Ultra-short-term forecasting for photovoltaic power plants and real-time key performance indicators analysis with big data solutions. Two case studies - PV Agigea and PV Giurgiu located in Romania

2020 ◽  
Vol 120 ◽  
pp. 103230 ◽  
Author(s):  
Simona-Vasilica Oprea ◽  
Adela Bâra
Keyword(s):  
Big Data ◽  
Real Time ◽  
Case Studies ◽  
Performance Indicators ◽  
Power Plants ◽  
Key Performance Indicators ◽  
Short Term ◽  
Photovoltaic Power ◽  
Short Term Forecasting

Implementation of a Digital Roadmap with Best-in-Class Applications

2021 ◽  
Author(s):  
John McIntosh ◽  
Renata Martin ◽  
Pedro Alcala ◽  
Stian Skjævesland ◽  
John Rigg
Keyword(s):  
Real Time ◽  
Case Studies ◽  
Performance Indicators ◽  
Key Performance Indicators ◽  
Third Party ◽  
Commercial Application ◽  
Time Data ◽  
Operating Model ◽  
Wide Range ◽  
Real Time Data

Abstract The paper describes a project known internally as "InWell" to address multiple requirements in Repsol Drilling & Completions. InWell is defined by a new Operating Model comprising Governance, People, Process, Functions and Technology. This paper addresses changes to the Technology element - often referred to as "Digitalization". The paper includes a discussion about the business transformation strategy and case studies for addressing three of 18 functionalities identified in the first round of development. The InWell development strategy followed four steps; identification of performance issues, envisioning of a future operating model, identification of functionalities required/supporting this operating model and matching to digital solutions. Our case studies focus on three functionalities provided by three separate companies, Unification of Planning and Compliance, Real Time Data aggregation and Key Performance Indicators. Each functionality was addressed with an existing commercial application customized to meet specific requirements. A corporate web-based Well Construction Process (WCP) was initially piloted and then extended to include all well projects. The WCP identifies the key Tasks that must be completed per project, and these are all tracked. Data from this application is used by a third-party Business Analytics application via an API. Real time data from many sites and a wide range of sources was aggregated and standardized, Quality Controlled and stored within a private secure cloud. The data collation service is an essential building block for current third-party applications such as the operating centre and is a prerequisite for the goal of increased automation. A suite of Operator specific Key Performance Indicators (KPIs) and data analytics services were developed for drilling and completions. Homogenized KPIs for all business units provide data for objective performance management and apples-to-apples comparison. Results are presented via custom dashboards, reports, and integrations with third party applications to meet a wide range of requirements. During a four-month Pilot Phase the InWell Project delivered € 2.5 million in tangible savings through improvements in operational performance. In the first 12 months € 16 million in savings were attributed to InWell. By 2022 forecast savings are expected to exceed € 60 million (Figures 1 & 2). The value of Intangible benefits is thought to exceed these objective savings. Figure 1 The Business Case for InWell – Actual & Projected Savings and Costs. Figure 2 InWell Services addressing Value Levers and quantified potential impact. A multi-sourced digital strategy can produce quick gains, is easily adapted, and provides high value at low risk. The full benefit of digital transformation can only be realised when supported by an effective business operating model.

scholarly journals The Impact of Data Filtration on the Accuracy of Multiple Time-Domain Forecasting for Photovoltaic Power Plants Generation

2020 ◽  
Vol 10 (22) ◽  
pp. 8265
Author(s):  
Stanislav A. Eroshenko ◽  
Alexandra I. Khalyasmaa ◽  
Denis A. Snegirev ◽  
Valeria V. Dubailova ◽  
Alexey M. Romanov ◽  
...  
Keyword(s):  
Power Generation ◽  
Time Domain ◽  
Power Plants ◽  
Multiple Time ◽  
Short Term ◽  
Forecasting Accuracy ◽  
Photovoltaic Power ◽  
Operational Forecasting ◽  
Short Term Forecasting ◽  
Forecasting Errors

The paper reports the forecasting model for multiple time-domain photovoltaic power plants, developed in response to the necessity of bad weather days’ accurate and robust power generation forecasting. We provide a brief description of the piloted short-term forecasting system and place under close scrutiny the main sources of photovoltaic power plants’ generation forecasting errors. The effectiveness of the empirical approach versus unsupervised learning was investigated in application to source data filtration in order to improve the power generation forecasting accuracy for unstable weather conditions. The k-nearest neighbors’ methodology was justified to be optimal for initial data filtration, based on the clusterization results, associated with peculiar weather and seasonal conditions. The photovoltaic power plants’ forecasting accuracy improvement was further investigated for a one hour-ahead time-domain. It was proved that operational forecasting could be implemented based on the results of short-term day-ahead forecast mismatches predictions, which form the basis for multiple time-domain integrated forecasting tools. After a comparison of multiple time series forecasting approaches, operational forecasting was realized based on the second-order autoregression function and applied to short-term forecasting errors with the resulting accuracy of 87%. In the concluding part of the article the authors from the points of view of computational efficiency and scalability proposed the hardware system composition.

