Day-Ahead Forecasting for Small-Scale Photovoltaic Power Based on Similar Day Detection with Selective Weather Variables

As photovoltaic (PV) power plants are an essential component of modern smart grids, the PV generation forecasting of such plants has recently been gaining interest. The forecasting results of PV power often suffer from large errors because of unusual weather conditions. In a learning-based forecasting model, the forecasting accuracy can be enhanced by using carefully selected data for training rather than all the data without any screening. That is, using a training set that only contains information obtained from similar days can help enhance the accuracy of learning-based PV forecasting. This paper proposes a forecasting method for small-scale PV generation. This method is based on long short-term memory; further, it detects similar days considering the different impacts of weather variables on PV power according to the day. This method can address issues caused by unnecessary learning from non-similar historical days. The simulation results demonstrate that the proposed method exhibits better performance than do existing similar day detection methods.

Download Full-text

Multi-Site Photovoltaic Forecasting Exploiting Space-Time Convolutional Neural Network

Energies ◽

10.3390/en12234490 ◽

2019 ◽

Vol 12 (23) ◽

pp. 4490 ◽

Cited By ~ 5

Author(s):

Jaeik Jeong ◽

Hongseok Kim

Keyword(s):

Neural Network ◽

New York ◽

Convolutional Neural Network ◽

Smart Grids ◽

Power Plants ◽

Temporal Correlation ◽

Space Time ◽

Percentage Error ◽

Virtual Power Plants ◽

Pv Generation

The accurate forecasting of photovoltaic (PV) power generation is critical for smart grids and the renewable energy market. In this paper, we propose a novel short-term PV forecasting technique called the space-time convolutional neural network (STCNN) that exploits the location information of multiple PV sites and historical PV generation data. The proposed structure is simple but effective for multi-site PV forecasting. In doing this, we propose a greedy adjoining algorithm to preprocess PV data into a space-time matrix that captures spatio-temporal correlation, which is learned by a convolutional neural network. Extensive experiments with multi-site PV generation from three typical states in the US (California, New York, and Alabama) show that the proposed STCNN outperforms the conventional methods by up to 33% and achieves fairly accurate PV forecasting, e.g., 4.6–5.3% of the mean absolute percentage error for a 6 h forecasting horizon. We also investigate the effect of PV sites aggregation for virtual power plants where errors from some sites can be compensated by other sites. The proposed STCNN shows substantial error reduction by up to 40% when multiple PV sites are aggregated.

Download Full-text

Forecast Quality of Physics-Based and Data-Driven PV Performance Models for a Small-Scale PV System

Frontiers in Energy Research ◽

10.3389/fenrg.2021.639346 ◽

2021 ◽

Vol 9 ◽

Author(s):

Moritz Stüber ◽

Felix Scherhag ◽

Matthieu Deru ◽

Alassane Ndiaye ◽

Muhammad Moiz Sakha ◽

...

Keyword(s):

Smart Grids ◽

Time Series Data ◽

Short Term Memory ◽

Data Driven ◽

Series Data ◽

Small Scale ◽

Pv System ◽

Advantages And Disadvantages ◽

Forecast Quality

In the context of smart grids, the need for forecasts of the power output of small-scale photovoltaic (PV) arrays increases as control processes such as the management of flexibilities in the distribution grid gain importance. However, there is often only very little knowledge about the PV systems installed: even fundamental system parameters such as panel orientation, the number of panels and their type, or time series data of past PV system performance are usually unknown to the grid operator. In the past, only forecasting models that attempted to account for cause-and-effect chains existed; nowadays, also data-driven methods that attempt to recognize patterns in past behavior are available. Choosing between physics-based or data-driven forecast methods requires knowledge about the typical forecast quality as well as the requirements that each approach entails. In this contribution, the achieved forecast quality for a typical scenario (day-ahead, based on numerical weather predictions [NWP]) is evaluated for one physics-based as well as five different data-driven forecast methods for a year at the same site in south-western Germany. Namely, feed-forward neural networks (FFNN), long short-term memory (LSTM) networks, random forest, bagging and boosting are investigated. Additionally, the forecast quality of the weather forecast is analyzed for key quantities. All evaluated PV forecast methods showed comparable performance; based on concise descriptions of the forecast approaches, advantages and disadvantages of each are discussed. The approaches are viable even though the forecasts regularly differ significantly from the observed behavior; the residual analysis performed offers a qualitative insight into the achievable forecast quality in a typical real-world scenario.

Download Full-text

Small-Scale Smart Electrical Grid Design, Construction, and Analysis

Volume 5: Education and Globalization ◽

10.1115/imece2016-65219 ◽

2016 ◽

Author(s):

Andrew Craig ◽

Xiaokuan Li ◽

Patrick Sesker ◽

Alex Mcinerny ◽

Thomas DeAgostino ◽

...

