Solar Power Forecasting Research Articles

Semi-asynchronous personalized federated learning for short-term photovoltaic power forecasting

Accurate forecasting for photovoltaic power generation is one of the key enablers for the integration of solar photovoltaic systems into power grids. Existing deep-learning-based methods can perform well if there are sufficient training data and enough computational resources. However, there are challenges in building models through centralized shared data due to data privacy concerns and industry competition. Federated learning is a new distributed machine learning approach which enables training models across edge devices while data reside locally. In this paper, we propose an efficient semi-asynchronous federated learning framework for short-term solar power forecasting and evaluate the framework performance using a CNN-LSTM model. We design a personalization technique and a semi-asynchronous aggregation strategy to improve the efficiency of the proposed federated forecasting approach. Thorough evaluations using a real-world dataset demonstrate that the federated models can achieve significantly higher forecasting performance than fully local models while protecting data privacy, and the proposed semi-asynchronous aggregation and the personalization technique can make the forecasting framework more robust in real-world scenarios.

Support Vector Quantile Regression for the Post-Processing of Meso-Scale Ensemble Prediction System Data in the Kanto Region: Solar Power Forecast Reducing Overestimation

Although the recent development of solar power forecasting through machine learning approaches, such as the machine learning models based on numerical weather prediction (NWP) data, has been remarkable, their extreme error requires an increase in the amount of reserve capacity procurement used for the power system safety. Hence, a reduction of the serious overestimation is necessary for efficient grid operation. However, despite the importance of the above issue, few studies have focused on the model design, suppressing serious errors, to the best of the authors’ knowledge. This study investigates a prediction model that can reduce the huge overestimation of the solar irradiance, which poses a risk to the power system. The specific approaches used are as follows: the employment of Support Vector Quantile Regression (SVQR), the utilization of Meso-scale Ensemble Prediction System (MEPS, Meso-scale EPS for the regions of Japan) data, which is based on the forecasts from Meso-scale Model (MSM) as explanatory variables, and the hyperparameter adjustment. The performance of the models is verified in the one day-ahead forecasting for surface solar irradiance at five sites in the Kanto region as the numerical simulation, where their forecasting errors are measured by the root mean square error (RMSE) and the 3σ error, which corresponds to the 99.87% quantile error of the order statistics. The test results indicate the following findings: the SVRs’ RMSE and 3σ error tend to be trade-offs in the case of varying the penalty of the regularization term; by using SVR as a post-processing tool for MSM or MEPS data, both of the score of their metrics can be improved from original data; the MEPS-based SVQR (MEPS-SVQR) could provide superior performance in both metrics in comparison with the MSM-based SVQR (MSM-SVQR) if the parameters are properly adjusted. Although the time period and the type of MEPS data used for the validation are limited, our report is expected to help the design of NWP-based machine learning models to enable short-term solar power forecasts with a low risk of overestimation.

Determination of the Concepts of Building a Solar Power Forecasting Model

Since in Ukraine there are fines for imbalances in solar power generation in the “day-ahead” energy market, the forecasting of electricity generation is an important component of the solar power plant operation. To forecast the active power generation of photovoltaic panels, a mathematical model should be developed, which considers the main factors affecting the volume of energy generation. In this article, the main factors affecting the performance of solar panels were analysed using correlation analysis. The data sets for the construction of the forecasting model were obtained from the solar power plant in the Kyiv region. Two types of data sets were used for the analysis of factors and model building: 10-minute time interval data and daily data. For each data set, the input parameters were selected using correlation analysis. Considering the determining factors, the models of finding the function of reflecting meteorological factors in the volume of electricity generation are built. It is established that through models with a lower discreteness of climatic parameters forecast it is possible to determine the potential volume of electricity production by the solar power plant for the day-ahead with a lower mean absolute error. The best accuracy of the model for predicting electric power generation over the 10-minute interval is obtained in the ensemble random of a forest model. It is determined that models without solar radiation intensity parameters on the input have an unsatisfactory coefficient of determination. Therefore, further research will focus on combining a model of forecasting the day-ahead solar radiation with 10-minutes discreteness with a model for determining the amount of electricity generation. The determined predicted values of solar radiation will be the input parameter of the forecasting model described in the article

Increasing the Accuracy of Hourly Multi-Output Solar Power Forecast with Physics-Informed Machine Learning.