Digitalization for Effective Performance Management Using Key Performance Indicators in Miscible Gas Injection Projects in South Oman

2021 ◽  
Author(s):  
Kamlesh Kumar ◽  
Tushar Narwal ◽  
Zaal Alias ◽  
Pankaj Agrawal ◽  
Ali Farsi ◽  
...  
Keyword(s):  
Performance Management ◽  
Real Time ◽  
Performance Indicators ◽  
Key Performance Indicators ◽  
Production Optimization ◽  
Utilization Efficiency ◽  
Analysis Tool ◽  
Time Data ◽  
Short Term ◽  
Web Based

Abstract South Oman has several pre-Cambrian reservoirs that are highly pressured (400-1000 bar), deep (3-5 km) and critically sour (H2S up to 10%). The combined STOIIP of these reservoirs makes it one of the largest gas EOR projects in the world. The objective here is to highlight the key performance indicators and digitalization techniques used for continuous and effective well, reservoir and facility management (WRFM) and production optimization, while honoring the facility constraints and gas export requirements. Real time pressure data such as tubing head pressures, injection/production rates along with other data including maps, static pressures and production logs are used to define an appropriate set of performance metric at various levels, e.g. reservoir, sector or well. Digitalization of surveillance data helps in real time production optimization such as offtake management based on creaming curves according to gas sink availability and facility constraints. Key business performance indicators include gas utilization efficiency; MGI performance indicators include incremental oil, throughput, instantaneous and cumulative voidage replacement ratios, gas breakthrough level and time, ratio of reservoir pressure to the target minimum miscibility pressure; and facility constraints are optimized through gas balance, along with tracking field performance against the initial FDP forecasts. Real time performance data is tracked using a commercial Real-Time Data Analysis tool (RTDA) and Database Analytics Visualization Tool (DAVT), with surveillance indicators targeted at well, reservoir and facility level. The above-defined Key Performance Indicators (KPI) are tracked against predictions from the field development plan in web-based portal developed at PDO (Nibras). Digitalization has enabled quick and effective monitoring of these KPI, short-term optimization of injection distribution and offtake rates to maximize oil production and overall value within facilities constraints and varying export gas commitments based on South Oman Gas Line (SOGL) network optimization. Using dimensionless plots and a standardized set of parameters help in developing a common understanding and benchmarking the MGI reservoir response with analogs and amongst different reservoirs. This work presents a set of performance KPIs and short-term optimization methodology using digitalization and LEAN framework that are tracked in a web-based portal, RTDA and DAVT. It provides means to facilitate offtake decisions to meet variable export requirements while honoring facilities constraints, assess reservoir performance, providing valuable insights that helps in speedy reservoir management decisions. This process has been replicated across PDO for all related MGI projects and can benefit other development types, e.g. chemical/steam injection.

scholarly journals Key Performance Indicators for an Energy Community Based on Sustainable Technologies

2021 ◽  
Vol 13 (16) ◽  
pp. 8789
Author(s):  
Giovanni Bianco ◽  
Barbara Bonvini ◽  
Stefano Bracco ◽  
Federico Delfino ◽  
Paola Laiolo ◽  
...  
Keyword(s):  
Performance Indicators ◽  
Urban Areas ◽  
High Performance ◽  
Power Plants ◽  
New Technologies ◽  
Storage Systems ◽  
Clean Energy ◽  
Key Performance Indicators ◽  
Energy Performance ◽  
Green Energy

As reported in the “Clean energy for all Europeans package” set by the EU, a sustainable transition from fossil fuels towards cleaner energy is necessary to improve the quality of life of citizens and the livability in cities. The exploitation of renewable sources, the improvement of energy performance in buildings and the need for cutting-edge national energy and climate plans represent important and urgent topics to be faced in order to implement the sustainability concept in urban areas. In addition, the spread of polygeneration microgrids and the recent development of energy communities enable a massive installation of renewable power plants, high-performance small-size cogeneration units, and electrical storage systems; moreover, properly designed local energy production systems make it possible to optimize the exploitation of green energy sources and reduce both energy supply costs and emissions. In the present paper, a set of key performance indicators is introduced in order to evaluate and compare different energy communities both from a technical and environmental point of view. The proposed methodology was used in order to assess and compare two sites characterized by the presence of sustainable energy infrastructures: the Savona Campus of the University of Genoa in Italy, where a polygeneration microgrid has been in operation since 2014 and new technologies will be installed in the near future, and the SPEED2030 District, an urban area near the Campus where renewable energy power plants (solar and wind), cogeneration units fed by hydrogen and storage systems are planned to be installed.

