Temporal feature decomposition fusion network for building energy multi-step prediction

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: 94, P. 110034 - 110034

Published: June 22, 2024

Language: Английский

---

EV load forecasting using a refined CNN-LSTM-AM

Electric Power Systems Research, Journal Year: 2024, Volume and Issue: 238, P. 111091 - 111091

Published: Sept. 23, 2024

Language: Английский

---

Network Security Situation Prediction Model Based on VMD Decomposition and DWOA Optimized BiGRU-ATTN Neural Network

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 129507 - 129535

Published: Jan. 1, 2023

The widespread adoption of Internet-of-Things (IoT) devices has resulted in a comprehensive transformation human life. However, the network security challenges posed by IoT have become increasingly severe, necessitating implementation effective mechanisms. Network situational awareness enables an state prediction for better formulation defense strategies. Existing methods are typically constrained sequence data, especially those sequences with high degree non-stationarity, leading to unstable predictions and low performance. Moreover, real-world application scenarios, often highly non-stationary. To address these challenges, we introduce novel hybrid model named Variational Mode Decomposition (VMD) - Dynamic Whale Optimization Algorithm (DWOA) Bidirectional Gated Recurrent Unit (BiGRU) Attention Mechanism (ATTN). proposed integrates VMD, BiGRU, ATTN, DWOA. Initially, processed using VMD decompose them into series subsequences, thus reducing non-stationarity original sequences. Subsequently, enhanced DWOA optimization algorithm is introduced tuning hyperparameters BiGRU-ATTN network. Ultimately, employed predict each which then aggregated yield final value. When compared several existing on public datasets, VMD-DWOA-BiGRU-ATTN method demonstrated improvement R ² values ranging from 6.34% 52.61%. These results substantiate that significantly enhances predictive

Language: Английский

---

Informer-based model predictive control framework considering group controlled hydraulic balance model to improve the precision of client heat load control in district heating system

Applied Energy, Journal Year: 2024, Volume and Issue: 373, P. 123951 - 123951

Published: July 19, 2024

Language: Английский

---

Machine learning para la predicción de energía eléctrica: una revisión sistemática de literatura

Ingeniería y Competitividad, Journal Year: 2024, Volume and Issue: 26(2)

Published: July 11, 2024

Este estudio presenta una Revisión Sistemática de la Literatura (RSL) sobre las técnicas inteligencia artificial (IA) aplicadas para predicción energía eléctrica. Las bases datos especializadas que se emplearon en esta revisión son Scopus, IEEE, ACM y Google Scholar. análisis ofreció perspectiva utilizadas este campo, lo facilitó identificación tendencias presentes desarrollo. Esto proporciona comprensión clara oportunidades venideras mejorar precisión eléctrica y, consecuencia, toma decisiones.Un hallazgo destacado fue el predominio del uso redes neuronales artificiales (RNA) como técnica más prevalente dentro campo Machine Learning aplicado a Esta preferencia justifica por capacidad inherente RNA identificar patrones complejos relaciones los datos, convierte herramienta valiosa Además, destaca importancia varios factores fundamentales eléctrica, recolectar relevantes representativos, abarquen tanto información histórica contextual. El preprocesamiento cual implica limpieza transformación recopilados prepararlos adecuadamente su modelado división crucial evitar sesgos evaluar manera precisa predictiva modelo.

---

A New Design of an Optimized Informer Wind Power Prediction Model Utilizing Wind Turbine Health Assessment

Journal of Circuits Systems and Computers, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 11, 2024

Wind power prediction is of significant value to the stability grid. Employing Informer model for wind yields better results than traditional neural networks, yet issues such as slow speed and insufficient accuracy persist. By utilizing a health assessment algorithm optimize model, both can be concurrently enhanced. Initially, matrix obtained by performing turbines based on operational data. Subsequently, this used encoding method Informer, improving speed. Simultaneously, decoding method, embedding vectors process are refined increase accuracy. Finally, conventional models optimized tested compared using four distinct datasets. The indicate that achieves an approximately 15% in about 100%

Language: Английский

---

PMformer: A Novel Informer-based Model for Accurate Long-Term Time Series Prediction

Information Sciences, Journal Year: 2024, Volume and Issue: unknown, P. 121586 - 121586

Published: Oct. 1, 2024

Language: Английский

---

Prediction of Total Phosphorus Concentration in Canals by GAT-Informer Model Based on Spatiotemporal Correlations

Water, Journal Year: 2024, Volume and Issue: 17(1), P. 12 - 12

Published: Dec. 24, 2024

The accurate prediction of total phosphorus (TP) is crucial for the early detection water quality eutrophication. However, predicting TP concentrations among canal sites challenging due to their complex spatiotemporal dependencies. To address this issue, study proposes a GAT-Informer method based on correlations predict in Beijing–Hangzhou Grand Canal Basin Changzhou City. begins by creating feature sequences each site time lag relationship concentration between sites. It then constructs graph data combining real river distance and correlation sequences. Next, spatial features are extracted fusing node using attention (GAT) module. employs Informer network, which uses sparse mechanism extract temporal efficiently simulating model was evaluated R2, MAE, RMSE, with experimental results yielding values 0.9619, 0.1489%, 0.1999%, respectively. exhibits enhanced robustness superior predictive accuracy comparison traditional models.

Language: Английский

---

Improving the Accuracy of Multi-Step Prediction of Railway Freight Volume Based on Informer Model

Published: Jan. 12, 2024

Language: Английский

---

Short-Term Photovoltaic Power Forecasting Using Ssa-Based Informer Combined with Fcm-Based Similar Day Selection and Mrsvd-Vmd Decomposition

Published: Jan. 1, 2024

Accurate photovoltaic power (PV) prediction technology is critical for ensuring the safe, stable and economic operation of grid. To maintain efficient system, this study proposes a short-term combined PV output forecasting model with high accuracy. Firstly, key meteorological factors that significantly affect are determined through Convergent Cross Mapping (CCM) method. Secondly, identified considered as input improved Fuzzy C-means algorithm (IFCM) classifying historical samples predicted days into three weather scenarios (Sunny, Cloudy Rainy). Thirdly, integration Multi-resolution Singular Value Decomposition Variational Mode (i.e. MRSVD-VMD) utilized to capture pattern trend involved original series, decomposing it several Intrinsic Functions (IMFs). Finally, Sparrow Search Algorithm (SSA) used optimize hyperparameters Informer model, which applied establish high-precision corresponding each IMF. The summation results all IMFs final outcome. experimental in plant Yunnan province, China indicate proposed CCM-IFCM-MRSVD-VMD-SSA-Informer method has highest accuracy stability compared other benchmark models, where evaluated index Mean Absolute Error, MAE) under 0.434, 0.480 0.363, respectively, validated effectiveness model.

Language: Английский

---