Machine Learning (ML)-based methods have been identified as capable of providing up to one day ahead Photovoltaic (PV) power forecasts. In this research, we introduce a generic physical model of a PV system into ML predictors to forecast from one to three days ahead. The only requirement is a basic dataset including power, wind speed and air temperature measurements. Then, these are recombined into physics informed metrics able to capture the operational point of the PV. In this way, the models learn about the physical relationships of the different features, effectively easing training. In order to generalise the results, we also present a methodology evaluating this physics informed approach. We present a study-case of a PV system in Denmark to validate our claims by extensively evaluating five different ML methods: Random Forest, Support Vector Machine, Convolutional Neural Networks (CNN), Long-Short Term Memory (LSTM) and a hybrid CNN–LSTM. The results show consistently how the best predictors use the proposed physics-informed features disregarding the particular ML-method, and forecasting horizon. However, also, how there is a threshold regarding the number of previous samples to be included that appears as a convex function.

Hourly Forecasting of Solar Photovoltaic Power in Pakistan Using Recurrent Neural Networks

The solar photovoltaic (PV) power forecast is crucial for steady grid operation, scheduling, and grid electricity management. In this work, numerous time series forecast methodologies, including the statistical and artificial intelligence-based methods, are studied and compared fastidiously to forecast PV electricity. Moreover, the impact of different environmental conditions for all of the algorithms is investigated. Hourly solar PV power forecasting is done to confirm the effectiveness of various models. Data used in this paper is of one entire year and is acquired from a 100 MW solar power plant, namely, Quaid-e-Azam Solar Park, Bahawalpur, Pakistan. This paper suggests recurrent neural networks (RNNs) as the best-performing forecasting model for PV power output. Furthermore, the bidirectional long-short-term memory RNN framework delivered high accuracy results in all weather conditions, especially under cloudy weather conditions where root mean square error (RMSE) was found lowest 0.0025, R square stands at 0.99, and coefficient of variation of root mean square error (RMSE) Cv was observed 0.0095%.

Critical Review of Data, Models and Performance Metrics for Wind and Solar Power Forecast

Global climatic changes and increased carbon footprints provided the main impetus for the decrease in the use of fossil fuels for electricity generation and transportation. Matured manufacturing technologies of solar PV panels and on-shore and off-shore windmills have brought down the cost of generation of electricity using solar energy on par with conventional fossil fuel. Initially, solar and wind power generation was envisioned for microgrids, serving small local communities. However, advancements in power electronics have now facilitated large solar and wind farms to be integrated with main power grids. In this context, hosting capacity, which is the amount of distributed energy resources a grid can accommodate, without significant infrastructure up-gradation, has gained importance. In determining the hosting capacity at a particular location, the uncertainties of wind and solar power generation play a role. Effective forecasting models using time-series weather data can be built to predict wind and solar power generation. This forecast is essential to ensure proper grid operation and control when renewable energy sources are already installed. The forecast is also useful in the planning stages for investment decisions and distribution system planning. While long-term forecasts are rarely needed for the operation of integrated grids, accurate short-term predictive models are necessary for scheduling. This paper presents an extensive review of various forecast models available in the literature. The study mainly focuses on the short-term forecast, providing a critical review of the duration of data used in each model and a synoptic comparison of their performance indices.

Overcoming the Chicken-and-Egg Problem in Prosumer Solar Power and Energy Storage Forecasting: The Case of Thailand

Overcoming the Chicken-and-Egg Problem in Prosumer Solar Power and Energy Storage Forecasting: The Case of Thailand

SYSTEM OF FORECASTING THE SMART HOME SOLAR BATTERY POWER

Controlling household appliances in automatic mode according to the specified logic and the needs of the residents of a smart home allows efficient use of electricity and saves time and money. The paper proposes a system of forecasting the smart home solar battery power based on the analysis of weather data. The cloudiness degree values obtained using the OpenWeatherMap online service for a given area and time are used. Forecasting is carried out by methods of correlation analysis. The Pearson coefficient and parameters of the regression line were determined using the experimentally measured power of the Buheshui JZ-110X69 photovoltaic module under different weather conditions. The close to one value of Pearson coefficient modulus indicates a close relationship and high correlation between cloudiness and solar panel power. Correction coefficients were introduced to eliminate the influence of the angle between sun rays and the normal to the panel surface on the photovoltaic module power. The correction coefficients were determined on a cloudless day on March 20, 2022, in Lviv for each hour as the ratio of the solar module power at the measurement time to the measured at 1 pm power. It was established that adjusting the photovoltaic module power value according to the time of day increases the degree of linear correlation and the angle of inclination of the regression line. The weather forecast and obtained linear regression equation were used to estimate the future power of the solar battery. The use of correction coefficients according to the time of day increases the accuracy of the smart home solar battery power forecast. The obtained results can be used to form recommended scenarios for the rational use of energy resources. Choosing the optimal scenarios for using household devices in accordance with the forecasted solar battery power makes it possible to balance the comfort necessary level and frugal energy management of a smart home. Key words : smart home, energy management, energy efficiency, correlation, regression line, forecasting.

