High penetration and output volatility of island wind and photovoltaics (PV) pose challenges to energy consumption and supply–demand balance, and cost-effective energy storage configuration. A coupled dispatch model for a wind–PV–storage system is proposed, which treats multiple canal units as virtual ‘loads’ that switch between generation and pumping under constraints of power balance and available water head model. Considering the variable reservoir–irrigation feature, a multi-objective model framework is developed to minimize both economic cost and storage capacity required. An augmented Lagrangian–Nash product enhanced NSGA-II (AL-NP-NSGA-II) algorithm enforces constraints of irrigation shortfall and overflow via an augmented Lagrangian term and allocates fair benefits across canal units through a Nash product reward. Moreover, updates of Lagrange multipliers and reward weights maintain power balance and accelerate convergence. Finally, a case simulation (3.7 MW wind, 7.1 MW PV, and 24 h rural load) is performed, where 440.98 kWh storage eliminates shortfall/overflow and yields 1.5172 × 104 CNY. Monte Carlo uncertainty analysis (±10% perturbations in load, wind, and PV) shows that increasing storage to 680 kWh can stabilize reliability above 98% and raise economic benefit to 1.5195 × 104 CNY. The dispatch framework delivers coordination of irrigation and power balance in island microgrids, providing a systematic configuration solution.
To improve performance in day-ahead photovoltaic forecasting under irradiance fluctuations, meteorological conditions transitions, bias existed between numerical weather prediction (NWP) and local data measurements (LDM), as well the limited ability of a single forecasting model on global patterns capture in complex and highly variable data, a residual-downscaled multi-scenario dynamic fusion (RD-MSDF) framework is proposed. Firstly, historical paired data of NWP and LDM are used to train LS-Boost models, so to learn the residual features among physically matched meteorological variables, further add which to the corresponding NWP baselines to generate site-adapted meteorological features. After standardization, principal component analysis (PCA) is applied to reduce the dimensionality of corrected meteorological variables and reduce feature redundancy. The retained principal components are further combined with historical power data, fluctuation-related features, and temporal descriptors, so to construct unified forecasting inputs. Secondly, to characterize scenario-dependent operating conditions, a hierarchical scenario representation is developed. Rule-based hard scenarios are constructed according to corrected irradiance levels, irradiance fluctuations, day/night states, and historical power-fluctuation characteristics, while a Gaussian mixture model is employed to obtain soft-scenario posterior membership probabilities. Thirdly, LSTM and RBF network are adopted as complementary base predictors to capture temporal dependence and local nonlinear variations, respectively. Their preliminary outputs are calibrated through a soft-scenario mixture-of-experts module and subsequently fused by a hard-scenario-based Scene-EWAF++ mechanism, which maintains independent fusion-weight buckets for different operating scenarios and updates the active weights according to recently observed prediction errors. Finally, the proposed framework is evaluated with a public dataset from a 20 MW photovoltaic plant across four seasonal test sets. The comparison results show that it can obtain lower forecasting errors than other methods under varying irradiance and power-fluctuation conditions. The observed error reduction can effectively save the backup capacity for scheduling, as well potential cost of supply and demand power balance dispatch and carbon emission.