Data-Driven Load Forecasting Using Machine Learning and Meteorological Data

Electrical load forecasting is very crucial for electrical power systems’planning and operation.Both electrical buildings’load demand and meteorological datasets may contain hidden patterns that are required to be investigated and studied to show their potential impact on load forecasting.The meteorological data are analyzed in this study through different data mining techniques aiming to predict the electrical load demand of a factory located in Riyadh,Saudi Arabia.The factory load and meteorological data used in this study are recorded hourly between 2016 and 2017.These data are provided by King Abdullah City for Atomic and Renewable Energy and Saudi Electricity Company at a site located in Riyadh.After applying the data pre-processing techniques to prepare the data,different machine learning algorithms,namely Artificial Neural Network and Support Vector Regression(SVR),are applied and compared to predict the factory load.In addition,for the sake of selecting the optimal set of features,13 different combinations of features are investigated in this study.The outcomes of this study emphasize selecting the optimal set of features as more features may add complexity to the learning process.Finally,the SVR algorithm with six features provides the most accurate prediction values to predict the factory load.

Spatial-temporal characteristics of drought detected from meteorological data with high resolution in Shaanxi Province, China

The spatial pattern of meteorological factors cannot be accurately simulated by using observations from meteorological stations(OMS) that are distributed sparsely in complex terrain. It is expected that the spatial-temporal characteristics of drought in regions with complex terrain can be better represented by meteorological data with the high spatial-temporal resolution and accuracy. In this study, Standard Precipitation Evapotranspiration Index(SPEI) calculated with meteorological factors extracted from ITPCAS(China Meteorological Forcing Dataset produced by the Institute of Tibetan Plateau Research, Chinese Academy of Sciences) was applied to identify the spatial-temporal characteristics of drought in Shaanxi Province of China, during the period of 1979–2016. Drought areas detected by SPEI calculated with data from ITPCAS(SPEI-ITPCAS) on the seasonal scale were validated by historical drought records from the Chinese Meteorological Disaster Canon-Shaanxi, and compared with drought areas detected by SPEI calculated with data from OMS(SPEI-OMS). Drought intensity, trend and temporal ranges for mutations of SPEI-ITPCAS were analyzed by using the cumulative drought intensity(CDI) index and the Mann-Kendall test. The results indicated that drought areas detected from SPEI-ITPCAS were closer to the historical drought records than those detected from SPEI-OMS. Severe and exceptional drought events with SPEI-ITPCAS lower than –1.0 occurred most frequently in summer, followed by spring. There was a general drying trend in spring and summer in Shaanxi Province and a significant wetting trend in autumn and winter in northern Shaanxi Province. On seasonal and annual scales, the regional and temporal ranges for mutations of SPEI-ITPCAS were different and most mutations occurred before the year 1990 in most regions of Shaanxi Province. The results reflect the response of different regions of Shaanxi Province to climate change, which will help to manage regional water resources.

Towards Increasing Data Availability for Meteorological Services: Inter-Comparison of Meteorological Data from a Synoptic Weather Station and Two Automatic Weather Stations in Kenya

Meteorological data is useful for varied applications and sectors ranging from weather and climate forecasting, landscape planning to disaster management among others. However, the availability of these data requires a good network of manual meteorological stations and other support systems for its collection, recording, processing, archiving, communication and dissemination. In sub-Saharan Africa, such networks are limited due to low investment and capacity. To bridge this gap, the National Meteorological Services in Kenya and few others from African countries have moved to install a number of Automatic Weather Stations (AWSs) in the past decade including a few additions from private institutions and individuals. Although these AWSs have the potential to improve the existing observation network and the early warning systems in the region, the quality and capacity of the data collected from the stations are not well exploited. This is mainly due to low confidence, by data users, in electronically observed data. In this study, we set out to confirm that electronically observed data is of comparable quality to a human observer recorded data, and can thus be used to bridge data gaps at temporal and spatial scales. To assess this potential, we applied the simple Pearson correlation method and other statistical tests and approaches by conducting inter-comparison analysis of weather observations from the manual synoptic station and data from two Automatic Weather Stations (TAHMO and 3D-PAWS) co-located at KMD Headquarters to establish existing consistencies and variances in several weather parameters. Results show there is comparable consistency in most of the weather parameters between the three stations. Strong associations were noted between the TAHMO and manual station data for minimum (r = 0.65) and maximum temperatures (r = 0.86) and the maximum temperature between TAHMO and 3DPAWS (r = 0.56). Similar associations were indicated for surface pressure (r = 0.99) and RH (r > 0.6) with the weakest correlations occurring in wind direction and speed. The Shapiro test for normality assumption indicated that the distribution of several parameters compared between the 3 stations were normally distributed (p > 0.05). We conclude that these findings can be used as a basis for wider use of data sets from Automatic Weather Stations in Kenya and elsewhere. This can inform various applications in weather and climate related decisions.