How Can Probabilistic Solar Power Forecasts Be Used to Lower Costs and Improve Reliability in Power Spot Markets? A Review and Application to Flexiramp Requirements

Net load uncertainty in electricity spot markets is rapidly growing. There are five general approaches by which system operators and market participants can use probabilistic forecasts of wind, solar, and load to help manage this uncertainty. These include operator situation awareness, resource risk hedging, reserves procurement, definition of contingencies, and explicit stochastic optimization. We review these approaches, and then provide a case study in which a method for using probabilistic solar forecasts to define needs for reserves is developed and evaluated. The case study has three parts. First, we describe building blocks for enhancing the Watt-Sun solar forecasting system to produce probabilistic irradiance and power forecasts. Second, relationships between Watt-Sun forecasts for multiple sites in California and the system’s need for flexible ramp capability (flexiramp) are defined by machine learning and statistical methods. Third, the performance of present methods to defining flexiramp requirements, which are not conditioned on weather and renewables forecasts, is compared with that of probabilistic solar forecast-based requirements, using a multi-timescale production costing model with an 1820-bus representation of the WECC power system. Significant potential savings in fuel and flexiramp procurement costs from using solar-informed reserve requirements are found.

Using probabilistic solar power forecasts to inform flexible ramp product procurement for the California ISO

How can independent system operators (ISOs) take advantage of probabilistic solar forecasts to lower generation costs and improve reliability of power systems? We discuss one three-step approach for doing so, focusing on how such forecasts might help the California Independent System Operator (CAISO) prepare unexpected net load ramps, where net load equals gross demand minus wind and solar production. First, we enhance an existing solar forecasting system to provide well-calibrated hours-ahead probabilistic forecasts. We then relate the degree of uncertainty reflected in the forecasted prediction intervals (independent variables) to error distributions for net load ramp forecasts for the CAISO real-time market (dependent variable) using machine learning and quantile regression. Projected ramp forecast errors conditioned on solar uncertainty are translated into flexible ramp requirements that therefore reflect real-time meteorological and solar conditions, improving on typical ISO procedures. Detailed descriptions are provided on the quantile regression and kth-nearest neighbor categorization methods for accomplishing that translation. Finally, a multiple time-scale look-ahead market simulation model is applied to a 118-bus IEEE Reliability Test System, modified to represent the CAISO generation mix and demand distributions. The model runs quantify how solar-conditioned ramp requirements can, first, decrease operating costs by reducing requirements compared to often conservative unconditional methods and, second, decrease generation scarcity events and consequently improve reliability by increasing flexibility requirements at times when unconditional forecast-based requirements understate actual ramp uncertainty. Solar-conditioned ramp requirements are found to reduce generation operating costs by about 2% for the test system (which would be equivalent to over $100 million per year for a CAISO-size system).

A robust auto encoder-gated recurrent unit (AE-GRU) based deep learning approach for short term solar power forecasting

The increasing presence of solar power plants shows its potency as one of the key renewable energy resource to fulfill energy needs of the community. This increasing presence can be effectively utilized by using efficient prediction of solar power plant output for stable grid operation. The availability of processors with high computational capability and access to baseline data provide edge to deep learning models for prediction of solar power plants. In this work, sequence to sequence auto-encoder (AE) and Gated Recurrent unit (GRU) based hybrid deep learning approach is developed, which further advances other recent works offered in solar power forecasting. The AE extracts the internal relationship of input solar data, minimizes the reconstruction error and taps the important feature correlation. Furthermore, GRU exploits the time dependencies of the data. For comprehensive evaluation, different deep learning models and their hybrid forms are analyzed for different prediction durations i.e. 24 h, 48 h, and 15 days prediction. The analysis compares models on three major performance indices, i.e., mean square error, mean absolute error, and R2errors with the same hyperparameters for this time series prediction. The outcome ascertains that the proposed AE-GRU based deep learning model performs better than other deep learning models.

