Maximum Absolute Error Research Articles

Combining Multi-Indirect Features Extraction and Optimized GPR Algorithm for Online State of Health Estimation of Lithium-ion Batteries

Abstract With the wide application of lithium batteries (LIBs) in electrified transportation and smart grids, especially in the pure electric vehicle industry, the accurate health maintenance monitor of LIBs has emerged as critical for safe battery operation. Although many data-driven methods with state of health (SOH) estimation for LIBs have been proposed, the problems of industrial generation and computational cost still need to be improved further. In contrast, this paper carried out a low-complexity SOH estimation method for LIBs. Specifically, the seven health indicators are extracted firstly to characterize battery health status from voltage, current, temperature and other data that can be obtained online. Then, the optimized gaussian process regression (GPR) algorithm is proposed with proper computational cost. Ultimately, combining a multi-indirect features extraction and optimized GPR algorithm, the online SOH estimation for LIBs was established and verified with NASA experiment data. The experimental results show that the maximum Mean Absolute Percentage Error (MAPE) of SOH estimation from the proposed method is 1.4496 and the minimum MAPE only reaches 0.5635. More importantly, the optimized GPR for SOH estimation can achieve a maximum 65.37% improvement under multiple evaluation criteria compared to traditional GPR. The method proposed in this paper is helpful for realizing on-line SOH estimation in battery management systems.

Collaborative control of circumferential yarn implantation system for large-scale 3D braiding equipment using fast nonsingular terminal sliding mode

To ensure high performance collaborative control of the circumferential yarn implantation system for a specific large-scale 3D braiding equipment, a multi-motor fast nonsingular terminal sliding mode (FNTSM) controller is proposed using time-delay estimation (TDE) and deviation coupling control (DCC). Then, comparative experiments were conducted among our proposed TDE-based FNTSM controller and PD controller and TDE-based PD controller. As shown in the experimental results, the root-mean-square error (RMSE) and maximum absolute error (MAE) ensured by our FNTSM controller tracking sine signal have decreased by 89.4% and 67.1% compared with the ones by PD controller; while the RMSE of our FNTSM controller tracking slope signal is ensured within 0.03°. Meanwhile, the RMSE and MAE of our FNTSM controller tracking sine signals has been reduced by 48.2% and 56.9% due to DCC, respectively. Afterwards, less than 3% error fluctuation has been observed with load using our FNTSM controller. High control performance and strong robustness have been experimentally observed.

Inverse design of cellular structures with the geometry of triply periodic minimal surfaces using generative artificial intelligence algorithms

Triply periodic minimal surfaces (TPMS) exhibit excellent mechanical and energy absorption properties due to their structural advantages. However, existing porous TPMS structural design methods are constrained to a forward process from structural parameters to mechanical properties. This study proposed an inverse design method that combines bidirectional generative adversarial networks (BiGAN) and mechanical performance targets, resulting in a combined TPMS structure of Primitive and IWP types with superior buffering and energy absorption capabilities. The results show that under a single load value target condition of the designed structure, the minimum deviation index (R2) between the load value corresponding to the displacement point and the target load value is only 0.987, and the maximum mean absolute percentage error (MAPE) is only 5.92 %. When considering the elastic modulus target, the approach successfully conducts two sets of combined structural designs meeting the requirements of both high and low elastic moduli. When targeting the specified load-displacement curve conditions, specifically when combining high elastic modulus with ascending plasticity, the designed structures exhibit an error of only 2.2 % compared to the target property. Moreover, the quasi-static uniaxial compression experiments conducted on additively manufactured designed structures confirm that the experimental curves match the target curves in terms of deformation trends and load value ranges. The success of this inverse design approach for cellular TPMS structures has the potential to expedite new structural material development processes.

