Energy Informatics最新文献

In smart substations, equipment protection and abnormal warning are crucial to the safe and stable operation of the power grid. In existing research, traditional methods such as LSTM (Long Short-Term Memory Network) are time-consuming and require large computing resources, and SVM (Support Vector Machine) is easy to fall into local optimization and limited generalization ability when processing high-dimensional data, making it difficult to efficiently realize deep feature mining and accurate early warning of multi-source data. In this study, it is proposed to achieve accurate early warning through “synchronization data of similar equipment” (i.e., simultaneous collection of operating data of the same type of equipment, such as load current and oil temperature of multiple transformers on the same bus, so as to facilitate mutual verification in case of abnormality) and multi-source data fusion technology. Firstly, high-precision sensors are used to collect electrical and non-electric data such as voltage, current, equipment temperature, and vibration in real time, and various data features are integrated into a unified vector (such as combining power trend and vibration frequency characteristics) through feature-level fusion, and then redundancy is removed by dimensionality reduction algorithms such as PCA. The core model uses the “Whale Optimization Extreme Learning Machine” (WOA-ELM): WOA simulates the initial parameters of the Whale Predation Behavior Optimization Extreme Learning Machine (ELM), and the ELM exerts fast learning and strong generalization capabilities to deeply mine the processed multi-source data features. Experiments show that compared with the traditional model, the accuracy of equipment anomaly identification is improved by about 20%, and the early warning response time is shortened by 30%, which significantly improves the efficiency and reliability of equipment protection in intelligent substation. It not only provides a stronger guarantee for the safe operation of equipment, but also shows potential application value in the early abnormal warning system, which can help the power system achieve more efficient preventive maintenance.

This paper demonstrates how explainable machine learning (XAI) can be operationalised as a methodological pathway for formalising engineering knowledge from high-frequency building operational data. We propose a modular pipeline that combines feature engineering, ensemble and sequence learners, SHAP attribution and uncertainty quantification to convert raw sensor streams into machine-readable knowledge artefacts (JSON schema) suitable for automation workflows such as fault detection and demand response. Using a monitored nearly Zero-Energy Building (nZEB) in Lisbon (12 months, 5-minute resolution), we (i) report model performance (LightGBM, Random Forest, SVR, Linear Regression, and LSTM) under time-aware 70/15/15 split and 5-fold temporal cross-validation; (ii) present SHAP-based global and local attribution analyses that identify stable seasonal drivers; and (iii) provide computational cost (training and inference times) and uncertainty quantification. Results show ensemble models achieve superior short-term forecasting accuracy while producing consistent, actionable attributions that can be encoded as reusable artefacts. We close by describing a JSON artefact schema and outlining how these artefacts could be integrated within digital twins and supervisory control systems.

The economic viability of distributed energy resources (DERs) like wind turbines and photovoltaic (PV) units is hampered by their low conversion efficiencies and ineffective energy management. This research aims to improve Maximum Power Point Tracking (MPPT) in a hybrid wind-PV power system to reduce voltage and current oscillations by using the properties of DC link behavior. The innovation in this study stems from creating an integrated MPPT supervisor that responds to both the partial shading of PV modules and the variability of wind simultaneously and without utilizing current or voltage sensors. An artificial neural network (ANN) is used for this purpose, and its parameters are tuned by the Particle Swarm Optimization (PSO) algorithm. The proposed strategy was implemented in simulations compared to dynamic P&O, standalone ANN, and hybrid PSO-ANN frameworks. Based on the simulations performed, the PSO-ANN controller outperforms other methods by achieving better efficiency in MPPT when using DC link voltage as input and lowering Total Harmonic Distortion (THD). Also, the controller reduces the DC link voltage ripple while attenuating current and voltage THD to 3% and 2%, respectively. Moreover, during islanded operation, the controller decreases the distortion by 1.27%, showing enhanced system stability without traditional feedback control.

Due to the global increase in the share of renewable energy sources in the share of generation in distribution electric networks around the world, an acute problem of power fluctuations has arisen. Battery energy storage systems make it possible to smooth out peak loads, compensate for the instability of renewable energy generation, provide backup power and optimize network operation, including in the context of reducing total active power losses in the network. However, their implementation is associated with a number of technical, economic and regulatory challenges that require an integrated approach to integration and management. The purpose of this article is to select a connection node and capacity of a battery storage system in a distribution electric network in terms of minimization of total daily active energy losses in the network. An algorithm is presented that allows selecting a connection node for the storage system in terms of minimization of total daily active energy losses. Using a test circuit of a distribution electric network as an example, a modeling of the connection of a battery energy storage system was carried out, the values of active power losses and the capacities of the battery system were obtained.

The increasing penetration of renewable energy introduces significant uncertainty to the operation of energy-intensive industries. To address this challenge, we propose a multi-timescale scheduling model for wind power integration in polysilicon industrial parks. The model consists of a day-ahead component, which minimizes operational costs by coordinating polysilicon reduction and water electrolysis hydrogen production while allocating reserve capacity. A chance-constrained programming approach is employed to determine reserve capacity, ensuring a balance between reliability and economic efficiency. The intraday component, benefiting from improved wind power forecasting accuracy, engages only the water electrolysis process to absorb excess renewable generation. By involving water electrolysis in both timescales, the model enhances operational flexibility and renewable utilization. Case studies demonstrate the effectiveness of the proposed approach, highlighting its potential to support low-carbon and cost-efficient operation in polysilicon industrial parks.

