Day-ahead forecasting of solar radiation is essential for grid balancing, real-time unit dispatching, scheduling and trading in the solar energy utilization system. In order to provide reliable forecasts of solar radiation, a novel hybrid model is proposed in this study. The hybrid model consists of two modules: a mesoscale numerical weather prediction model (WRF: Weather Research and Forecasting) and Kalman filter. However, the Kalman filter is less likely to predict sudden changes in the forecasting errors. To address this shortcoming, we develop a new framework to implement a Kalman filter based on the clearness index. The performance of this hybrid model is evaluated using a one-year dataset of solar radiation taken from a photovoltaic plant located at Maizuru, Japan and Qinghai, China, respectively. The numerical results reveal that the proposed hybrid model performs much better in comparison with the WRF-alone forecasts under different sky conditions. In particular, in the case of clear sky conditions, the hybrid model can improve the forecasting accuracy by 95.7% and 90.9% in mean bias error (MBE), and 42.2% and 26.8% in root mean square error (RMSE) for Maizuru and Qinghai sites, respectively.

The millimeter-wave cloud radar, ceilometer, and disdrometer have been widely used to observe clouds and precipitation. However, there are some drawbacks when those three instruments are solely employed due to their own limitations, such as the fact that radars usually suffer from signal attenuation and ceilometers/disdrometers cannot provide measurements of the hydrometeors of aloft clouds and precipitation. Thus, in this paper, we developed an integrated technology by combining and utilizing the advantages of three instruments together to investigate the vertical structure and diurnal variation of warm clouds and precipitation, and the raindrop size distribution. Specifically, the technology consists of appropriate data processing, quality control, and retrieval methods. It was implemented to study the warm clouds and precipitation in South China during the pre-flood season of 2016. The results showed that the hydrometeors of warm clouds and precipitation were mainly distributed below 2.5 km and most of the rainfall events were very light with a rain rate less than 1 mm h⁻¹, however, the stronger precipitation primarily contributed the accumulated rain amount. Furthermore, a rising trend of cloud base height from 1000 to 1900 BJT was found. The cloud top height and cloud thickness gradually increased from 1200 BJT to reach a maximum at 1600 BJT (Beijing Standard Time, UTC+8), and then decreased until 2000 BJT. Also, three periods of the apparent rainfall on the ground of the day, namely, 0400–0700 BJT, 1400–1800 BJT, and 2300–2400 BJT were observed. During three periods, the raindrops had wider size spectra, higher number concentrations, larger rain rates, and higher water contents than at other times. The hydrometeor type, size, and concentration were gradually changed in the vertical orientation. The raindrop size distributions of warm precipitation in the air and on the ground were different, which can be expressed by $\gamma$ distributions N(D) = 1.49 × 10⁴D^−0.9484exp(−6.79D) in the air and N(D) = 1.875 × 10³D^0.862exp(−2.444D) on the ground, where D and N(D) denote the diameter and number concentration of the raindrops, respectively.

This work introduces a novel method to generate probabilistic hub‐height wind speed forecasts aimed at power output prediction. We employ state‐of‐the‐art convolutional variational autoencoders (CVAEs) trained with historical wind speed observations, multivariable outputs (wind speed, direction, temperature, pressure, and humidity) from a numerical weather prediction (NWP) model and spatio‐temporal encodings. After training, we exploit the CVAE data generating capabilities to produce probabilistic forecasts from the same deterministic dynamical NWP model. The resulting probabilistic forecast provides an insight into the uncertainty of the original deterministic input that compensate for errors due to numerical discretization, inaccuracies in initial/boundary conditions and parameterizations and, most importantly wind speed fluctuations due to complex terrain features. To show the performance of the proposed model, we validate the approach with forecasted and observed data for fifteen sites in a wind farm in Awaji Island, Japan, in a challenging zone with complex topography and highly fluctuating wind patterns. We show that the proposed method provides improved wind speed forecasts from both deterministic (reduced root mean square error) and probabilistic (reduced continuous ranked probability score) standpoints. We also use ensemble forecast validation methods to assess the statistical properties of the CVAE and conclude that in zones with rapidly changing wind speed dynamics, the CVAE ensemble performs in a superior way when compared to the analog ensemble method. Finally, we also provide an insight on how the results can be used to obtain reliable wind power estimations.

