An improved CRNN for Vietnamese Identity Card Information Recognition

This paper proposes an enhancement of an automatic text recognition system for extracting information from the front side of the Vietnamese citizen identity(CID)card.First,we apply Mask-RCNN to segment and align the CID card from the background.Next,we present two approaches to detect the CID card’s text lines using traditional image processing techniques compared to the EAST detector.Finally,we introduce a new end-to-end Convolutional Recurrent Neural Network(CRNN)model based on a combination of Connectionist Temporal Classification(CTC)and attention mechanism for Vietnamese text recognition by jointly train the CTC and attention objective functions together.The length of the CTC’s output label sequence is applied to the attention-based decoder prediction to make the final label sequence.This process helps to decrease irregular alignments and speed up the label sequence estimation during training and inference,instead of only relying on a data-driven attention-based encoder-decoder to estimate the label sequence in long sentences.We may directly learn the proposed model from a sequence of words without detailed annotations.We evaluate the proposed system using a real collected Vietnamese CID card dataset and find that our method provides a 4.28%in WER and outperforms the common techniques.

State of charge estimation for electric vehicles using random forest

This paper introduces an innovative approach to addressing a critical challenge in the electric vehicle(EV)industry—the accurate estimation of the state of charge(SOC)of EV batteries under real-world operating conditions.The electric mobility landscape is rapidly evolving,demanding more precise SOC estimation methods to improve range prediction accuracy and battery management.This study applies a Random Forest(RF)machine learning algorithm to improve SOC estimation.Traditionally,SOC estimation has posed a formidable challenge,particularly in capturing the complex dependencies between various parameters and SOC values during dynamic driving conditions.Previous methods,including the Extreme Learning Machine(ELM),have exhibited limitations in providing the accuracy and robustness required for practical EV applications.In contrast,this research introduces the RF model,for SOC estimation approach that excels in real-world scenarios.By leveraging decision trees and ensemble learning,the RF model forms resilient relationships between input parameters,such as voltage,current,ambient temperature,and battery temperatures,and SOC values.This unique approach empowers the model to deliver precise and consistent SOC estimates across diverse driving conditions.Comprehensive comparative analyses showcase the superiority of the RF over ELM.The RF model not only outperforms in accuracy but also demonstrates exceptional robustness and reliability,addressing the pressing needs of the EV industry.The results of this study not only underscore the potential of RF in advancing electric mobility but also suggest a promising integration of the SOC estimation approach into the battery management system of BMW i3.This integration holds the key to more efficient and dependable electric vehicle operations,marking a significant milestone in the ongoing evolution of EV technology.Importantly,the RF model demonstrates a lower Root Mean Squared Error(RMSE)of 5.902,8%compared to 6.312,7%for ELM,and a lower Mean Absolute Error(MAE)of 4.432,1%versus 5.111,2%for ELM across rigorous k-fold cross-validation testing,reaffirming its superiority in quantitative SOC estimation.

Hybrid firefly algorithm-neural network for battery remaining useful life estimation

Accurately estimating the remaining useful life(RUL)of batteries is crucial for optimizing maintenance,preventing failures,and enhancing reliability,thereby saving costs and resources.This study introduces a hybrid approach for estimating the RUL of a battery based on the firefly algorithm–neural network(FA–NN)model,in which the FA is employed as an optimizer to fine-tune the network weights and hidden layer biases in the NN.The performance of the FA–NN is comprehensively compared against two hybrid models,namely the harmony search algorithm(HSA)–NN and cultural algorithm(CA)–NN,as well as a single model,namely the autoregressive integrated moving average(ARIMA).The comparative analysis is based mean absolute error(MAE)and root mean squared error(RMSE).Findings reveal that the FA–NN outperforms the HSA–NN,CA–NN,and ARIMA in both employed metrics,demonstrating su-perior predictive capabilities for estimating the RUL of a battery.Specifically,the FA–NN achieved a MAE of 2.5371 and a RMSE of 2.9488 compared with the HSA–NN with a MAE of 22.0583 and RMSE of 34.5154,the CA–NN with a MAE of 9.1189 and RMSE of 22.4646,and the ARIMA with a MAE of 494.6275 and RMSE of 584.3098.Additionally,the FA–NN exhibits significantly smaller maximum errors at 34.3737 compared with the HSA–NN at 490.3125,the CA–NN at 827.0163,and the ARIMA at 1.16e+03,further emphasizing its robust performance in minimizing prediction inaccuracies.This study offers important insights into battery health management,showing that the proposed method is a promising solution for precise RUL predictions.

Forecasting solar power generation using evolutionary mating algorithm-deep neural networks

This paper proposes an integration of recent metaheuristic algorithm namely Evolutionary Mating Algorithm (EMA) in optimizing the weights and biases of deep neural networks (DNN) for forecasting the solar power generation. The study employs a Feed Forward Neural Network (FFNN) to forecast AC power output using real solar power plant measurements spanning a 34-day period, recorded at 15-minute intervals. The intricate nonlinear relationship between solar irradiation, ambient temperature, and module temperature is captured for accurate prediction. Additionally, the paper conducts a comprehensive comparison with established algorithms, including Differential Evolution (DE-DNN), Barnacles Mating Optimizer (BMO-DNN), Particle Swarm Optimization (PSO-DNN), Harmony Search Algorithm (HSA-DNN), DNN with Adaptive Moment Estimation optimizer (ADAM) and Nonlinear AutoRegressive with eXogenous inputs (NARX). The experimental results distinctly highlight the exceptional performance of EMA-DNN by attaining the lowest Root Mean Squared Error (RMSE) during testing. This contribution not only advances solar power forecasting methodologies but also underscores the potential of merging metaheuristic algorithms with contemporary neural networks for improved accuracy and reliability.

上位效应对遗传算法可靠性的影响（英文）

目的:探讨遗传算法的局限性和实用性,并分析基于相互作用产生的上位效应对遗传算法可靠性的影响。创新点:1.指出遗传算法缺陷的根源;2.基于测试样本函数定义目标函数,以判断遗传算法的适用性。方法:1.基于非上位效应函数(表1)和上位效应函数(表2),以及非上位效应函数F4和上位效应函数F6的结构图来验证遗传算法可靠性;2.通过计算样本函数(公式(1))和遗传算法流程(图3)表达遗传算法的工作原理。3.利用克洛弗函数(公式(2))和计算不同结构角下的函数分布(图4),进一步判断匹配度(表3)和计算效率(表4);定义新的目标函数(公式(9))和一组新的变量(公式(10))来实现变量相关性解离。结论:1.对当前遗传算法存在的不足给出了独到见解,并认为正定性的假设并非可以保证遗传算法实际的有效性和优化性。2.定义成本代价函数用以判断遗传算法可靠性,并分别考虑上位性和非上位性效应两种情形。当成本代价函数在非上位性效应下时,遗传算法是有效的;否则,可以把N维函数降级为N个一维函数,从而采用更简单的算法来判断。基于一些通用的基准,进一步设计三类样本函数来证实以上判断,且这些样本函数适合于上位性效应情形和非上位效应情形。3.遗传算法的瓶颈在于主算子和相干匹配性;可以通过破坏某些结构来实现变量关系的解离,从而抑制相干匹配性对遗传算法的影响。希望相关读者在处理实际优化问题时能验证作者关于上位效应的定性结论,并给出更可靠的方法来表征这种效应。