Maximum Error Value Research Articles

Evaporating temperature estimation of refrigeration systems based on vibration data-driven soft sensors

The evaluation of the operating conditions of refrigeration compressors once installed in household appliances is challenging due to the need to install pressure transducers, a process which requires system evacuation and refrigerant reintroduction. In addition, changes in the piping modify the characteristics of the original product. This paper proposes a soft-sensing technique based on vibration measurements of the compressor surface to predict the evaporating temperature. Different machine learning (ML) techniques are evaluated as data-driven prediction models, namely multilayer perceptron (MLP) neural networks, least squares boosting, generalized additive model, random forest, extreme learning machine, and random vector functional link neural networks. These techniques were applied to data obtained from a test rig designed to emulate compressor operation in a refrigeration system, with an operating envelope from -30°C to -10°C for the evaporating temperature and from 34°C to 54°C for the condensing temperature. The results showed that, with a single vibration measurement point, it was possible to use an MLP technique to estimate the evaporating temperature with a root mean squared error of 1.74°C in a non-intrusive way. For the other prediction techniques, the errors were a bit higher than for the MLP, but the maximum error value was about 2.5°C in all cases.

Comparative assessment of FSM and MPSIAC models in quantifying soil sedimentation rates in the semi-humid and alpine regions of northern Iran.

Soil erosion and sediment yield is a global problem that increasingly contributes to soil degradation. Although erosion analysis requires the availability of erosion and sedimentation data, the lack of sediment monitoring stations and the resulting limitations in collecting sediment measurements have necessitated the use of experimental models in many areas. The present study aimed to compare Factorial Scoring Model (FSM) and Modified Pacific South-West Inter-Agency Committee (MPSIAC) model for estimating erosion in the Mazdaran Basin (Firoozkuh, Iran). For this purpose, the required maps were prepared for both models, and the sediment rate was estimated using the two models to compare their efficiency using the corresponding maximum error (ME) and coefficient of determination (R2) values. The results showed that considering sediment based on the FSM model, the studied catchment consisted of regions with a high and very high sediment yield, while the MPSIAC model identified regions with low, medium, and high sediment yield. With an R2 value of 0.62 and an ME value of 2.24, the MPSIAC model provided more accurate estimates of the sediment yield in the studied area. Using the MPSIAC model, sediment yield was 6687.86 tons per year or the equivalent of 2.64 tons/ha per year.

Intelligent optimization prediction of screw conveyor speed for earth pressure balance shield machine based on complete ensemble empirical mode decomposition of adaptive noise and deep learning

The excavation stratum of the earth pressure balance (EPB) shield machine is complex and changeable. For ensuring the excavated interface stabilization, the rotational speed of screw conveyor must be precisely controlled to maintain the balance of earth pressure in sealed cabin and to avoid accidents such as ground collapse or upliftment. A data-driven intelligent optimal model is presented by the paper, which organically integrates the complete ensemble empirical mode decomposition of adaptive noise (CEEMDAN), sparrow search algorithm (SSA) and long short-term memory (LSTM) network to achieve accurate prediction of screw conveyor rotation speed. Firstly, Pearson correlation analysis method is used to analyze the historical data of shield tunneling, and the tunneling parameters strongly related to the change of screw conveyor rotation speed are obtained as model input. Secondly, the CEEMDAN is used to decompose the obtained parameter data set into several modal components, and then the denoised data is reorganized to capture the changing characteristics of the original data and reduce the complexity of the data. Then, the parameters such as the learning rate of the LSTM network are optimized by using the strong global search ability of SSA to enhance the prediction precision further in the model. Finally, based on decomposed and reorganized data, SSA-LSTM is used to establish a rotation speed intelligent prediction model realize precise controlling of screw conveyor rotation speed. The simulation experiment is carried out by using the construction data of a section of Beijing Metro Line 10, and the results of three evaluation indexes of the model are given: Mean Absolute Error (MAE) is 0.02878, Mean Absolute Percentage Error (MAPE) is 0.00882 and R2 is 0.99909, which shows that the model has good prediction performance. The control effect of the method on the earth pressure balance of the sealed cabin is tested, and the maximum error value is 0.25%, which verifies the engineering applicability of the method.