Wireless Backup System of Meteorological Data Transmission

By using wireless communication technology, when wired network of grassroots station is failure, wireless backup system of meteorological data transmission can automatically switch, and transmit the data at the station with line fault to the destination by short message way. The system has simple operation and low operation cost, and transmission fault of real-time meteorological data due to line problem at the station can be effectively decreased. At present, the system has been tested successfully and put into business operation.

Development of Management and Application System of Basic Meteorological Data

Based on convenience and safety of historical data application, B/S mode is used instead of database management structure of C/S mode, and it can not only combine database and network, but also realize the safe use of historical data online. High-level programming language is used to develop a online management and application system of historical meteorological data based on B/S mode. System data import function can import ground report file sequence of Xuzhou （ including five county-level stations） into the database, construct Oracle database of Xuzhou and the five stations since 1953, and establish data tables of time, day, ten-day, monthly, quarterly and annual historical data, weather information and so forth. Management software of database server is established to realize instruction-level management and scheduling of database and a balanced distribution of resources among users. At the same time, a Web-based management application interface is set up to meet users＇ needs to retrieve a variety of repositories, and it provides statistical query of time, day, ten-day, monthly, quarterly and annual historical data and climate data for each meteorological element, thereby meeting the needs of meteorological research and all sectors of society for statistical query of meteorological data.

Development of Multilevel Meteorological Data Acquisition System for Radiation Emergency Decision Support

This paper describes a development of multilevel meteorological data acquisition system implemented at Kalpakkam coastal site for atmospheric dispersion and model validation studies. Meteorological data are one of the most important inputs into any air dispersion model. As a part of atmospheric dispersion modeling studies and developing a methodology to forecast the site-specific dispersion characteristics, the real time monitoring of meteorological parameters assumes significance. This is achieved by erecting met towers instrumented at multilevel and single level at different locations with sensors for measuring various meteorological parameters. Real-world data logging applications involve not only just acquiring and recording signals, but also combination of offline analysis, display, report generation and data sharing. This paper covers development of low cost compact MMDAS （modular meteorological data acquisition system）, its performance evaluation, field deployment test and data comparison analysis with fast response and high accuracy internationally acclaimed sonic anemometer. The system is based on embedded modules from Advantech and is designed to acquire analogue and digital signals from a multilevel instrumented met tower. The collected data are transferred from remote base station to central server for storage and further processing using wireless interface. MMDAS has many advantages like cost effectiveness, less complex signal conditioning electronics and easy maintenance. This system has good application during radiation emergency as well as site specific meteorological data collection and model validation studies.

Research on Spatial Statistical Downscaling Method of Meteorological Data Applied to Photovoltaic Prediction

Aiming at the low spatial resolution of meteorological data output from a numerical model in photovoltaic power prediction,a geographically weighted statistical downscaling method considers the influence factors such as normalized vegetation index(NDVI),digital elevation model(DEM),slope direction,longitude and latitude is proposed.This method is based on the correlation between meteorological data and NDVI,DEM,slope direction,latitude and longitude,and introduces DEM and local Moran index to improve the regression model,and obtains 100∗100 m high-resolution meteorological spatial distribution data.Finally,combining the measured data of the study area and the established EOF iterative downscaling method to verify and compare the downscaling results.The results show that the error between the downscaled meteorological data and the measured value is smaller,and the comprehensive downscaling accuracy of the geographically weighted regression method is higher,and the model fitting effect is better.Therefore,this method can effectively improve the influence of errors caused by lower resolution,and provide a more reliable meteorological basis for the prediction of photovoltaic power.

Meteorological Data Generation for the Numerical Simulation of Stratified Flow in the Joumine Reservoir, Tunisia

Dynamic numerical simulation of water conditions is useful for reservoir management. In remote semi-arid areas, however, meteorological and hydrological time-series data needed for computation are not frequently measured and must be obtained using other information. This paper presents a case study of data generation for the computation of thermal conditions in the Joumine Reservoir, Tunisia. Data from the Wind Finder web site and daily sunshine duration at the nearest weather stations were utilized to generate cloud cover and solar radiation data based on meteorological correlations obtained in Japan, which is located at the same latitude as Tunisia. A time series of inflow water temperature was estimated from air temperature using a numerical filter expressed as a linear second-order differential equation. A numerical simulation using a vertical 2-D （two-dimensional） turbulent flow model for a stratified water body with generated data successfully reproduced seasonal thermal conditions in the reservoir, which were monitored using a thermistor chain.