Adaptive Nonsingular Fast Terminal Sliding Mode Control for Shape Memory Alloy Actuated System

Due to its high power-to-weight ratio, low weight, and silent operation, shape memory alloy (SMA) is widely used as a muscle-like soft actuator in intelligent bionic robot systems. However, hysteresis nonlinearity and multi-valued mapping behavior can severely impact trajectory tracking accuracy. This paper proposes an adaptive nonsingular fast terminal sliding mode control (ANFTSMC) scheme aimed at enhancing position tracking performance in SMA-actuated systems by addressing hysteresis nonlinearity, uncertain dynamics, and external disturbances. Firstly, a simplified third-order actuator model is developed and a variable gain extended state observer (VGESO) is employed to estimate unmodeled dynamics and external disturbances within finite time. Secondly, a novel nonsingular fast terminal sliding mode control (NFTSMC) law is designed to overcome singularity issues, reduce chattering, and guarantee finite-time convergence of the system states. Finally, the ANFTSMC scheme, integrating NFTSMC with VGESO, is proposed to achieve precise position tracking for the prosthetic hand. The convergence of the closed-loop control system is validated using Lyapunov’s stability theory. Experimental results demonstrate that the external pulse disturbance error of ANFTSMC is 8.19°, compared to 19.21° for the comparative method. Furthermore, the maximum absolute error for ANFTSMC is 0.63°, whereas the comparative method shows a maximum absolute error of 1.03°. These results underscore the superior performance of the proposed ANFTSMC algorithm.

Regression prediction of critical exhaust volumetric flow rate in tunnel fire with two-point extraction ventilation based on neural network method

In this study, a Genetic Algorithm-Backpropagation Neural Network (GA-BPNN) model was developed to predict critical exhaust volumetric flow rate in tunnel fires with two-point extraction ventilation system. Seven influencing factors served as inputs for the model, while the critical exhaust volumetric flow rate was the output. Experimental data from two reduced-scale tunnels were utilized to both train and validate the GA-BPNN model. The results demonstrate a satisfactory prediction performance, with key metrics such as Mean Absolute Percentage Error (MAPE), Relation Coefficient (R2), and Root Mean Square Error (RMSE) achieving values of 0.0384, 0.9989, and 3.5258, respectively. Importantly, the model maintains a maximum Absolute Percentage Error (APE) below 10 %. In contrast, the traditional BPNN model exhibited an APE of approximately 18.9 %. Moreover, the predictive results of the GA-BPNN model were compared with those of a previously semi-empirical formula. It was observed that the semi-empirical predictions exhibited APE exceeding 20 % for certain data points. This further underscores the superiority of the GA-BPNN model in predicting critical exhaust volumetric flow rates in tunnel fires with two-point ventilation system. These findings affirm the feasibility and effectiveness of GA-BPNN regression model as a reliable method for predicting critical exhaust volumetric flow rates under multiple factors.

A feasible numerical computation based on matrix operations and collocation points to solve linear system of partial differential equations

In this work, a system of linear partial differential equations with constant and variable coefficients via Cauchy conditions is handled by applying the numerical algorithm based on operational matrices and equally-spaced collocation points. To demonstrate the applicability and efficiency of the method, four illustrative examples are tested along with absolute error, maximum absolute error, RMS error, and CPU times. The approximate solutions are compared with the analytical solutions and other numerical results in literature. The obtained numerical results are scrutinized by means of tables and graphics. These comparisons show accuracy and productivity of our method for the linear systems of partial differential equations. Besides, an algorithm is described that summarizes the formulation of the presented method. This algorithm can be adapted to well-known computer programs.

Anti Wind‐Up and Robust Data‐Driven Model‐Free Adaptive Control for MIMO Nonlinear Discrete‐Time Systems

ABSTRACTThis article addresses a solution to one of the main challenges of online data‐driven control (DDC) methods: reducing the sensitivity of the model‐free adaptive control (MFAC) method to initial conditions and control parameters with the new control cost function and added the output error rate and integral along with a new anti‐wind up strategy for multi‐input multi‐output (MIMO) systems. The parameters introduced to the new control law have been validated using the boundary‐input boundary‐output (BIBO) approach to design and converge the controller. The simulation findings on a nonlinear auto‐regressive moving average model with exogenous inputs (NARMAX) system with triangular control input demonstrate that the proposed control rule will outperform to prototype MFAC. Furthermore, to analyze the sensitivity of the controller to the initial conditions and the uncertainties of the control parameters, 30 Monte Carlo simulations were performed with random initial conditions in the presence of disturbance in the control input, and output noise, and the results were compared with the prototype MFAC and conventional PID controller using standard criteria such as integral time absolute error, standard deviation, steady‐state error, and mean maximum error, which shows a noticeable superiority of proposed controller relative to the prototype MFAC.

