Proposed Identification Method Research Articles

Weed Identification Using Deep Learning in Vegetable Farming

Weed identification in vegetable plantation is more challenging than crop weed identification due to their random plant spacing. So far, little work has been found on identifying weeds in vegetable plantation. Traditional methods of crop weed identification used to be mainly focused on identifying weed directly; however, there is a large variation in weed species. This paper proposes a new method in a contrary way, which combines deep learning and image processing technology. Firstly, a trained CenterNet model was used to detect vegetables and draw bounding boxes around them. Afterwards, the remaining green objects falling out of bounding boxes were considered as weeds. In this way, the models focus on identifying only the vegetables and thus avoid handling various weed species. Furthermore, this strategy can largely reduce the size of training image dataset as well as the complexity of weed detection, thereby enhancing the weed identification performance and accuracy. To extract weeds from the background, a colour index-based segmentation was performed utilizing image processing. The employed colour index was determined and evaluated through Genetic Algorithms (GAs) according to Bayesian classification error. During the field test, the trained CenterNet model achieved a precision of 95.6%, a recall of 95.0%, and a F1 score of 0.953, respectively. These experiment results demonstrate the feasibility of using the proposed method for the ground-based weed identification in vegetable plantation.

Defect identification method for capacitive-inductive dual-mode sensors with lift-off interference

The capacitive-inductive dual-mode probe provides a defect detection solution that allows simultaneous or sequential eddy current detection and capacitive imaging detection, known as capacitive-inductive dual-mode detection. The dual-mode detection combines the advantages of the two detection techniques and can be used to detect multiple types of defects in “insulator-conductor” multilayer hybrid structures. However, its high sensitivity poses a challenge as the probe's responses a susceptible to interference from lift-off variations, leading to non-relevant indications in the results that is not indicative of defects or anomalies. In this study, the Lift-off Characteristics (LoC) of the hybrid structures are studied, and an effective method for identifying relevant indications such as defect and anomaly for dual-mode sensor response signals with lift-off interference by combining the testing data of both detection modes and the LoC curve is proposed. A series of validation experiments were conducted to demonstrate the effectiveness of the proposed method in defect identification and classification. The results confirmed that the method effectively identified defect from lift-off interference and significantly reduced the occurrence of false judgments, thereby enhancing the reliability of the dual mode detection. Furthermore, the feasibility of utilizing the method to identify relevant indication when the sensor is tilted was also demonstrated.

Design and analysis of shadow detection and removal scheme in real-time still images

This research’s main objective is to study and evaluate the detection and removal of undesired shadows from still images since these shadows might mask important information caused by light sources and other obstructions. A variety of methods for detecting and eliminating shadows as well as object tracking approaches based on movement estimation and identification are investigated. This includes shadow removal methods like background subtraction, which are intended to improve obstacle recognition of the source item and increase the accuracy of shadow removal from objects. When new items enter the frame, they are first distinguished from the background using a reference frame. The tracking procedure is made more difficult by the merging of the shadow with the foreground object. The approach highlights the difficulties in object detection owing to frequent occurrences of obstacles by using morphological procedures for shadow identification and removal. The proposed approach uses feature extraction is also discussed, highlighting its importance in image processing research and the use of suggested methods to get over obstacles in image sequences. The proposed method for shadow identification and removal offers a novel approach to improve image processing when dealing with still images. The purpose of this technique is to better detect and remove shadows from images, which will increase the precision of object tracking and detection. Depending on the type of images being processed, the process begins with initializing a background model, which is based on a static image background.