Statistical Monitoring and Early Forecasting of the Earthquake Sequence: Case Studies after the 2019 M 6.4 Searles Valley Earthquake, California

2020 ◽  
Vol 110 (4) ◽  
pp. 1781-1798 ◽  
Author(s):  
Yosihiko Ogata ◽  
Takahiro Omi
Keyword(s):  
Real Time ◽  
Case Studies ◽  
Seismic Activity ◽  
Southern California ◽  
Aftershock Sequence ◽  
Earthquake Sequence ◽  
Short Term ◽  
Probability Forecast ◽  
Background Rate ◽  
High Background

ABSTRACT This study considers the possible implementation of the operational short-term forecasting, and analysis of earthquake occurrences using a real-time hypocenter catalog of ongoing seismic activity, by reviewing case studies of the aftershocks of the Mw 6.4 Searles Valley earthquake that occurred before the Mw 7.1 Ridgecrest earthquake. First, the short-term prediction of spatiotemporal activity is required in real time along with the background seismic activity over a wide region to obtain practical probabilities of large earthquakes; snapshots from the continuous forecasts during the Searles Valley and Ridgecrest earthquake sequence are included to monitor the growth and migration of seismic activity over time. We found that the area in and around the rupture zone in southern California had a very high background rate. Second, we need to evaluate whether a first strong earthquake may be the foreshock for a further large earthquake; the rupture region in southern California had one of the highest such probabilities. Third, short-term probability forecast of early aftershocks are much desired despite the difficulties with data acquisition. The aftershock sequence of the Mw 6.4 Searles Valley event was found to significantly increase the probability of a larger earthquake, as seen in the foreshock sequence of the 2016 MJMA 7.4 Kumamoto, Japan, earthquake. Finally, detrending the temporal activity of all the aftershocks by stretching and shrinking the ordinary time scale according to the rate given by the Omori–Utsu formula or the epidemic-type aftershock sequence model, we observe the spatiotemporal occurrences in which seismicity patterns may be abnormal, such as relative quiescence, relative activation, or migrating activity. Such anomalies should be recorded and listed for the future evaluation of the probability of a possible precursor for a large aftershock or a new rupture nearby. An example of such anomalies in the aftershocks before the Mw 7.1 Ridgecrest earthquake is considered.

Efficient computation of key performance indicators in a distance learning university

2019 ◽  
Vol 47 (2) ◽  
pp. 96-105 ◽  
Author(s):  
Riccardo Pecori ◽  
Vincenzo Suraci ◽  
Pietro Ducange
Keyword(s):  
Big Data ◽  
Distance Learning ◽  
Performance Indicators ◽  
Key Performance Indicators ◽  
Data Representation ◽  
Efficient Computation ◽  
Educational Decisions ◽  
Content Type ◽  
Educational Sources ◽  
And Performance

Purpose Managing efficiently educational Big Data, produced by Virtual Learning Environments, is becoming a compelling necessity, especially for those universities providing distance learning. This paper aims to propose a possible framework to compute efficiently key performance indicators, summarizing the trends of students’ academic careers, by using educational Big Data. Design/methodology/approach The framework is designed and implemented in a distributed fashion. The parallel computation of the indicators through Map and Reduce nodes is carefully described, together with the workflow of data, from the educational sources to a NoSQL database and to the learning analytics engine. Findings This framework was tested at eCampus University, an Italian distance learning institution, and it was able to significantly reduce the amount of time needed to compute key performance indicators. Moreover, by implementing a proper data representation dashboard, it resulted in a useful help and support for educational decisions and performance analyses and for revealing possible criticalities. Originality/value The framework proposed integrates for the first time, to the best of the authors’ knowledge, a set of modules, designed and implemented in a distributed fashion, to compute key performance indicators for distance learning institutions. It can be used to analyze the dropouts and the outcomes of students and, therefore, to evaluate the performances of universities, which can, in turn, propose effective improvements toward enhancing the overall e-learning scenario.

Sign in / Sign up

Export Citation Format

Share Document