Dynamic formulation and inertia fast estimation of a 5-DOF hybrid robot

As the main driving mechanism of a hybrid robot, the parallel mechanism is a nonlinear time-varying system. The load inertia of its actuated joints changes with the configuration of the robot. Analyzing and fitting the inertia variation is of great significance to the design and control of hybrid robots. By taking a hybrid robot named TriMule as an example, the variation of load inertia of each actuated joint in the whole workspace is first revealed based on the dynamic analyses of the robot. Then two methods based on the circular and elliptical membership are proposed to calculate fitted inertia over the whole workspace using inertia information at a few configurations. Finally, the fitting methods of the two membership functions are compared and discussed. The results show that the maximum value and global mean value of the fitting error of the elliptical membership method are 39.18% (51.23%) and 65.79% (81.25%) for actuated joint-1 (joint-2 and joint-3) lower than those of circular membership method, which promise a better global fitting accuracy. The proposed method can be used to estimate the joint load inertia or other control variables affected by inertia in a quick manner, allowing the algorithm to be easily integrated into the robot control system.

A comparative study on state-of-charge estimation for lithium-rich manganese-based battery based on Bayesian filtering and machine learning methods

In this paper, a comparative study on the state of charge (SOC) estimation of the lithium-rich manganese-based battery (LRMB) has been conducted by systematically considering the equivalent circuit model (ECM), aging state and algorithms. The results show that the first-order RC model combined with the Extended Kalman filter (EKF) is more suitable for the SOC estimation of the LRMB when the battery decays less severely. The maximum value of root mean square error of SOC estimation error no exceeds 1.6 % under various dynamic operating conditions, which can well meet the practical application. Besides, the first-order model based EKF algorithm occupies least computing resource and simultaneously possesses the best convergence velocity against the initial SOC value deviation. When the LRMB suffers a severe aging, the SOC estimation performance decreases significantly with the max error exceeding 6 %, to overcome this, a new algorithm combined deep learning and Extend Kalman filter (EKF) is proposed, which exhibit high SOC estimation accuracy with the RMSE less than 0.16 %. In summary, a systematic study about the SOC estimation of LRMB has been carried out, which can provide great guidance for the future engineering application of LRMB.

Evaluation of an Active Disturbance Rejection Controller for Ophthalmic Robots with Piezo-Driven Injector.

Retinal vein cannulation involves puncturing an occluded vessel on the micron scale. Even single millinewton force can cause permanent damage. An ophthalmic robot with a piezo-driven injector is precise enough to perform this delicate procedure, but the uncertain viscoelastic characteristics of the vessel make it difficult to achieve the desired contact force without harming the retina. The paper utilizes a viscoelastic contact model to explain the mechanical characteristics of retinal blood vessels to address this issue. The uncertainty in the viscoelastic properties is considered an internal disturbance of the contact model, and an active disturbance rejection controller is then proposed to precisely control the contact force. The experimental results show that this method can precisely adjust the contact force at the millinewton level even when the viscoelastic parameters vary significantly (up to 403.8%). The root mean square (RMS) and maximum value of steady-state error are 0.32 mN and 0.41 mN. The response time is below 2.51 s with no obvious overshoot.

Application of Fashion Element Trend Prediction Model Integrating AM and EfficientNet-b7 Models in Art Design

The poor application effect of traditional sports training methods and the difficulty of recording data due to the time and space constraints of sports make it difficult for trainers to improve their learning outcomes. Based on this, the study proposes to apply human posture recognition in sports teaching design, and use VGGNet-19 as a feature extractor and OpenCV open source software to capture posture movements, and introduce the concepts of joint angle and movement similarity to design a sports assessment system for physical education based on the geometric spatial feature variability analysis of posture based on limb angle information. The testing outcomes demonstrate that the study’s improved gesture recognition algorithm has a recognition rate of more than 90% on gesture movements, and the maximum recognition error value (0.010) is smaller than that of the dynamic time-regularised gesture algorithm (0.014) and the convolutional neural network algorithm (0.017). The assessment system is also better able to improve students’ professional performance and satisfaction, with its average professional score and satisfaction reaching 86 and 92%, which is significantly better than other comparative algorithms. The method is effective in providing trainers with data-based training scenarios and helping them to improve their learning in sport.