Physics-Based Aircraft Dynamics Identification Using Genetic Algorithms

This research introduces a physics-based identification technique utilizing genetic algorithms. The primary objective is to derive a parametric matrix, denoted as A, describing the time-invariant linear model governing the longitudinal dynamics of an aircraft. This is achieved by proposing a fitness function based on the properties of the transition matrix and taking advantage of some of the capabilities of the genetic algorithm, mainly those of restricting the search ranges of the unknowns. In this case, such unknowns are related to the type of aircraft and flight conditions that are considered during the identification process. The proposed identification method is validated with a reliable nonlinear model that can be found in the literature, as well as with the calculation of the trim condition and linearization generally used in aircraft dynamics. In summary, this study suggests that the genetic algorithm provided with the adequate fitness function could be an appealing alternative for aircraft model identification, even when limited data are available. Furthermore, in some cases, linearization using a genetic algorithm can be more efficient than classical methods.

A workload identification method of industrial robot combining dynamic model and convolutional neural network

The motion joint system of industrial robot has obvious nonlinear characteristics, high dimensional running data and limited number of experimental samples. If mature neural network algorithm is used for identifying the workload of robots, it is easy to have over-learning problems, which seriously restricts the generalization ability of load identification method. In this paper, a load identification method combining the dynamic model of industrial robots with the neural network data model is proposed. A UR5 robot is used for carrying out multiple dynamic load testings to verify the effectiveness of the proposed identification method. Firstly, the workload identification by CNN algorithm is given, and the influence of parameters of prediction model and different neutral network are analyzed. Then the classical dynamic model of industrial robots with multi-degrees of freedom is established. The identified workload by dynamic model is also analyzed. At last, the deterministic information such as velocity and displacement is extracted from the calculation results of the dynamic model as the initial anchoring value. Then convolutional neural network is applied for compensating the residual highly nonlinear information. An improved mixing combination method is also proposed. This method can effectively deal with the interaction of different types of information in the data, and preliminarily cooperate the dynamic model and convolutional neural network. This provides a basic framework and method for solving the problem of parameter identification in multi-degree-of-freedom systems with small samples.

Experiment-free physical hybrid neural network approach for battery pack inconsistency estimation

With the increasing scale of battery systems, the impact of battery inconsistency due to aging on battery pack performance becomes increasingly significant. To achieve high-precision battery pack modeling, we propose an in-situ estimation method for battery inconsistency parameters. The proposed method utilizes current and voltage data recorded by the battery management system (BMS). The respective terminal voltage errors are used as the loss function, and the equivalent circuit model and the fully connected neural network are combined to realize the estimation of inconsistency parameters. Through optimization algorithm, the inconsistency parameters of all cells in the battery pack are simultaneously estimated. The estimation results of full capacity, high-end capacity, and internal resistance exhibit high accuracy with root-mean-square error (RMSE) values of 0.82% (0.393 Ah), 0.70% (0.336 Ah), and 3.34% (0.097 mΩ), respectively. The battery pack model constructed using the estimation results maintains high accuracy across various operating conditions, demonstrating the effectiveness of the proposed method for inconsistency estimation and battery pack modeling.

The role of time and space in the identification of left behind regions: a case study of Denmark

Abstract This paper provides an overview of the state of identification of left behind regions and proposes a new methodology for identifying them. Despite increased interest in rising regional inequality, the identification of left behind regions differs greatly, complicating policymaking aimed at addressing these regional challenges. Based on the substantial literature that has emerged in recent years on left behind regions, the paper argues for the importance of both time and space as key elements when identifying left behind regions. This paper applies the proposed identification method to Denmark; it uses longitudinal data from 1980 to 2018 and includes the current wave of interregional inequality and typically adopted variables for identifying left behind regions across the literature.

