainty referring to yearly GHI estimates. The general uncertainty information is provided in PDF data reports, and on request it can be more accurately specified with regard to the region of interest. The model uncertainty already includes the uncertainties related to the measurements used for the model validation.
Interannual variability. Weather changes year-by-year, in longer-term cycles and has also stochastic nature. Therefore, solar radiation, air temperature and PV energy yield in each year can deviate from the long-term average to some extent, and this is called interannual variability.
It can be calculated from the historical time series as a standard deviation of the series of annual values. If the interannual variability for a period of N years is being considered, then the STDEV is to be divided by the square root of N typically one year, 10 years, or the total expected lifetime of the solar energy asset. For single year this uncertainty is highest, and it decreases with number of years.
In P90 energy calculation, the case of variability that can be expected at any single year is typically assumed. On request, calculation of variability over longer period 10, 20 or 25 years is also provided. Optimally, interannual variabilityof PV power production is calculated from full historical time series.
In case that TMY data is used this is not possible and therefore a less accurate assumption of GHI variability is applied. Uncertainty of energy simulation model.
This considers the imperfections of PV energy simulation models, which provide values of expected energy yield. Various uncertainty factors affecting PV energy production e. The final P90 Pxx is obtained by combining P50 with all factors of uncertainty expressed for the same exceedance level It is quite common to see the uncertainty expressed in terms of standard deviation STDEV , which represents a confidence interval equivalent to approximately Different calculation approaches may give different results Solargis offers 3 type of hourly datasets that can be used for simulation of expected energy output for P50, P90, and other Pxx scenarios.
If expressed in hourly intervals, it has values per each year value for the leap years of data available. The sample dataset below has more than , values for each parameter. Download sample data file for hourly time series CSV, The benefit of TMY is size of the data file allowing faster speed of calculation.
The disadvantage is the loss of various less typical weather patterns. In a simplified way, it can be considered that it represents a year that can occur once in 10 years.
Thus, it is suitable for simulation of conservative PV energy yield scenarios. This dataset is generated by concatenating months representing lower summaries of solar radiation so that the annual value is close to P90 taking into account a combined effect of the solar model uncertainty and GHI interannual variability that can be observed at any single year.
0コメント