Short-Term Solar Power Forecasting and Uncertainty Analysis Using Long and Short-Term Memory

As a green and renewable resource, the scale of solar energy is increasing. Huge amount of solar energy penetration in a distribution system causes drastic changes in the load forming a duck shape load curve which may cause stability problems. Accurate load forecasting and demand response procedures are the key tasks of power distribution system management. For this reason, this paper proposes a new mechanism for optimized operation of solar microgrid based on deep learning long and short-term memory (LSTM) and demand response (DR). LSTM is used for load prediction, and linear integer programming is used to implement load scheduling. The proposed DR procedure is used for load management, the proposed method is implemented in a modified IEEE-12 bus radial distribution network. In a dynamic pricing environment, not only the stability of the voltage is guaranteed, but the electricity cost is minimized. This paper solves the dual objectives that is flattening the duck curve and electricity cost minimization.

Deep learning-based probabilistic anomaly detection for solar forecasting under cyberattacks

Recently, cyberattacks on solar power forecasting emerge with the increase of solar penetration, which may lead to substantial economic losses and power system reliability issues. This paper presents a novel probabilistic anomaly detection framework to effectively and accurately detect cyberattacks in solar power forecasting, which consists of three major components. First, a convolutional neural network is used to extract spatial correlations among solar farms at the bottom of the model. Second, a long short-term memory network is used to capture the temporal dependencies within solar power data and generate deterministic solar power forecasts, which is then converted to probabilistic forecasts through pinball loss optimization. Finally, the probabilistic solar power forecasts are used for anomaly detection. The effectiveness of the proposed framework is validated by using 16 solar farms with a variety of data integrity attacks. Numerical results of case studies show that the developed probabilistic anomaly detection methodology could effectively detect data integrity attacks in solar power forecasting with a relatively high accuracy.

Operational day-ahead solar power forecasting for aggregated PV systems with a varying spatial distribution

Accurate forecasts of the power production of distributed photovoltaic (PV) systems are essential to support grid operation and enable a high PV penetration rate in the electricity grid. In this study, we analyse the performance of 12 different models that forecast the day-ahead power production in agreement with market conditions. These models include regression, support vector regression, ensemble learning, deep learning and physical based techniques. In addition, we examine the effect of aggregating multiple PV systems with a varying inter-system distance on the forecast model performance. The models are evaluated both on their technical and economic performance. From a technical perspective, the results show a positive effect from both an increasing inter-system distance and a larger sized PV fleet on the model performance, which was not the case for the economic assessment. Furthermore, the ensemble and deep learning models perform better than the alternatives from a technical point of view. For the economic assessment, the results indicate the superiority of the physical based model, followed by the deep learning models. Lastly, our findings show the importance of considering the user's objective when assessing solar power forecast models.

Characteristic profile

The rapid uptake of rooftop solar photovoltaic systems is introducing many challenges in the management of distribution networks, energy markets, and energy storage systems. Many of these problems can be alleviated with accurate short term solar power forecasts. However, forecasting the power output of distributed rooftop solar PV systems can be challenging, since many complex local factors can affect solar output. A common approach when forecasting such systems is to extract the daily seasonality from the time series using some form of seasonality model, and then forecast only the residuals that remain after seasonality extraction. In this work, we explore in detail the effectiveness of three commonly used seasonality models, and we propose a new one, called the "characteristic profile". We find that when seasonality models are integrated into the forecasting process, significant gains in forecast accuracy may be obtained - particularly for machine learning based approaches, which have a reduction in forecast error of 5-25%. Among the seasonality models, the characteristic profile demonstrates the highest forecast accuracy, resulting in reductions in forecast error of 8% or more compared to forecasting models that do not take seasonality into account. The benefits of this approach are particularly pronounced when forecasting solar PV systems that are curtailed, suffer from local shading, or consist of multiple sets of panels having different orientations and tilts. Our results are demonstrated on a high resolution dataset obtained from 258 sites in Western Australia over the course of a full year.

Demand response of grid-connected microgrid based on metaheuristic optimization algorithm

ABSTRACT Demand Response Programs (DRPs) have gained the microgrid (MG) operators phenomenal attention for mobilizing the end-users in alleviating the uncertainties associated with renewable energy sources. In this research work, the effective MG energy management system (EMS) in conjunction with price-driven DRPs is proposed to achieve synergistic coordination between the energy providers and consumers and reduce operational costs. The flexible price elasticity model is implemented instead of using the price elasticity models with a preordained constant value like most existing literature. This results in a more realistic characterization of customer responsiveness to energy price changes and promotes the DRPs under a grid-connected MG environment. With this regard, a stochastic day-ahead energy management strategy is proposed to incorporate four distinct DRPs and schedule the MG distributed energy resources. The proposed strategy is verified on a practical 3-feeder MG test system with a majority of 50% industrial load considered. The short-term scheduling period of 15 min ahead solar and wind power forecast is considered to optimize the microgrid dispatch costs accurately. The intermittent nature of renewable energy sources is addressed by employing a stochastic-based scenario generation and reduction approach. A novel and nature-inspired Black Widow Optimization (BWO) is applied to determine the optimal scheduling configuration. The effectiveness of the BWO is validated in terms of rate of convergence, computational time, and solution efficacy. Finally, the best DRP has been chosen based on its technical and economic performance indices by employing a multi-criteria decision-making approach.