The rapid advancement of smart grids necessitates robust dynamic assessment of wireless communication link quality, which faces dual challenges: complex electromagnetic interference (EMI) and the need for effective multi-source temporal data correlation modeling. Traditional methods relying on manual expertise and existing data-driven models often inadequately capture intricate multi-source temporal features. To address these limitations, this paper proposes a novel contrastive learning-based model for wireless link quality assessment in smart grids. Our framework employs Link Quality Indicator (LQI), Received Signal Strength Indicator (RSSI), and Signal-to-Noise Ratio (SNR) as multi-view inputs. A cross-view semantic alignment strategy is introduced to extract noise-robust shared features across these heterogeneous indicators. Furthermore, we design a hybrid attention temporal encoder integrating Long Short-Term Memory (LSTM) networks, adaptive channel attention, and global temporal attention modules. This cascaded architecture achieves deep fusion of local dynamic feature enhancement and global long-range dependency modeling. Experimental validation on 48 hours of continuously collected real-world communication link data demonstrates that the proposed model outperforms baseline methods, achieving accuracy improvements of 2.5% to 7.7% with validated statistical significance. Specifically, for abnormal link states, the model maintains a high recall rate of over 92.1%, ensuring reliable fault detection. While maintaining high overall stability, we observe minor performance degradation under conditions of extreme burst noise or high rates of missing data. Crucially, it exhibits substantially enhanced robustness and generalization capability, particularly in identifying abnormal link states under challenging EMI conditions.

With the rapid development of renewable energy, photovoltaic power generation has become a key part of the global energy transition. Short-term photovoltaic prediction is critical for intra-day real-time power grid dispatching, and enhancing its accuracy is a key research focus. However, existing methods still have limitations in handling complex nonlinear relationships in photovoltaic temporal data. To tackle this, this paper proposes a new model combining Long Short-Term Memory (LSTM), Differential Transformer (DiffTransformer), and Multi-Objective Escape Algorithm (MOESC) for short-term photovoltaic power prediction optimization: Preprocessed data is input into the LSTM-Differential Transformer model, with the Differential Transformer encoder capturing fine-grained temporal changes via optimized multi-head attention and rotary positional encoding, and the LSTM decoder integrating local temporal information for power prediction. Subsequently, Pareto-improved MOESC performs multi-objective optimization on the model’s key parameters (balancing RMSE, MAE, and R²), with the optimal parameters selected from the Pareto frontier. Experiments based on the Guoneng Rixin photovoltaic dataset show that, with user-defined weights (RMSE: 30%, MAE: 30%, R²: 40%), this method outperforms XGBoost, LightGBM, SVR, LSTM, GRU and the unoptimized LSTM-Differential Transformer model in photovoltaic power prediction. It not only can effectively improve prediction accuracy but also exhibits better stability compared with the unoptimized LSTM-Differential Transformer model.

Accurate short-term wind power prediction (WPP) is critical for power system stability but remains challenging due to the inherent non-linearity and volatility of wind series. This study proposes a novel framework, SVMD-RIME-TCN-BiGRU, to address these challenges. First, the Maximal Information Coefficient (MIC) is used to select high-correlation features and eliminate redundancy. Second, Successive Variational Mode Decomposition (SVMD) decomposes raw data into successive intrinsic modes, effectively mitigating non-stationarity and avoiding the mode-mixing issues of traditional methods. Third, a hybrid Temporal Convolutional Network-Bidirectional Gated Recurrent Unit (TCN-BiGRU) model is constructed to extract spatiotemporal features. Crucially, the RIME optimization algorithm is introduced to automatically tune the key hyperparameters of the TCN-BiGRU, avoiding local optima. Experimental results on a Xinjiang wind farm dataset demonstrate that the proposed model achieves a Root Mean Square Error (RMSE) of 7.6882 and an R² of 0.9813. It significantly outperforms baseline models (including LSTM, TCN, and Transformer) and other hybrid variants, reducing errors by over 38% compared to the TCN-BiGRU baseline. This validates the framework’s reliability and accuracy for practical power dispatching.

Critical infrastructure systems such as the electric grid are increasingly cyber-physical, where communication and control are tightly intertwined with the physics-based processes of power flow. To ensure safe and resilient operation of these cyber-physical systems, a variety of sensors and analyses are required for monitoring and detection of abnormal or malicious behavior to achieve full cyber-physical situational awareness (CPSA). To share this collected data with different analysis platforms, whether intrusion detection systems or state estimation algorithms, secure data exchange is essential. Designing secure data exchange across interconnected systems of systems (SoS) can be challenging without considering unique characteristics of the underlying cyber and physical processes. It is important to consider different types of communication protocols, frequency of communications, and types of communications (e.g., sensor measurements, control commands). In this paper, we provide design principles and experimental results for secure and resilient data exchange across distributed sensors and analytics in decentralized, cyber-physical energy systems. Specifically, secure data exchange technologies such as IPFS, synchronic web, multichain, and storage/sharing principles are presented and experimental results are provided to assess their applicability to exemplar distributed CPSA sensors.

The study examines the impact of financial inclusion, fintech and foreign direct investment (FDI) on energy efficiency during 2004–2022 for ten emerging economies which include Brazil, China, Russia, South Africa, India, Malaysia, Indonesia, Thailand, Mexico and Turkey. Fixed effect ordinary least squares (FEOLS), fully modified ordinary least squares (FMOLS) and dynamic ordinary least squares (DOLS) techniques were applied to quantify the relationship among the variables in long run. Findings suggest that financial inclusion, fintech and FDI are important factors driving energy efficiency. Financial inclusion provides opportunities toinvestors to invest in energy-efficient technologies at a reduced cost. Availability of infrastructure for fintech is shown to have a favorable impact on energy efficiency. Therefore, investment in digital infrastructure should be prioritized which will increase the availability of fintech services. Policymakers should also take steps to channelize FDI inflows into research and development (R&D) which would help in developing energy efficient technologies.