Keyword(s):

Smart Grid ◽

Smart Grids ◽

Energy Demand ◽

Power Plants ◽

Theoretical Models ◽

Industrial Building ◽

Small Scale ◽

Lead Acid Battery ◽

Research Activities ◽

Lead Acid

As society moves into the digital age, the expectation of instantaneous electricity at the flip of a switch is more prominent than ever. The traditional electric grid has become outdated and Smart Grids are being developed to deliver reliable and efficient energy to consumers. However, the costs involved with implementing their infrastructure often limits research to theoretical models. As a result, an undergraduate capstone design team constructed a small-scale 12 VDC version to be used in conjunction with classroom and research activities. In this model Smart Grid, two houses act as residential consumers, an industrial building serves as a high-load demand device, and a lead-acid battery connected to a 120 VAC wall outlet simulates fossil fuel power plants. A smaller lead-acid battery provides a microgrid source while a photovoltaic solar panel adds renewable energy into the mix and can charge either lead-acid battery. All components are connected to a National Instruments CompactRIO system while being controlled and monitored via a LabVIEW software program. The resulting Smart Grid can run independently based on constraints related to energy demand, cost, efficiency, and environmental impact. Results are shown demonstrating choices based on these constraints, including a corresponding weighting according to controller objectives.

Download Full-text

SUBSTANTIATION AND RESULTS OF PRACTICAL RESEARCHES OF THE PLOUGH FOR DIGGING POTATOES IN WET WEATHER CONDITIONS

Vestnik of Kazan state agrarin university ◽

10.12737/2073-0462-2020-75-80 ◽

2020 ◽

Vol 14 (4) ◽

pp. 75-80

Author(s):

Petr Smirnov ◽

Nikolay Pushkarenko ◽

Mihail Smirnov ◽

Evgeniy Alekseev

Keyword(s):

Power Plants ◽

Analytical Calculation ◽

Weather Conditions ◽

The Body ◽

Small Scale ◽

Soil Reaction ◽

Central Russia ◽

Average Hardness ◽

Graphic Analysis

In Central Russia, wet conditions for potato harvesting occur on average once every 5 ... 6 years. The task is practically not feasible for small-scale agricultural production - peasant farming (peasant farms) and smallholding, because of they are limited both in power plants (tractors) and in potato-harvesting equipment. Therefore, it is proposed to dig up potatoes with a very common three-unit plow, followed by picking in a mechanized way. For the turnover of the tuberous layer, a trellised conical blade is proposed. A graphic analysis of the turnover of the reservoir with the determination of the secretive zones of loss of tubers was carried out, an analytical calculation with the task of stabilizing the course of the plow on the potato ridge, since, compared with field plowing, the plow body plank is placed in the furrow and the soil reaction to it is not sufficiently implemented. Formulas for calculating the area of the field board are given. For calculations, the average hardness of the soil of the ridge and furrow of the potato plantings was determined experimentally with an absolute soil moisture of 31%. According to the calculations, enlarged field boards were made and installed deeper than the body stroke. Field boards with an increased area of 1.5 ... 1.6 times stabilized the plow. The changes made to the frame, the support wheel and the trailed (mounted) plow system are also shown.

Download Full-text

Virtual Energy Storage in RES-Powered Smart Grids with Nonlinear Model Predictive Control

Energies ◽

10.3390/en14041082 ◽

2021 ◽

Vol 14 (4) ◽

pp. 1082 ◽

Cited By ~ 1

Author(s):

Dimitrios Trigkas ◽

Chrysovalantou Ziogou ◽

Spyros Voutetakis ◽

Simira Papadopoulou

Keyword(s):

Energy Storage ◽

Smart Grids ◽

Real Data ◽

Weather Conditions ◽

Energy Sources ◽

Small Scale ◽

Auxiliary Power ◽

Predictive Controller ◽

Central Energy ◽

Control And Management

The integration of a variety of heterogeneous energy sources and different energy storage systems has led to complex infrastructures and made apparent the urgent need for efficient energy control and management. This work presents a non-linear model predictive controller (NMPC) that aims to coordinate the operation of interconnected multi-node microgrids with energy storage capabilities. This control strategy creates a superstructure of a smart-grid consisting of distributed interconnected microgrids, and has the ability to distribute energy among a pool of energy storage means in an optimal way, formulating a virtual central energy storage platform. The goal of this work is the optimal exploitation of energy produced and stored in multi-node microgrids, and the reduction of auxiliary energy sources. A small-scale multi-node microgrid was used as a basis for the mathematical modelling and real data were used for the model validation. A number of operation scenarios under different weather conditions and load requests, demonstrates the ability of the NMPC to supervise the multi-node microgrid resulting to optimal energy management and reduction of the auxiliary power devices operation.