Effect of Anisotropy on Residual Stress Measurement of 316L Stainless Steel by Ultrasonic Surface Wave

Fe–Cr alloys are widely used in power, petroleum, and manufacturing industries for their good resistance to corrosion and oxidation at high temperatures. Ultrasound is the only nondestructive method so far to measure the residual stress of in-service components. However, for parts with material anisotropy, such as materials processed by rolling, the measurement accuracy is highly restrained. In this paper, a rolled 316L stainless steel sample is used to study the influence of texture on the measurement of residual stress by ultrasonic surface wave. The experimental results show that the propagation velocity of surface waves in the sample has anisotropic characteristics. The wave velocity parallel to the rolling direction (0°) is the maximum, and the wave velocity perpendicular to the rolling direction (90°) is the minimum, thereby affecting the measurement accuracy. It is found that reducing the frequency of surface waves can reduce the influence of anisotropy. Therefore, a low-frequency method and modified formula are used to improve the measurement accuracy. The maximum error in the rolling direction is reduced from 21.3 to 3.6 MPa, and the maximum relative error is also reduced from 45.4 to 9.0%. The modified formula can further reduce the influence of anisotropy, with the maximum error value further reduced to 2.3 MPa, the maximum relative error reduced to 4.9%, and the surface wave detection accuracy effectively improved.

Enhancing azimuth estimation in low-resolution underwater acoustic lens systems for advanced communication and positioning

Here we aimed to enhance azimuth estimation to accurately locate a communication partner’s position by utilizing a low-resolution underwater acoustic lens system for communication. We proposed a method to accurately estimate the direction-of-arrival (DoA) by finding the center of gravity of the focal distribution using two adjacent elements of the transducer array. Ranging was to be based on time-of-flight (ToF). The method’s effectiveness was evaluated by two-dimensional finite difference time domain simulation. The lens’s focused sound field was calculated and the array was designed based on the result. DoA and ToF were estimated from the received signals and evaluated by comparing them with the true values under conditions involving 1-, 2-, and 3- users. The results confirmed DoA estimation with an error of almost 4°, even when using a lens system with an azimuth resolution of 10°. As the number of users increased, the maximum error value increased, though slightly.

Analytical Approach for Forecasting the Load Capacity of the EN AW-7075-T6 Aluminum Alloy Joints Created Using RFSSW Technology.

To ensure the high reliability of aircraft structures, the Refill Friction Stir Spot Welding (RFSSW) process must be characterized by a high load capacity of the welds and a small standard deviation of the load capacity spread. This allows us to obtain uniform functional properties in the connections, ensuring the high quality of the process. This work aims to select the most favorable technological parameters for the welding process of EN AW-7075-T6 Alclad aluminum alloy sheets, which are used for the production of aircraft structures. The best networks were calculated using the Statistica 13.3 program. The obtained results were compared with the results of previous investigations. It has been shown that a model using neural networks allows for the determination of connection parameters with much greater accuracy than the classical model. The maximum error in estimating the load capacity of the connection for the mathematical model was 6.13%, and the standard deviation was 14.51%. In the case of neural networks, the maximum error value did not exceed 1.55%, and the standard deviation was 3.74%. It was shown that, based on the neural model, it is possible to determine the process parameters that ensure the required quality capacity of the process, ensuring a probability of obtaining the required load capacity of the connections amounting to P = 0.999935 with a defect rate of 0.0065%. This possibility is not provided by the classical model due to its large error in estimating the process spread and the high sensitivity of the process input parameters to the output parameters.