Optimization of electric spring operational strategy to minimize electricity bill

This paper aims at using electric spring to reduce the monthly electricity bill by developing its optimal operational strategy to capitalize on the variable energy cost over a day. An hourly-sampled optimization algorithm is presented to devise an optimum daily strategy for electric spring as a tool to smartly manage energy among the loads, utility, and storage devices, provided that load and solar power forecasts are available. Uncertainty in forecasts is dealt with by embedding certain stochastic processes with the proposed optimization algorithm to identify an envelope of probable operational strategies. The development and deployment of the proposed optimization algorithm are presented on a realistic load profile of a typical net-zero energy house in South Australia. The obtained results are analyzed under different connection configurations. Consequently, a significant reduction in the monthly electricity bill is realized.

Short Term Solar Power and Temperature Forecast Using Recurrent Neural Networks

Short Term Solar Power and Temperature Forecast Using Recurrent Neural Networks

Impact of Soiling on Energy Yield of Solar PV Power Plant and Developing Soiling Correction Factor for Solar PV Power Forecasting

Across the world, the geographical conditions are varied, and the characteristics of dust depend on the local environmental conditions. The solar power generators must incorporate the soiling losses in their estimation for power output and therefore a methodology was developed to estimate the soiling correction factor. After extensive research, a comprehensive review was presented on the effect of soiling on performance of PV plants along with case studies of soiling experiments around the world. A soiling experiment was designed to develop the soiling correction factor. A methodology to calculate the soiling correction factor, which can be implemented in any location, was developed by analyzing the data from the soiling experiment. The effect of rainfall, humidity and wind on soiling was analyzed and documented. The performance of one 20 kWp PV plant was monitored to study the effect of weather-related parameters on the performance. The soiling correction factor varied from -1.36% to 3.67% during the period between June 2018 and June 2019 in Chennai. It was observed that the average PV conversion efficiency of the 20-kW plant was 11.75% and the average PR was 75%. It was observed that the correlation between module temperature and DC power; between humidity and DC power; between GTI and DC power varied every month. The soiling factor developed could be incorporated into the short-term day ahead solar forecasting model. The developed methodology could be applied at the any large-scale solar power plant around the world for yield assessment, designing as well as operational forecasting purposes.

Regional Solar Irradiance Forecast for Kanto Region by Support Vector Regression Using Forecast of Meso-Ensemble Prediction System

From the perspective of stable operation of the power transmission system, the transmission system operators (TSO) needs to procure reserve adjustment power at the stage of the previous day based on solar power forecast information from global horizontal irradiance (GHI). Because the reserve adjustment power is determined based on information on major outliers in past forecasts, reducing the maximum forecast error in addition to improving the average forecast accuracy is extremely important from the perspective of grid operation. In the past, researchers have proposed various methods combining the numerical weather prediction (NWP) and machine learning techniques for the one day-ahead solar power forecasting, but the accuracy of NWP has been a bottleneck issue. In recent years, the development of the ensemble prediction system (EPS) forecasts based on probabilistic approaches has been promoted to improve the accuracy of NWP, and in Japan, EPS forecasts in the mesoscale domain, called mesoscale ensemble prediction system (MEPS), have been distributed by the Japan Meteorological Agency (JMA). The use of EPS as a machine learning model is expected to improve the maximum forecast error, as well as the accuracy, since the predictor can utilize various weather scenarios as information. The purpose of this study is to examine the effect of EPS on the GHI prediction and the structure of the machine learning model that can effectively use EPS. In this study, we constructed the support vector regression (SVR)-based predictors with multiple network configurations using MEPS as input and evaluated the forecast error of the Kanto region GHI by each model. Through the comparison of the prediction results, it was shown that the machine learning model can achieve average accuracy improvement while reducing the maximum prediction error by MEPS, and knowledge was obtained on how to effectively provide EPS information to the predictor. In addition, machine learning was found to be useful in improving the systematic error of MEPS.