Download Full-text

Machine Learning Schemes for Anomaly Detection in Solar Power Plants

10.20944/preprints202112.0481.v1 ◽

2021 ◽

Author(s):

Mariam Ibrahim ◽

Ahmad Alsheikh ◽

Feras M. Awaysheh ◽

Mohammad Dahman Alshehri

Keyword(s):

Machine Learning ◽

Anomaly Detection ◽

Smart Grids ◽

Power Plants ◽

Short Term Memory ◽

Solution Space ◽

Machine Learning Techniques ◽

Industrial Growth ◽

Trade Offs ◽

Learning Techniques

The rapid industrial growth in solar energy is gaining increasing interest in renewable power from smart grids and plants. Anomaly detection in photovoltaic (PV) systems is a demanding task. In this sense, it is vital to utilize recent advances in machine learning to accurately and timely detect different anomalies and condition monitoring. This paper addresses this issue by evaluating different machine learning techniques and schemes and showing how to apply these approaches to solve anomaly detection and detect faults on photovoltaic components. For this, we apply distinct state-of-the-art machine learning techniques (AutoEncoder Long Short-Term Memory (AE-LSTM), Facebook-Prophet, and Isolation Forest) to detect faults/anomalies and evaluate their performance. These models shall identify the PV system's healthy and abnormal actual behaviors. Our results provide clear insights to make an informed decision, especially with experimental trade-offs for such complex solution space.

Download Full-text

Impacts of Weather on Short-Term Metro Passenger Flow Forecasting Using a Deep LSTM Neural Network

Applied Sciences ◽

10.3390/app10082962 ◽

2020 ◽

Vol 10 (8) ◽

pp. 2962

Author(s):

Lijuan Liu ◽

Rung-Ching Chen ◽

Shunzhi Zhu

Keyword(s):

Neural Network ◽

Short Term Memory ◽

Weather Conditions ◽

Urban Transport ◽

Weather Data ◽

Weather Variables ◽

Short Term ◽

Passenger Flow ◽

Input Variables ◽

Main Station

Metro systems play a key role in meeting urban transport demands in large cities. The close relationship between historical weather conditions and the corresponding passenger flow has been widely analyzed by researchers. However, few studies have explored the issue of how to use historical weather data to make passenger flow forecasting more accurate. To this end, an hourly metro passenger flow forecasting model using a deep long short-term memory neural network (LSTM_NN) was developed. The optimized traditional input variables, including the different temporal data and historical passenger flow data, were combined with weather variables for data modeling. A comprehensive analysis of the weather impacts on short-term metro passenger flow forecasting is discussed in this paper. The experimental results confirm that weather variables have a significant effect on passenger flow forecasting. It is interesting to find out that the previous variables of one-hour temperature and wind speed are the two most important weather variables to obtain more accurate forecasting results on rainy days at Taipei Main Station, which is a primary interchange station in Taipei. Compared to the four widely used algorithms, the deep LSTM_NN is an extremely powerful method, which has the capability of making more accurate forecasts when suitable weather variables are included.

Download Full-text

Hour-Ahead Solar Irradiance Forecasting Using Multivariate Gated Recurrent Units

Energies ◽

10.3390/en12214055 ◽

2019 ◽

Vol 12 (21) ◽

pp. 4055 ◽

Cited By ~ 8

Author(s):

Jessica Wojtkiewicz ◽

Matin Hosseini ◽

Raju Gottumukkala ◽

Terrence Lynn Chambers

Keyword(s):

Power Generation ◽

Solar Irradiance ◽

Cloud Cover ◽

Power Plants ◽

Solar Power ◽

Short Term Memory ◽

Time Of Day ◽

Sequential Data ◽

Weather Variables ◽

Gated Recurrent Units

Variation in solar irradiance causes power generation fluctuations in solar power plants. Power grid operators need accurate irradiance forecasts to manage this variability. Many factors affect irradiance, including the time of year, weather and time of day. Cloud cover is one of the most important variables that affects solar power generation, but is also characterized by a high degree of variability and uncertainty. Deep learning methods have the ability to learn long-term dependencies within sequential data. We investigate the application of Gated Recurrent Units (GRU) to forecast solar irradiance and present the results of applying multivariate GRU to forecast hourly solar irradiance in Phoenix, Arizona. We compare and evaluate the performance of GRU against Long Short-Term Memory (LSTM) using strictly historical solar irradiance data as well as the addition of exogenous weather variables and cloud cover data. Based on our results, we found that the addition of exogenous weather variables and cloud cover data in both GRU and LSTM significantly improved forecasting accuracy, performing better than univariate and statistical models.

Download Full-text