Comparative analysis of equivalent circuit battery models for electric vehicle battery management systems

Lithium-ion batteries need to be controlled by a Battery Management System (BMS) to operate safely and efficiently. BMS controls parameters, such as current, voltage, temperature, state of charge (SoC),state of health (SoH), state of power (SoP) and etc. The battery models and several prediction algorithms that the BMS uses to carry out these checks are essential to the system's performance. Therefore, the battery model is crucial to the BMS. This model is used to optimize the performance, capacity, lifetime and safety of the battery. Using the accurate battery model for BMS and electric vehicles can improve energy efficiency, extend battery life and reduce safety risks. Therefore, it is important that the model can accurately reflect the battery behavior under different load conditions. In this study, the performance of Rint, Partnership for a New Generation of Vehicles (PNGV), Thevenin, and Dual Polarization (DP) battery models, which are widely known in the literature, to simulate static and dynamic voltage behavior is compared. A 18650 NMC battery was used for this purpose, and Hybrid Pulse Power Characterization (HPPC), Dynamic Stress Test (DST), Worldwide Harmonised Light Vehicle Test Procedure (WLTP), and Constant Current (CC) discharge tests were performed. The performance of the models for the four tests is compared. The maximum error values for WLTP are 2.98 % in Rint, 1.32 % in PNGV, 2.80 % in Thevenin, and 1.09 % in DP. Comparing the performances of models for all tests, it is found that the DP model is the most accurate model under both constant and dynamic current conditions.

Construction of environmental vibration prediction model for subway transportation based on machine learning algorithm and database technology

Vibrations generated in the metro transport environment are mainly caused by, vibrations generated by the interaction between the metro and the track during operation. and the change of vibration factors will affect the normal operation of the subway. However, it is difficult to have a model that can achieve the characteristics of high accuracy, fast computing speed and wide range of use in the traditional metro rail transportation environment prediction. Therefore, this research uses database theory and machine learning algorithms to predict the vibration of subway transportation environment. The experimental results show that the average difference between the whole prediction value and the real value is 1.4 dB, of which the maximum difference error value is 0.29%, the maximum error difference is 8.2%, and the approximate value is 6.2 dB, and the four averages predicted in 40 m are relatively small as 1.6 dB, and the average error value of prediction ability between 40 and 100 m is 1.72 dB, and the experimental prediction value and real value are in good agreement. The agreement between the experimental prediction and the real value is very good. Therefore, the model is able to predict the vibration model of the subway transportation environment with a high degree of agreement and accuracy.

A novel multi-factor fuzzy membership function - adaptive extended Kalman filter algorithm for the state of charge and energy joint estimation of electric-vehicle lithium-ion batteries

In view of the unmeasurable state parameters of electric-vehicle lithium-ion batteries, this paper investigates a novel multi-factor fuzzy membership function - adaptive extended Kalman filter (MFMF-AEKF) algorithm for the online joint estimation of the state of charge and energy. Strong nonlinear characteristics of model parameters are characterized by considering multiple processing factors of electrochemical and diffusion effects for lithium-ion batteries and constructing an optimized multifactor coupling model. In the proposed MFMF-AEKF method, multi-space-scale factors are introduced to realize the numerical analysis of the multi-factor coupled model parameters and state estimation under dynamic working conditions of electric-vehicle lithium-ion batteries. The proposed MFMF-AEKF algorithm estimates the state of charge (SOC) with the overall best mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE), and maximum error (ME) values of 1.822 %, 4.322 %, 1.947 %, and 2.954 %, respectively, under challenging working conditions. And The MAE, MAPE, RMSE, and ME values for the state of energy (SOE) are 0.617 %, 1.711 %, 0.695 %, and 1.011 %, respectively. Both state estimation results are better than the traditional method. The proposed MFMF-AEKF algorithm has higher estimation accuracy which provides a feasible estimation algorithm for the joint SOC and SOE of lithium-ion batteries.