Fault Diagnosis of Distributed Energy Distribution Network Based on PSO-BP

With the increasing scale of distribution network at distribution time, its complexity grows geometrically, and its fault diagnosis becomes more and more difficult. Aiming at the slow convergence and low accuracy of traditional backpropagation neural network in dealing with single-phase ground faults, the study proposes a backpropagation neural network based on improved particle swarm optimization. The model optimizes the weights and acceleration constants of the particle swarm algorithm by introducing dynamic coefficients to enhance its global and local optimization seeking ability. It is also applied in optimizing the parameters of backpropagation neural network and constructing the routing model and ranging model for fault diagnosis about distributed energy distribution network. The simulation results revealed that the maximum absolute error of the improved method is 0.08. While the maximum absolute errors of the traditional backpropagation neural network and the particle swarm optimized backpropagation neural network were 0.65 and 0.10, respectively. The fluctuation of the relative errors of the research method was small under different ranges of measurements. At 8.0 km, the minimum relative error was 0.39% and the maximum relative error was 2.81%. The results show that the improved method proposed in the study significantly improves the accuracy and stability of fault diagnosis and localization in distribution networks and is applicable to complex distribution network environments. The method has high training efficiency and fault detection capability and provides an effective tool for distribution network fault management.

Addressing multidimensional highly correlated data for forecasting in precision beekeeping

In recent years, there have been relevant advances in precision beekeeping. These advances are mainly focused on proposing sensor systems that collect crucial information for bee welfare, creating integrated architectures that allow beekeepers to monitor the current state of their hive through real-time data. However, there is a lack of predictive models that would allow beekeepers to anticipate specific events that endanger bee welfare and lead to a decline in productivity. Specifically, predictive approaches accounting for the high correlation among internal variables of beehives have not been developed to date. To address this research gap, multivariate predictive models, including auto-regressive state-space and time series models, have been implemented and applied to four different hives from the we4bee project. These models aim to predict the internal variables of beehives (four different temperatures, humidity, and weight) by utilizing the meteorological conditions to which the hives are exposed. A cross-validation adapted to time series data was employed for model generalization assessment. Prediction models based on vector time series exhibited superior performance in forecasting internal hive variables compared to multivariate auto-regressive state-space models. Overall, the approach based on the vector error correction model yielded the best balance between fit, prediction, and computational cost. The VEC-based approach produces predictions with maximum mean absolute errors of 177 (312)g in weight, 3.366 (3.802)% in humidity, and 1.122 (1.685)°C in temperature at 1 (3)-days ahead when dealing with beehives exhibiting a high degree of correlation in their internal variables. Moreover, the VEC-based approach requires less than a second to perform the time series fitting process, which makes it particularly interesting for application in big data environments. The integration of such models into a decision support system would meet the need of beekeepers to anticipate potential threats to the welfare of their bee colonies, streamlining their monitoring processes while eliminating the need for continuous inspections.

Unsteady-state modeling and analysis of a spray-assisted low-temperature desalination system

The spray-assisted low-temperature desalination (SLTD) technology driven by low-grade heat sources has demonstrated excellent thermodynamic and economic performances. However, existing studies are based on steady-state analyses and cannot describe dynamic system performance under fluctuations of the operating conditions. This work presents an unsteady-state analysis of the SLTD system driven by low-grade waste-heat sources. A dynamic model is developed and validated against experimental data. The model is observed to predict dynamic temperature profiles accurately with a maximum absolute error of ±0.3 °C. Then, the model is applied to analyze the responses of the SLTD system to different disturbances. Results show that the first evaporator is the most sensitive to the heat source temperature fluctuation, while locations farther away from the disturbance have longer response times and smaller fluctuations. When the heat source temperature for a 4-effect system fluctuates by 10 %, the response time and temperature fluctuation amplitude of the 1st effect are 579 s and 8 %, respectively, while those of the 4th effect are 1007 s and 3 %, respectively. Besides, the systems with smaller heat transfer areas, fewer numbers of effect, and higher heat source temperatures have faster response speeds. The results provide useful information for system operation and control strategy exploitation.