An intelligent identification method based on self-adaptive mechanism regulated neural network for chemical process

BackgroundThe complex characteristics of chemical process, such as multivariable, nonlinear, time-varying and strong coupling often lead to the poor effect of traditional identification theory in practical application. The development of deep learning in recent years has brought a breakthrough for nonlinear system identification, but more progress is still needed. MethodsThis paper proposes a nonlinear identification method based on self-adaptive mechanism regulated Long Short-Term Memory (LSTM) network for chemical process dynamic simulation. First, to increase the reliability of the application of neural network (NN) for chemical process identification and improve the generalization ability, the known differential equation describing the mechanism is taken as a regularization term to constrain the output of the NN. Then, a specific training method is proposed, which introduces trainable self-adaptive weights to force the neural network to focus on the regions with large training error. In addition, a semi-supervised network training method is proposed for the case that some parameters in the mechanism equation are unknown. Finally, a dynamic virtual device (VD) model is established, which can simulate the dynamic response of controlled objects. Significant findingsTo evaluate the efficiency of the developed identification method, various comparative experiments are conducted on pH neutralization and continuous stirred tank reactor (CSTR) processes. The experimental results show that the proposed identification method can obtain a nonlinear dynamic model with robustness, high accuracy and strong generalization ability.

Identifying the switching topology of dynamical networks based on adaptive synchronization.

This paper proposes an approach for identifying unknown switching topology in a complex dynamical network. The setup is divided into two components: a primary drive network and a specialized response network equipped with switched topology observers. Each class of observers is dedicated to tracking a specific topology structure. The updating law for these observers is dynamically adjusted based on the operational status of the corresponding topology in the drive network-active if engaged and dormant if not. The sufficient conditions for successful identification are obtained by employing adaptive synchronization control and the Lyapunov function method. In particular, this paper abandons the generally used assumption of linear independence and adopts an easily verifiable condition for accurate identification. The result shows that the proposed identification method is applicable for any finite switching periods. By employing the chaotic Lü system and the Lorenz system as the local dynamics of the networks, numerical examples demonstrate the effectiveness of the proposed topology identification method.

Rice Identification Under Complex Surface Conditions with CNN and Integrated Remote Sensing Spectral-Temporal-Spatial Features

Accurate and effective rice identification has great significance for the sustainable development of agricultural management and food security. This paper proposes an accurate rice identification method that can solve the confused problem between fragmented rice fields and the surroundings in complex surface areas. The spectral, temporal, and spatial features extracted from the created Sentinel-2 time series were integrated and collaboratively displayed in the form of visual images, and a convolutional neural network model embedded with integrated information was established to further mine the key information that distinguishes rice from other types. The results showed that the overall accuracy, precision, recall, and F1-score of the proposed method for rice identification reached 99.4%, 99.5%, 99.5%, and 99.5%, respectively, achieving a better performance than the support vector machine classifier. Therefore, the proposed method can effectively reduce the confusion between rice and other types and accurately extract rice distribution information under complex surface conditions.

High-Fidelity Model Identification for Synchronous Reluctance Motor Drives

This paper presents an accurate model identification method for the synchronous reluctance machine (SynRel), considering the nonlinear magnetic behavior and spatial harmonics. The availability of the high-fidelity magnetic model is essential for high-performance control and modeling the vibro-acoustic behavior of the electric drive, allowing to predict not only the average torque but also the torque ripple behavior versus rotor position. The magnetic coenergy variation is calculated based on the flux linkages in order to yield the torque ripple. During the identification, the current disturbances introduced by the spatial effects and the inverter nonlinearities are properly suppressed by the merged proportional-integral (PI) controller with a repetitive technique. Besides, the inverter nonlinearity and digital control effects are carefully addressed to improve identification accuracy. The effectiveness of the proposed identification method is demonstrated by the comparison of the finite element (FE) simulation results and experiments for two SynRel prototypes.