This work introduces a novel error correction method for short‐term, hub‐height wind speed forecasting systems aimed at power output prediction. We present a multivariable neural network that is trained to reduce the error in wind speed predictions out of a numerical weather prediction (NWP) model, by exploiting hidden information in additional atmospheric variables, that is, wind direction, temperature, and pressure. The unique layout of the network was influenced by that of denoising autoencoders, and their ability to learn mapping functions. The predicted values from the NWP model, which incorporate errors due to numerical discretization, inaccuracies in initial/boundary conditions and parametrizations, complex terrain features, etc., are mapped to a more accurate prediction in which the errors have been reduced. To show the performance of the proposed model, training and validation are carried out with 4 years of forecasted and observed data for fifteen sites in a wind farm in Awaji Island, Japan, in a challenging zone with complex topography and therefore complicated, highly fluctuating wind patterns. Moreover, a single variable (i.e., wind speed) network is also implemented in order to expose the contribution and usefulness of including additional atmospheric variables. The results show a considerable reduction in the root mean square error as well as an increase in the correlation coefficient. As expected, it is found that multiple meteorological variables as inputs offer a huge advantage when compared with the equivalent single‐variable correction method.

The mesoscale Weather Research and Forecasting (WRF) model has been widely employed to forecast day-ahead rainfalls. However, the deterministic predictions from the WRF model incorporate relatively large errors due to numerical discretization, inaccuracies in initial/boundary conditions and parameterizations, etc. Among them, the uncertainties in parameterization schemes have a huge impact on the forecasting skill of rainfalls, especially over the Sichuan Basin which is located east of the Tibetan Plateau in southwestern China. To figure out the impact of various parameterization schemes and their interactions on rainfall predictions over the Sichuan Basin, the Global Forecast System data are chosen as the initial/boundary conditions for the WRF model and 48 ensemble tests have been conducted based on different combinations of four microphysical (MP) schemes, four planetary boundary layer (PBL) schemes, and three cumulus (CU) schemes, for four rainfall cases in summer. Compared to the observations obtained from the Chinese ground-based and encrypted stations, it is found that the Goddard MP scheme together with the asymmetric convective model version 2 PBL scheme outperforms other combinations. Next, as the first step to explore further improvement of the WRF physical schemes, the polynomial chaos expansion (PCE) approach is then adopted to quantify the impacts of several empirical parameters with uncertainties in the WRF Single Moment 6-class (WSM6) MP scheme, the Yonsei University (YSU) PBL scheme and the Kain-Fritsch CU scheme on WRF rainfall predictions. The PCE statistics show that the uncertainty of the scaling factor applied to ice fall velocity in the WSM6 scheme and the profile shape exponent in the YSU scheme affects more dominantly the rainfall predictions in comparison with other parameters, which sheds a light on the importance of these schemes for the rainfall predictions over the Sichuan Basin and suggests the next step to further improve the physical schemes.

Under different water vapor and dynamic conditions, and the influence of topographies and atmospheric environments, stratiform precipitation over South China and the Tibetan Plateau can produce different features. In this study, stratiform precipitation vertical characteristics, bright-band (BB) microstructures, and the vertical variations of the raindrop size distribution (DSD) over a low-altitude site (Longmen site, 86 m) in South China and a high-altitude site (Nagqu site, 4507 m) on the Tibetan Plateau were comprehensively investigated and compared using measurements from a Ka-band millimeter-wave cloud radar (CR), a K-band microrain radar (MRR), and a Parsivel disdrometer (disdrometer). A reliable BB identification scheme was proposed on the basis of CR variables and used for stratiform precipitation sample selection and further statistics and analysis. Results indicate that melting layers over the Longmen are much higher and slightly thicker than those over the Nagqu due to significant differences in atmospheric conditions. For stratiform precipitation, vertical air motions and radar variables over the two sites show different variation trends from cloud top to the ground. Vertical air motions are very weak in the stratiform precipitation over the Longmen, whereas updrafts are more active over the Nagqu. Above the melting layer, radar equivalent reflectivity factor $Z_e$ (mean Doppler velocity $V_M$) gradually increases (decreases) as height decreases over the two sites, but the aggregation rate for ice particles over the Longmen can be faster. In the melting layer, $Z_e$ ($V_M$) at the BB bottom/center over the Longmen is larger (smaller) than those over the Nagqu for the reason that melted raindrops in the melting layers over the Longmen are larger than those over the Nagqu. Below the melting layer, profiles of radar variables and DSDs show completely different behaviors over the two sites, which reflects that the collision, coalescence, evaporation, and breakup processes of raindrops are different between the two sites. Over the Longmen, collision and coalescence dominate the precipitation properties; in particular, from 2.0–2.8 km, the breakup process competes with collision–coalescence processes but later is overpowered. In contrast, due to the lower BB heights over the Nagqu, collision and coalescence dominate raindrop properties. Comparisons of raindrop spectra suggest that the concentration of small (medium-to-large) raindrops over the Nagqu is much higher (slightly lower) than that over the Longmen. Therefore, the mass-weighted mean diameter $D_m$ (the generalized intercept parameter $N_w$) over the Nagqu is smaller (larger) than that over the Longmen.