Application of Meteorological Data Operation System(MDOS)in Meteorological Operation

Meteorological data operation system(MDOS)is a comprehensive business platform that integrates data transmission and monitoring,quality control,processing,storage and distribution,and data services.The operation and implementation of MDOS has further improved the quality control system of meteorological data in Yunnan Province,adjusted the business layout of meteorological data,and made the availability ratio of meteorological data continue to increase.The availability ratio of meteorological data from national stations rose from 99.68% to 99.96%,and that of regional stations increased from 95.33% to 99.08% from 2016 to 2020.MDOS not only solves the problem of inconsistent data and quality control methods of various business platforms,but also provides guarantees for more refined and accurate weather forecasts and climate predictions.

Determination of Kolmogorov Entropy of Chaotic Attractor Included in One-Dimensional Time Series of Meteorological Data

The 1970-1985 day to day averaged pressure dataset of Shanghai and the extension method in phase space are used to calculate the correlation dimension D and the second-order Renyi entropy K2 of the approximation of Kolmogorov's entropy, the fractional dimension D = 7.7-7.9 and the positive value K2 - 0.1 are obtained. This shows that the attractor for the short-term weather evolution in the monsoon region of China exhibits a chaotic motion. The estimate of K2 yields a predictable time scale of about ten days. This result is in agreement with that obtained earlier by the dynamic-statistical approach.The effects of the lag time i on the estimate of D and K2 are investigated. The results show that D and K2 are convergent with respect to i. The day to day averaged pressure series used in this paper are treated for the extensive phase space with T = 5, the coordinate components are independent of each other; therefore, the dynamical character quantities of the system are stable and reliable.

Geographically Weighted Regression and Secondary Variables for Mapping of Meteorological Data

GA （geostatistical analyst） is an indispensable tool to analyze various and plenty of data in GIS （geographic information system）. Spatial distribution is the most effective factor for predicting of meteorological maps at the point of performance or reliability of the model. Generally, classical interpolation methods may not be sufficient to produce accurate maps. GA is more considerable in this state. Secondary variables affect the precious of prediction models especially meteorological data mapping. In this study 245 meteorological data stations have been evaluated to produce precipitation model maps in Turkey. Long term （25 years） mean annual and monthly precipitation data from Turkish State Meteorological Service and elevation, slope and aspect values from DEM （Digital Elevation Model） were registered. OK （Ordinary Kriging）, OCK （Ordinary Co-Kriging） and GWR （Geographically Weighted Regression） have been used as a method to compare the models. With the study if there are effects of secondary variables to precipitation models have been illustrated on the prediction maps. Besides comparing statistical values, regional effects of secondary variables have been determined and illustrated on the maps numerically. As a result to define precipitation distribution spatially R2 values between measured and predicted values have been calculated 0.55 for Kriging, 0.67 for OCK and 0.86 for GWR. Cross validation indicated that GWR interpolation yields the smallest prediction error with elevation, slope and aspect. Spatial distribution of meteorological stations is also other important factor for similar studies.

Effect of Meteorological Data Assimilation on Regional Air Quality Forecasts over the Korean Peninsula

The Weather Research and Forecasting model coupled with Chemistry(WRF-Chem),a type of online coupled chemistry-meteorology model(CCMM),considers the interaction between air quality and meteorology to improve air quality forecasting.Meteorological data assimilation(DA)can be used to reduce uncertainty in meteorological field,which is one factor causing prediction uncertainty in the CCMM.In this study,WRF-Chem and three-dimensional variational DA were used to examine the impact of meteorological DA on air quality and meteorological forecasts over the Korean Peninsula.The nesting model domains were configured over East Asia(outer domain)and the Korean Peninsula(inner domain).Three experiments were conducted by using different DA domains to determine the optimal model domain for the meteorological DA.When the meteorological DA was performed in the outer domain or both the outer and inner domains,the root-mean-square error(RMSE),bias of the predicted particulate matter(PM)concentrations,and the RMSE of predicted meteorological variables against the observations were smaller than those in the experiment where the meteorological DA was performed only in the inner domain.This indicates that the improvement of the synoptic meteorological fields by DA in the outer domain enhanced the meteorological initial and boundary conditions for the inner domain,subsequently improving air quality and meteorological predictions.Compared to the experiment without meteorological DA,the RMSE and bias of the meteorological and PM variables were smaller in the experiments with DA.The effect of meteorological DA on the improvement of PM predictions lasted for approximately 58-66 h,depending on the case.Therefore,the uncertainty reduction in the meteorological initial condition by the meteorological DA contributed to a reduction of the forecast errors of both meteorology and air quality.

