• Energy Research
• 2025-2025close
• GB close
• AU close
• MY close
• Applied Energy close

This paper presents a novel coordinated Conservation Voltage Reduction (CVR) and Electric Vehicle Demand Control (EVDC) method for energy-efficient power system operation. The goal of coordinated CVR-EVDC is to minimise the total energy consumed by the distribution network through coordination of the scheduling of EV charging, the OLTC transformer tap position, and the switching of the capacitors on the network. Using publicly available statistics of journey length and travel time of car drivers in the UK, a stochastic model is developed for EV battery State-of-Charge (SOC) and its availability for home charging to generate EV charging demand profiles for different levels of EV penetration. These are then used in conjunction with Particle Swarm Optimisation (PSO) and a relaxation equality constraint to explore the potential for optimally coordinating CVR operation and the scheduling of charging of EVs on LV distribution networks. Results for five different LV network scenarios show that a coordinated CVR-EVDC approach enables more significant energy savings than with independent operation of CVR and EVDC – 4% greater with dumb charging and 3% greater with valley-filling smart charging. This study shows that distribution network operators can manage the networks by coordinating CVR and EVDC to manage consumer charging requirements for better energy savings.

© 2024Vehicle-to-grid (V2G) charging, where vehicles can send power to the grid, can provide valuable services to energy systems and network operators. However, social acceptance is an essential and overlooked barrier which must be addressed if V2G is to be successfully deployed. This study investigates the factors that govern attitudes towards V2G, and how electric vehicle (EV) ownership and participation in V2G changes them. For the first time, this includes survey data from users who had experience using a V2G charger, comparing the response of V2G users (n=49) with EV owners (n=520) and non-EV owners (n=1091). We show that time and EV ownership have lowered concerns around range anxiety, and that EV ownership and V2G trial participation leads to a 15%–35% increase in stated willingness to participate in V2G or Smart Charging as compared to a 2013 baseline. Additionally, it is demonstrated that the strongest single predictor for V2G willingness is whether the consumer believes that V2G can contributes to a stable electricity system. These results suggest that education around V2G benefits and allowing consumers to test V2G before committing could be key factors in increasing adoption. We also highlight the importance of data privacy, which for some consumers contributes towards a negative attitude towards V2G. We release the raw survey data and code with this manuscript.

This paper presents the first comparative study evaluating towing and onsite replacement strategies for heavy maintenance of floating offshore wind (FOW) turbines. The towing maintenance strategy is characterised by a Markov chain and implemented within a computationally-efficient operation and maintenance (O&M) model. This model includes all key phases of the towing strategy: transit-to-site, turbine disconnection, towing-to-port, component replacement, towing-to-site, turbine connection, and transit-to-port. Additionally, the paper provides the first spatial assessment of heavy maintenance for FOW turbines in the North Sea. Evaluation across the ScotWind area shows that onsite replacement can reduce turbine downtime, especially for quick heavy maintenance operations like blade and gearbox replacements. However, for longer operations, such as generator and pitch and hydraulic system replacements, onsite solutions are more effective than towing only when O&M vessels can operate in wave heights over 1.5 metres. Otherwise, a mixed heavy maintenance strategy is recommended, combining onsite replacements for blades and gearboxes with towing for generators and pitch and hydraulic systems. The average turbine availability reduction with the mixed strategy is 0.39%, followed by the fully towing strategy at 0.43%, and the fully onsite replacement strategy at 0.46%.

Electricity consumption has stochastic variabilities driven by the energy market volatility. The capability to predict electricity demand that captures stochastic variances and uncertainties is significantly important in the planning, operation and regulation of national electricity markets. This study has proposed an explainable deeply-fused nets electricity demand prediction model that factors in the climate-based predictors for enhanced accuracy and energy market insight analysis, generating point-based and confidence interval predictions of daily electricity demand. The proposed hybrid approach is built using Deeply Fused Nets (FNET) that comprises of Convolutional Neural Network (CNN) and Bidirectional Long-Short Term Memory (BILSTM) Network with residual connection. The study then contributes to a new deep fusion model that integrates intermediate representations of the base networks (fused output being the input of the remaining part of each base network) to perform these combinations deeply over several intermediate representations to enhance the demand predictions. The results are evaluated with statistical metrics and graphical representations of predicted and observed electricity demand, benchmarked with standalone models i.e., BILSTM, LSTMCNN, deep neural network, multi-layer perceptron, multivariate adaptive regression spline, kernel ridge regression and Gaussian process of regression. The end part of the proposed FNET model applies residual bootstrapping where final residuals are computed from predicted and observed demand to generate the 95% prediction intervals, analysed using probabilistic metrics to quantify the uncertainty associated with FNETS objective model. To enhance the FNET model’s transparency, the SHapley Additive explanation (SHAP) method has been applied to elucidate the relationships between electricity demand and climate-based predictor variables. The suggested model analysis reveals that the preceding hour’s electricity demand and evapotranspiration were the most influential ...