基于H无穷状态反馈控制的插秧机曲线路径跟踪控制器研究

The accuracy of curved path-tracking for headland turning of transplanters is crucial to maintaining the row spacing precision required for rice planting. To address this issue, a method based on H-infinity state feedback control is proposed. In this method, the requirement of robustness is transformed into linear matrix inequalities (LMIs) to optimize the gain coefficients of the control law. The simulation test show that this method outperforms the Linear Quadratic Regulator (LQR) when facing uncertain parameters (longitudinal speed and cornering stiffness) and path curvature disturbance. In addition, the field test results show that when the transplanter tracks a 1/4 circular arc path with a radius of 2 meters, the mean value of the absolute lateral error and the absolute heading angle error using this controller are 0.029 m and 3.69°, respectively. The maximum absolute lateral error is 0.072 m, and 64% of the absolute lateral error are less than 0.04 m, meeting practical requirements. Compared with the LQR controller with feed forward control, the mean value of the absolute lateral error is reduced by 36%. This method meets the accuracy and robustness requirements for unmanned rice transplanter turning at the headland.

An anti-swing control method combining deep learning prediction models with a multistate fractional-order terminal sliding mode controller for wave motion compensation devices

Shipborne multidimensional wave motion compensation devices eliminate the relative motion of cargo between supply and supplied ships. However, the random sway of ships and complicated disturbances still cause various cargo swings. Therefore, an anti-swing control method is proposed that combines deep learning prediction models with a multistate fractional-order terminal sliding mode controller for wave motion compensation devices. First, the constrained workspaces and anti-swing trajectories of cargo based on deep learning prediction models are constructed to decrease the swings of cargo originating from ship sway and inaccurate control parameters. Second, a multistate fractional-order terminal sliding mode controller is presented to eliminate the disturbance-caused cargo swings. Third, the stability of the multistate sliding mode controller is proven. Finally, the effectiveness of the anti-swing control method is verified through simulations and experiments, and the experimental results demonstrate that the maximum amplitude of cargo swings is only a few millimetres and that the maximum mean absolute error of the payload swings is less than four millimetres.

Battery state-of-health estimation incorporating model uncertainty based on Bayesian model averaging

Accurately estimating the state-of-health (SOH) of lithium-ion batteries is crucial for efficient, reliable, and safe use. However, the degradation of these batteries involves complex and intricate failure mechanisms that cannot be fully captured by a single model. To tackle this challenge, we propose an SOH estimation method based on Bayesian Model Averaging (BMA), which effectively accounts for both parameter and model uncertainties by combining estimations from different model implementations. It provides both point-value estimates and the probability distribution estimates, delivering results in a fraction of a second. Evaluated on three open-source datasets, the maximum absolute error of SOH estimation is within 0.03, and the Continuous Ranked Probability Score is within 0.015. Compared to optimal individual models, the proposed BMA method reduces the prediction error of point estimates by half to two-thirds. Additionally, the prediction error for probability estimation decreases by an order of magnitude. Moreover, the comparison studies with respect to the Gaussian Process Regression model and the Quantile Regression Forests model demonstrate the applicability and superiority of proposed method. These results highlight the potential of the BMA method to advance battery SOH estimation and facilitate reliable battery management.

Closed-loop state of charge estimation of Li-ion batteries based on deep learning and robust adaptive Kalman filter

The state of charge (SOC) is among the most crucial monitoring states in battery management systems. To accurately and robustly estimate the battery SOC, a closed-loop SOC estimation method based on deep neural networks is proposed in this study. Initially, a hybrid deep learning model, TGMA, is established by combining a temporal convolutional network (TCN) with a gated recurrent unit (GRU) and incorporating a multi-head self-attention mechanism (MHA) to estimate SOC. Subsequently, to further enhance the accuracy of estimating the battery SOC and the ability to resist outliers, the pre-estimated SOC from the deep learning model TGMA is used as the noisy observation in the Kalman filter, and robust factor and noise adaptive filtering are introduced into the Kalman filter, proposing a closed-loop estimation method based on deep learning and robust adaptive Kalman filter (RAKF). Finally, through comparison with other similar methods, the proposed TGMA-RAKF method is validated to have higher accuracy, generalization, and robustness against outliers under different operating conditions and temperatures, with root mean square error, mean absolute error, and maximum absolute error all less than 1 %, 1 %, and 2.5 % respectively.