Depth-based vessel position fixing by means of a neural network

A depth-based vessel position fixing method on the basis of a neural network is proposed. The network takes as an input a sequence of depth values measured by an echo-sounder and predicts vessel latitude and longitude for the moment of the latest depth measurement. The neural network has a fully-connected feedforward architecture with several layers which satisfies conditions of the universal approximation in compliance with the Stone-Weierstrass theorem. The Adamax algorithm for the neural network training with controlling a maximum value of position error at each epoch is implemented. Modeling is conducted with the Python programming language and the Tensorflow library. The model surface of seabed is performed as a second-order polynomial. Training samples on the basis of virtual soundings at the coordinate net knots with the space resolution not worse than one cable are obtained. After samples obtaining the training of the neural network is conducted. A validation set is not used. Several neural networks are trained. They have different number of hidden layers and different number of neurons per each hidden layer. After training the test procedure is performed. Test samples are generated in the assumption that a vessel is moving along meridians which are not used at the stage of the preliminary soundings survey. The cases of mean and random test meridians are considered. The random meridians are obtained with a uniform random number generator. As the result, all the tested neural networks have shown approximately identical navigational accuracy which is close to the accuracy for the training set.

Goniometric accuracy and compensation methods for full-space laser positioning systems

The accurate large-scale positioning system is a novel full-space laser positioning system that has the advantages of automatic and multi-node parallel measurement and sub-millimeter positioning accuracy, and it has become an important component in the digital manufacturing of large equipment. Goniometric accuracy is particularly important for the evaluation of system function and accuracy improvement. An improved goniometric accuracy calibration method is proposed to assess the goniometric accuracy by using multiple measurement points to fit the mapping relationship between the goniometric measurement of the system and the goniometric measurement of the multi-tooth indexing table. The feasibility of the accuracy assessment method is verified by establishing a geometric model and performing modeling and simulation analysis. The method is verified using an all-space laser positioning system, and the dynamic performance of the base station is optimized to improve the overall goniometric accuracy. The maximum value of the goniometric error is 0.008″ and the goniometric accuracy is 0.026″ after fitting the curve by the mapping relation, demonstrating high stability, and the fluctuation of the goniometric error after optimization is improved from the original ±15″ to ±7.5″, verifying the effectiveness of the goniometric compensation method. This method solves the problem of coaxiality calibration in goniometric evaluation and provides the basis for the traceability of the goniometric accuracy.

A Transient-State Lumped Parameter Thermal Model for Brushless Wound Field Switched Flux Machines

In this paper, the transient-state lumped parameter thermal model (LPTM) of the wound field switched flux (WFSF) machine is characterized for the first time. The modelling methodology for the LPTM is given in details, including the process of determining the parameters of the LPTM, based on which two LPTMs are compared with the experiments, i.e. the LPTM-I and LPTM-II without and with consideration of the practical winding position in the stator slot. It is found that the practical positions of the DC and AC coils in the slots is important to improve the accuracy of the LPTM, due to the existence of slot empty space caused by overlapped DC and AC windings. By considering the practical positions of the DC and AC coils in the slots, the maximum value of the absolute error between the LPTM model predicted temperature and the measured values of the AC winding are 3.4°C, 5.23°C, 3.18°C and 5.22°C under DC test, steady state test, cycling test-I and cycling test-II, respectively, whilst they are 3.95°C, 2.00°C, 2.28°C and 4.89°C for the DC winding temperature error. The average value of the absolute error between the LPTM model predicted temperature and the measured values of the AC winding are 0.88°C, 1.25°C, 0.85°C and 1.87°C under DC test, steady state test, cycling test-I and cycling test-II, respectively, whilst they are 1.08°C, 1.47°C, 1.06°C and 2.73°C for the DC winding temperature. The temperature of the WFSF machine during a 9-hour experiment can be predicted in 27.8s by the LPTM, which has a more than 1100 times faster calculation speed.