Assessment of Meteorological Threats to the Coordinated Search and Rescue of Unmanned/Manned Aircraft

The architecture and working principle of coordinated search and rescue system of unmanned/manned aircraft,which is composed of manned/unmanned aircraft and manned aircraft,were first introduced,and they can cooperate with each other to complete a search and rescue task.Secondly,a threat assessment method based on meteorological data was proposed,and potential meteorological threats,such as storms and rainfall,can be predicted by collecting and analyzing meteorological data.Finally,an experiment was carried out to evaluate the performance of the proposed method in different scenarios.The experimental results show that the coordinated search and rescue system of unmanned/manned aircraft can be used to effectively assess meteorological threats and provide accurate search and rescue guidance.

PNN based crop disease recognition with leaf image features and meteorological data

An automatic crop disease recognition method was proposed in this paper,which combined the statistical features of leaf images and meteorological data.The images of infected crop leaves were taken under different environments of the growth periods,temperature and humidity.The methods of image morphological operation,contour extraction and region growing algorithm were adopted for leaf image enhancement and spot image segmentation.From each image of infected crop leaf,the statistical features of color,texture and shape were extracted by image processing,and the optimal meteorological features with the highest accuracy rate were obtained and selected by the attribute reduction algorithm.The fusion feature vector of the image was formed by combining the statistical features and the meteorological features.Then the probabilistic neural networks(PNNs)classifier was adopted to evaluate the classification accuracy.The experimental results on three cucumber diseased leaf image datasets,i.e.,downy mildew,blight and anthracnose,showed that the crop diseases can be effectively recognized by the integrated application of leaf image processing technology,the disease meteorological data and PNNs classifier,and the recognition accuracy rate was higher than 90%,which indicated that the PNNs classifier trained on the disease feature coefficients extracted from the crop disease leaves and meteorological data could achieve higher classification accuracy.

Quality control and homogenization of daily meteorological data in the trans-boundary region of the Jhelum River basin

Many studies such as climate variability, climate change, trend analysis, hydrological designs, agriculture decision-making etc. require long-term homogeneous datasets. Since homogeneous climate data is not available for climate analysis in Pakistan and India, the present study emphases on an extensive quality control and homogenization of daily maximum temperature, minimum temperature and precipitation data in the Jhelum River basin, Pakistan and India. A combination of different quality control methods and relative homogeneity tests were applied to achieve the objective of the study. To check the improvement after homogenization, correlation coefficients between the test and reference series calculated before and after the homogenization process were compared with each other. It was found that about 0.59%, 0.78% and 0.023% of the total data values are detected as outliers in maximum temperature, minimum temperature and precipitation data, respectively. About 32% of maximum temperature, 50% of minimum temperature and 7% of precipitation time series were inhomogeneous, in the Jhelum River basin. After the quality control and homogenization, 1% to 11% improvement was observed in the infected climate variables. This study concludes that precipitation daily time series are fairly homogeneous, except two stations （Naran and Gulmarg）, and of a good quality. However, maximum and minimum temperature datasets require an extensive quality control and homogeneity check before using them into climate analysis in the Jhelum River basin.

SOME PROBLEMS AND TREATMENT IN THE REGRESSION ANALYSIS OF METEOROLOGICAL DATA

In this paper,some problems of regression analysis in the meteorological application are discussed and main reasons for statistical inference failures are analysed.We may find the failure problems with diagnos- tic method and solve them by different treatment.It has been proved that the treatment make the accuracy and stability of forecasting improved greatly.

