Purpose Solar power generation exhibits significant time-varying seasonal fluctuations and nonlinear trends, which pose a challenge to existing forecasting techniques. Therefore, this paper proposes a novel grey forecasting framework that integrates seasonal decomposition, nonlinear modeling, and time-varying adaptation, namely the seasonal time-varying discrete grey power model. Design/methodology/approach Firstly, seasonal dummy variables are introduced to separate periodic fluctuations from the trend. Secondly, the seasonal effect is innovatively decomposed into fixed and time-varying components, precisely capturing the temporal evolution patterns across different periods. Concurrently, the double-power index structure quantifies the nonlinear driving effects of both historical values and temporal factors on the current state. The hiking optimization algorithm is employed for hyper-parameter tuning. Findings The new model was validated using datasets from the three major economies of the U.S., India, and China, and compared with six competitive models representing different types. Results show that the new model exhibits superior predictive accuracy. Furthermore, robustness tests across multiple time windows further validate the stability of the model under varying scenarios. Practical implications Forecasts for 2025–2027 indicate sustained growth in solar power generation across all three nations, but the magnitude and patterns of growth exhibit significant differences. The research findings provide a reliable decision support tool for solar power planning and grid dispatch in the global energy transition. Originality/value The new model effectively captures the complex dynamics in solar power generation data by integrating seasonal time trend terms, dummy variables, and power index terms into the discrete grey modeling system.
Purpose This study examines relational closeness between university students' academic achievement, represented by cumulative grade point average (CGPA) and mental health, well-being, sociodemographic variables using a self-reported student survey. Anxiety, depression, stress, sleep quality, diet quality, physical energy level, age, gender, department and scholarship status were analyzed. Relational and predictive relevance was evaluated without assuming causality, with an ergonomic interpretation of well-being indicators. Design/methodology/approach Grey relational analysis (GRA) estimated relational degrees between CGPA and selected variables. Deng's, absolute, relative, comprehensive and negative GRA models were applied, with sensitivity analyses across distinguishing coefficient values. Cronbach's alpha and exploratory factor analysis (EFA) examined whether anxiety, depression and stress represented a common structure. They were combined into a risk construct defined as “mental health risk.” Seven machine learning algorithms classified this construct. Repeated randomized train-test experiments assessed accuracy, precision, F1-score and receiver operating characteristic (ROC) area under the curve (AUC). Findings Anxiety, depression and stress showed relatively high relational degrees with CGPA compared with the other variables. These findings suggest that academic achievement is related not only to demographic characteristics but also to psychological well-being and ergonomics-related conditions of students. According to the machine learning results, AdaBoost achieved the highest mental health risk classification performance, with an accuracy of 82.98%. Originality/value This study integrates comparative GRA modeling, statistical construction of a mental health risk indicator, repeated machine learning evaluation and ergonomic interpretation. The framework links relational analysis with predictive classification while distinguishing analytical risk from clinical diagnosis.
Purpose This paper aims to optimize the structural parameters of a newly developed coaxial electromagnetic buffer of artillery that integrates an eddy-current recoil brake and a spring-type recuperator to achieve low-recoil and stable braking performance. Design/methodology/approach The dynamic equations governing the recoil and counter-recoil processes are formulated, followed by a finite element model incorporating the nonlinear hysteresis and temperature-dependent characteristics of NdFeB permanent magnets. Multi-response grey hierarchical clustering is then proposed to identify key structural parameters. Based on the screened design variables, a Kriging surrogate model is constructed, and a hybrid multi-objective optimization strategy combining the multi-island genetic algorithm and sequential quadratic programming is developed to determine the optimal structural design. Findings The finite element model achieves high prediction accuracy, with dynamic response errors below 5% under three charge conditions. The optimized design reduces the peak recoil resistance from 39.81 kN to 34.68 kN, representing a 12.89% reduction, while satisfying the prescribed recoil displacement constraint. Practical implications The proposed electromagnetic buffer provides an effective alternative to conventional hydraulic recoil systems. The physically validated finite element model and surrogate-based optimization framework can support the structural design and performance improvement of recoil mechanisms for artillery systems. Originality/value This work proposes a multi-response grey hierarchical clustering method for design-variable sensitivity analysis, enabling comprehensive variable screening by jointly accounting for their influence across multiple performance responses.
Purpose The traffic flow system is a complex dynamic system, and the traffic flow data have complex spatiotemporal characteristics. Therefore, to deeply explore the spatiotemporal characteristics of traffic flow data, a novel spatiotemporal multivariate partial grey prediction model for multi-segment traffic flow is proposed, which is suitable for short-term traffic flow prediction. Design/methodology/approach The model adopts grey relational degree analysis to screen multi-segment traffic flow data and further selects time-series data from different periods based on the periodic patterns of traffic flow to capture the spatiotemporal characteristics of traffic datasets. Matrix analysis is then performed for parameter estimation and model construction. To enhance model applicability, particle swarm optimization is employed to optimize the background value coefficient, thereby reducing prediction errors and improving prediction accuracy. Findings The proposed model achieves mean absolute percentage error (MAPE) values below 7% across all evaluated scenarios. Short-term traffic flow prediction is examined over three time periods from three perspectives. The predicted trends closely align with the actual traffic flow sequences, confirming the model’s strong predictive performance. Originality/value The effectiveness of the new model is illustrated by three effectiveness analysis cases. The results are better than comparison models that include several multivariate grey prediction models and deep learning models, which shows the effectiveness of the proposed model. The simulation results of three effectiveness analysis cases are applied to the short-time traffic flow prediction problem.
Purpose Rapid urbanization, population growth, and shifting consumption patterns have placed increasing pressure on urban water supply systems, making reliable forecasting of monthly clean water delivery essential for effective infrastructure planning and sustainable resource management. To address this need, this study proposes a novel seasonal grey model to forecast monthly urban water demand. Design/methodology/approach This study proposes a novel Seasonal Optimized New Information Priority Conformable Fractional Grey Bernoulli Model, integrating a two-step moving average seasonal decomposition, a conformable fractional accumulation operator satisfying the new information priority principle, a nonlinear grey Bernoulli differential equation, and the Jaya algorithm for simultaneous parameter tuning. Findings The proposed model is validated on the monthly observations of clean water demand in Istanbul and benchmarked against several forecasting models including the Seasonal Autoregressive Integrated Moving Average (SARIMA) model and Long Short-Term Memory (LSTM). The results reveal that the proposed model outperforms traditional forecasting models in terms of both accuracy and predictive power. Its transferability is further examined through two additional case studies on monthly water supply in Izmir, Turkey and monthly residential water consumption in Austin, Texas, where the model likewise achieves highly accurate forecasts. Research limitations/implications A limitation of this study stems from the unavailability of complete annual data for the most recent years — 2024 and 2025 for Istanbul, and 2025 for Izmir and Austin, where only partial-year observations were available at the time of analysis. As with any data-driven methodology, this constraint may affect the accuracy of forecasts for subsequent years. Originality/value This study is the first to integrate the new information priority conformable fractional accumulation operator into a seasonal grey Bernoulli framework and the first to apply it to urban water demand forecasting, preserving computational simplicity while ensuring that recent observations receive greater influence.