Validation of OpenMC Code Criticality Value Calculation for GFR Reactor with UN-PuN Fuel

The use of OpenMC code needs to be validated with other code, so that an accurate and valid criticality value is generated. The validation value is affected by the minimum number of particles used. Determination of the minimum number of particles is carried out by varying the number of particles so that they have a convergent value, which is in the range of 100-50000 particles. The results of determining the minimum number of particles are then used to find the minimum number of cycles consisting of active and inactive batches. This study uses the criticality value parameter in the form of an effective multiplication factor (k-eff) as a benchmark for code accuracy. The k-eff values generated by OpenMC and SRAC are then compared and the validation error value is searched. The error value is found by calculating the k-eff (Δk-eff) difference between the two codes. OpenMC code can be said to be validated if it has an error value of less than 1% against the calculation results of the SRAC code. The calculation results of determining the minimum number of particles show k-eff values in the range of 35000 - 50000 minimal changes significantly or not fluctuating. The calculation of k-eff and entropy in cycle 500 shows a convergent value at the active batch value of >30. Based on the data cycle, researchers used 100 active batches and 30 inactive batches to perform validation calculations. The results of the validation calculation using UN-PuN fuel with a plutonium material composition of 10% showed a maximum k-eff error value of 0.819%. The Δk-eff obtained is 0.008.

Attitude Algorithm of Gyroscope-Free Strapdown Inertial Navigation System Using Kalman Filter.

A gyroscope-free strapdown inertial navigation system (GFSINS) solves the carrier attitude through the reasonable spatial combination of accelerometers, with a particular focus on the precision of angular velocity calculation. This paper conducts an analysis of a twelve-accelerometer configuration scheme and proposes an angular velocity fusion algorithm based on the Kalman filter. To address the sign misjudgment issue that may arise when calculating angular velocity using the extraction algorithm, a sliding window correction method is introduced to enhance the accuracy of angular velocity calculation. Additionally, the data from the integral algorithm and the data from the improved extraction algorithm are fused using Kalman filtering to obtain the optimal estimation of angular velocity. Simulation results demonstrate that this algorithm significantly reduces the maximum value and standard deviation of angular velocity error by one order of magnitude compared to existing algorithms. Experimental results indicate that the algorithm's calculated angle exhibits an average difference of less than 0.5° compared to the angle measured by the laser tracker. This level of accuracy meets the requirements for attitude measurement in the laser scanning projection system.

Kinetic regularities of filtration drying of corn alcohol distillery stillage

We studied the kinetic regularities of filtration drying of corn alcohol distillery stillage. The kinetic curves characterizing the process of filtration drying of corn alcohol distillery stillage at different parameters of the stationary layer of the studied material and thermal agent were obtained: at different heights of the wet material, and at different temperatures of the thermal agent and the velocity of its movement through the stationary layer. The calculated dependences for determining the intensity of the filtration drying process in two conditional drying periods were proposed. The equations for the calculation of drying time separately for each conditional drying period were given. The maximum value of the relative error is 19.42%, while the average relative error is 5.37%, which is an acceptable deviation in practical calculations of experimental and industrial drying installations.

Design of Sustainable Construction Cost Estimation System Based on Grey Theory — BP Model

China’s construction industry is developing rapidly, and market investment is out of control. Therefore, accurate and rapid project cost estimation is the focus of the industry. The grey theory system is a method to study the uncertainty problem of little data and poor information. It can excavate different information hidden in the system observation data to achieve correct description and understanding. The research will improve the BP neural network based on grey theory, and verify the exactitude and advantage of the model from performance comparison and empirical analysis. This study can make the project cost forecast more reasonable. Results show that compared with the traditional model, the maximum error value decreased by 0.52, the research model’s maximum mean square error decreased by 0.09, and the variance decreased by 0.13. In the regression analysis, the [Formula: see text]-value of the model fitting was above 0.95, the [Formula: see text]-value of the verification fitting reached 0.9205. The [Formula: see text]-value of the overall fitting reached 0.9475, and the fitting effect was excellent. The convergence speed of the research model is faster than other models, and the fitness value is stable at 1.36. In the empirical analysis, this model can review project cost investment budgets in some areas and reduce project costs. This can improve work efficiency. The model also has certain practical significance for the sustainable development of construction project cost estimates.