Modeling and control strategy optimization of battery pack thermal management system considering aging and temperature inconsistency for fast charging

Maintaining the battery pack’s temperature in the desired range is crucial for fulfilling the thermal management requirements of a battery pack during fast charging. Furthermore, the temperature difference, temperature gradient, aging loss and energy consumption of the battery pack should be balanced to optimize its performance. This paper establishes the liquid cooling thermal management system model for an electric vehicle’s battery pack, which accurately characterizes the temperature distribution and electrical and aging characteristics of the battery pack. The discrepancies between the experimental and simulated results of battery’s fast charging-cooling suggest that maximum absolute temperature error is 2 °C, and the maximum relative voltage error is 1.5 %. Based on the non-dominated sorting genetic algorithm II algorithm, an optimization model with aging and temperature gradient of cells as objective functions is established. Three different thermal management strategies are created according to the optimization results. The comparison of these thermal management methods with the constant temperature cooling strategy demonstrates that the minimum temperature gradient strategy enhances the temperature consistency of the battery pack and reduces the maximum temperature gradient by 27 %. The minimum aging thermal management strategy depicts an aging loss of 0.091 %, reducing the charging time by 272 s and the maximum temperature gradient by 2.0 °C. The balanced thermal management strategy enables the battery pack to balance the temperature gradient and aging loss by optimizing the charging time, battery pack temperature difference, energy consumption and other indicators. The weight of each indicator is determined by its information entropy, which can be replaced according to the diverse needs to achieve different balanced thermal management strategies, providing a reference for the actual application of thermal management strategies.

Hybrid algorithm GA-PSO-BP in secure localization technology for wireless sensor networks

ABSTRACT Wireless sensors face problems such as poor positioning accuracy in their applications. A hybrid indoor space safety positioning technology is proposed for this purpose. This technology uses genetic models and particle swarm optimization models to optimize the parameters of neural networks. Considering the vulnerability of wireless sensors to Sybil attacks during the positioning process, communication encryption and location awareness are used for optimization. In the positioning experiment, the proposed method had maximum positioning errors of 0.23 m and 0.43 m on the first and second floors, respectively, which were superior to other methods. In the positioning security experiment, the positioning errors of the first and second floors after being attacked by Sybil were 4 m and 5 m respectively, while the proposed security positioning technology had an error range of 1 m. In terms of security comparison, when there are 31–35 attack samples, the proposed method has a maximum root mean square error and average absolute error of 0.72 m and 0.79 m, respectively, which is superior to other models. It can be seen that the proposed technology has excellent positioning effect and stability. The research content will provide important technical support for the practical deployment and security management of wireless sensor networks.

Aggregate uniformity of recycled asphalt mixtures with RAP from refined decomposition process: Comparison with routine crushing method

In consideration of skyrocketing price in pavement materials and ever-increasing attention to environment preservation, the utilization of reclaimed asphalt pavement (RAP) into asphalt mixtures has gained more and more favor. However, the uniformity of virgin and RAP aggregate in recycled asphalt mixtures (RAM) has not yet been studied due to the difficulty in informational extraction of multi-structures. To address this, white dolomite was used as virgin aggregate to be distinguished from RAP aggregate based on their color difference. The shear gyratory compactor specimens of RAM were prepared and saw-cut, and then meso-structures including virgin aggregate, RAP aggregate and asphalt mortar were extracted using digital image processing (DIP) technology. The DIP processes mainly involved retention of blue channel for image graying, newly proposed double-threshold OTSU method for image segmentation and image morphology processing. The aggregate uniformity index (AUI) was proposed based on the coefficient of variation of aggregate area ratio. The influences of RAP that were pre-processed by refined decomposition (RD) and routine crushing (RC) and its content on the aggregate uniformity were evaluated. The results demonstrate the efficacy of the improved DIP approach, as evidenced by the maximum mean absolute error between laboratory-measured and image-estimated gradation that is lower than 7.8 %. The addition of RAP into asphalt mixtures is adverse to the aggregate uniformity. The higher the RAP content, the inferior the aggregate uniformity of RAM, which attributes to the agglomeration of RAP materials. RAP agglomeration leads to the heterogeneity of RAP aggregate and thus dominates the heterogeneity of the whole aggregate. Therefore, the aggregate uniformity of RAM produced by RD-RAP is superior to that produced by RC-RAP. The superior degree is 9 %.