Estimation of Sand Transportation Rate for Fixed and Semi-Fixed Dunes Using Meteorological Wind Data

Taibus County, Inner Mongolia, China, lies in a farming-pastoral ecotone, where severe wind erosion and various aeolian sand hazards are prevalent and fixed and semi-fixed sand dunes occur frequently. This study was conducted to investigate the relationships between sand transportation rate and wind speed for the fixed and semi-fixed sand dunes based on field measurements. The annual quantity of soil erosion by wind was estimated using meteorological wind data. The results indicated that the sand transportation rate in Taibus County in 2000 was 57.38 kg cm-1 year-1 for the semi-fixed dunes and 4.56 kg cm-1 year-1 for the fixed dunes. The total duration of erosive winds covered 12.5% of the time of the year, and spring posed the highest potential of sand transportation. Wind with low speed (≤ 17 m s-1) and high frequency plays a dominant role in sand transportation, while strong wind (≥ 17 m s-1) with low frequency significantly enhanced the sand transportation. Erosive wind speed, directions, and frequency were three crucial dynamic factors influencing sand hazards in the farming-pastoral ecotone. The dominant factors intensifying sand and dust storms in Taibus County might be related to the favorable wind condition in combination with the durable drought, which led to land desertification and vegetation degradation.

Identification of Dust Sources in Iraq Using Meteorological Surface Data and Satellite Data

The research included employment of the meteorological surface data such as surface pressure and wind speed and direction to determine the sources of dust in Iraq through the study and analysis of dust phenomena that strike Iraq in 2011 and at different times of the year. The results showed that Iraq was suffering to No. （18） dust storms in 2011. The highest wind speed recorded at the beginning of the composition of most the dust storms ranged of 6-10 m/s and the lowest wind speed values ranged of 10-14 m/s, while the direction of the wind in No. （14） dust storms were flowed from NW （northwest）, only No. （4） dust storms were flowed from southwest SW （southwest）, due to the fact that the prevailing wind in Iraq is NW in most days of the year. Atmospheric pressure values recorded for the day of dust storms are ranged of 960-990 millibar, due to the existence Iraq under high pressure system most days of the year. Satellite data of Meteosat 9 are used to determine the sources of dust over Iraq. Used the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations CALIPSO data to determine the vertical section of the dust, during a dust storm.

Evaluation of Water Losses by Evaporation in the Nakanbe Basin

A numerical approach to heat and mass transfer in a large water reservoir is presented. This water reservoir is likened to a parallelepiped reservoir whose vertical and lower walls are adiabatic and impermeable. The equations that govern natural convection in water are solved by the finite volume method and Thomas’salgorithm. The adequacy between the velocity and pressure fields is ensured by the SIMPLE algorithm. We are going to evaluate the water losses by evaporation from three dams in the Nakanbé basin in Burkina Faso for a period of thirty years, that is to say from January 1, 1991, to March 15, 2020. The three dams have a rate of evaporation greater than 40% of the volume of water stored. Indeed the rate of evaporation in each dam increases with the water filling rate in the reservoir: we have observed the following results for each dam in the Nakanbé basin;for the date of 02/27/1988 to 03/13/2020., the Loumbila dam received a total volume of stored water of 22.02 Mm3 and 10.57 Mm3 as the total volume of water evaporated at the same date. At the Ouaga dam (2 + 3), it stored a water volume of 4.06 Mm3 and evaporated 2.03 Mm3 of its storage volume from 01/01/1988 to 05/07/2016. Finally, with regard to the Bagré dam, it stored 745.16 Mm3 of water and 365.13 Mm3 as the volume of water evaporated from 01/01/1993 to 03/31/2020.

Numerical Modeling and Technico-Economic Analysis of a Hybrid Energy Production System for Self-Consumption: Case of Rural Area in the Comoros

This study aims to provide electricity to a remote village in the Union of Comoros that has been affected by energy problems for over 40 years. The study uses a 50 kW diesel generator, a 10 kW wind turbine, 1500 kW photovoltaic solar panels, a converter, and storage batteries as the proposed sources. The main objective of this study is to conduct a detailed analysis and optimization of a hybrid diesel and renewable energy system to meet the electricity demand of a remote area village of 800 to 1500 inhabitants located in the north of Ngazidja Island in Comoros. The study uses the Hybrid Optimization Model for Electric Renewable (HOMER) Pro to conduct simulations and optimize the analysis using meteorological data from Comoros. The results show that hybrid combination is more profitable in terms of margin on economic cost with a less expensive investment. With a diesel cost of $1/L, an average wind speed of 5.09 m/s and a solar irradiation value of 6.14 kWh/m2/day, the system works well with a proportion of renewable energy production of 99.44% with an emission quantity of 1311.407 kg/year. 99.2% of the production comes from renewable sources with an estimated energy surplus of 2,125,344 kWh/year with the cost of electricity (COE) estimated at $0.18/kWh, presenting a cost-effective alternative compared to current market rates. These results present better optimization of the used hybrid energy system, satisfying energy demand and reducing the environmental impact.