Fluidic Multichambered Actuator and Multiaxis Intrinsic Force Sensing.

Soft robots have morphological characteristics that make them preferred candidates, over their traditionally rigid counterparts, for executing physical interaction tasks with the environment. Therefore, equipping them with force sensing is essential for ensuring safety, enhancing their controllability, and adding autonomy. At the same time, it is necessary to preserve their inherent flexibility when integrating sensory units. Soft-fluidic actuators (SFAs) with hydraulic actuation address some of the challenges posed by the compressibility of pneumatic actuation while maintaining system compliance. This research further investigates the feasibility of utilizing the incompressible actuation fluid as the means of actuation and of multiaxial sensing. We have developed a hyperelastic model for the actuation pressure, acting as a baseline pressure. Any disparities from the baseline have been mapped to external forces, using the principle of pressure-based fluidic soft sensor. Computed tomography imaging has been used to examine inner deformation and validate the analytically derived actuation-pressure model. The induced stresses within the SFA are examined using COMSOL simulations, contributing to the development of a calibration algorithm, which accounts for geometric and cross-sectional nonlinearities and maps pressure variations with tip forces. Two force types (concentrated and distributed) acting on our SFA under different configurations are examined, using two experimental setups described as "Point Load" and "Distributed Force." The force sensing algorithm achieves high accuracy with a maximum absolute error of 0.32N for forces with a magnitude of up to 6N.

Target decomposition-led light-weighted offline training strategy-aided data-driven state-of-charge online estimation during constant current charging conditions over battery entire lifespan

This paper develops a novel target decomposition-led light-weighted offline training strategy-aided data-driven state-of-charge (SOC) online estimation method during constant current (CC) charging conditions over battery entire lifespan. Firstly, real SOC is conceptually decomposed into base SOC and SOC error. Subsequently, taking voltage and real SOC of initial CC charging cycle as input and output, machine learning algorithm is adopted to offline establish base SOC acquisition model without considering battery aging. Thirdly, the errors between real SOC and acquired base SOC are calculated, where the extremely similar distribution of SOC error against different battery degradation with two aging-dependent peaks can be clearly observed. Following this precious characteristic, a SOC error calculation model is further built only via several typical CC charging cycles with base SOC and battery capacity as input. Finally, the acquired base SOC is compensated by the computed SOC error for real SOC calculation. The validation results demonstrate that the proposed target decomposition-led method has overwhelming advantages in light-weighted offline training and accurate SOC online estimation during CC charging conditions, where maximum mean absolute error and maximum root mean squared error of SOC estimation results over the same type of batteries’ entire lifespan are only 0.98 % and 1.2 %, respectively.

Multidimensional ultrasonic vibration machining modeling of SiCp/Al cutting forces with different volume fractions: Experiments and numerical simulations

The cutting properties of the particles of aluminum-based silicon carbide (SiCp/Al) and the base material are so disparate that it is challenging to ensure the surface quality of machining. The cutting force significantly influences the quality of the machined surface. Therefore, real-time cutting force prediction is paramount for ensuring the workpiece's desired surface quality. This paper presents a 3D ultrasonic milling force model, established from four aspects: cutting formation force, extrusion friction force, SiC particle crushing force, and ultrasonic impact force. The model is based on the ultrasonic action of the cutting process and the unique removal characteristics of the material. A 3D ultrasonic milling system was constructed to experimentally verify the milling force model. The effects of each parameter on the milling force were analyzed through orthogonal experiments, which resulted in the effects of depth of cut, rotational speed, volume fraction, ultrasonic amplitude, and feed rate. The ultrasonic amplitude, feed rate, and other factors were found to influence the milling force to varying degrees. The milling force was observed to be 42.47 %, 22.37 %, 10.85 %, 12.67 %, and 11.64 % affected by the factors mentioned above, respectively. The single-factor experiment was conducted to analyze the cutting force at different machining parameters. The experimental results and the MATLAB numerical simulation results exhibited a similar trend of change. The maximum absolute error between the two was 12.71 %, with an average absolute error of 3.735 %.