Strain/displacement field reconstruction and load identification of bracket arm structure of high-speed maglev train based on superposition conversion algorithm

The bracket arm structure, subjected to long-term vertical levitation and transversal guidance forces, is a key load-bearing component of high-speed maglev trains. It is crucial to monitor strain, displacement, and load in real time to evaluate structural health. However, it is almost impossible to measure directly at the service site of trains due to the intricacy of the bracket arm structure and the constrained installation area. Hence, it is an effective and feasible approach to reconstruct the strain/displacement field using a limited amount of strain information from FBG sensors. In order to reconstruct the strain/displacement field and identify the load on complicated structures, a load–strain/displacement superposition conversion algorithm is proposed in this paper. The modal coordinate and the mode-shape matrix are swapped out for the load weight and the load–strain/displacement matrix, respectively. Thus, it gets around the difficulty of choosing the mode order in traditional modal theory. In addition, the algorithm can achieve full-field reconstruction of both strain and displacement by measuring only part of the nodal strain response, reducing the redundancy of the traditional algorithm. Finally, numerical simulations and experimental tests were carried out to verify the feasibility of the proposed method for field reconstruction and load identification.

Fire identification based on improved multi feature fusion of YCbCr and regional growth

Fire is one of the mutable hazards that damage properties and destroy forests. However, fire-like objects easily influence many existing image-based fire detection methods. In order to improve the performance of fire identification, this paper proposes a new fire identification algorithm by merging fire segmentation and multi feature fusion of fire. First, the improved YCbCr models in the reflection and non-reflection environment are constructed according to the color model. Simultaneously, the reflection and non-reflection conditions can be judged according to the segmented area. Second, the seed points are determined according to the weighted average of centroid of each connected region. Simultaneously, the fine segmentation of fire image is implemented according to the entropy of average contrast and uniformity within the connected region. Experiments show that the segmentation is not affected by image noises. Finally, the quantitative indicators of fire identification are given according to the coefficient of variation of area, the dispersion of centroid and the circularity. Cases show that the proposed identification method of fire not only accurately identifies fire, but also has a lower computation complexity than the deep learning method.

Bridge Frequency Identification Using Cross-Power Spectra of Vehicle Vibrations from Multiple Moving Speeds

In recent years, a rapid bridge health monitoring technology has been developed using an instrumental moving vehicle. Using recorded vehicle vibration data, bridge frequencies are identified for bridge health monitoring or finite element model updating. Target bridge frequencies with significant amplitudes in the vehicle’s vibration frequency spectra are expected to be found. However, in the coupled vehicle–bridge interaction (VBI) system, bridge vibration-relevant vehicle dynamics might not be noticeable. The bridge frequency would be difficult to identify because of the potential influence of road roughness. To resolve this difficulty, a novel bridge frequency identification method is proposed to mitigate the negative effects of road roughness. First, theoretical derivations are done to ascertain the VBI system dynamic characteristics considering road surface roughness. Our findings showed that the road roughness-relevant vehicle dynamics are closely related with the traveling speed, whereas the bridge frequency remains approximately constant. Theoretical investigations indicated that cross-power spectra between vehicle dynamics at multiple moving speeds are effective to mitigate the negative effects of road roughness. Presumably, it is feasible to identify the target bridge frequency from the cross-power spectra. Both the dynamic characteristics of the VBI system and the effectiveness of the proposed method for bridge frequency identification were examined using finite element simulations and laboratory experiments. Compared to existing methods, the proposed method is widely applicable to real-world situations and difficulties.

Parameter identification of photovoltaic converter models

The work aims to develop a simple and effective method for identifying the parameters of photovoltaic converter (PV-cell) models for their wide practical application. An exponential model having one diode and five parameters is adopted as the basic model of the photovoltaic converter to facilitate its easy integration into MatLab/Simulink simulation software. To identify the parameters of models of photovoltaic converters, an original method based on finding the minimum of the root-mean-square error between model and experimental volt-ampere characteristics was used. The effectiveness of the method was confirmed by determining the parameters of different models of photovoltaic modules based on various technologies: thin-film Shell ST40, monocrystalline Shell SP70, and polycrystalline Kyocera KC200GT. A developed technique for constructing mathematical models of photovoltaic converters in order to reproduce their electrical characteristics is presented. The solution is easily implemented in Excel spreadsheet software with the “Search for Solution” add-on installed. The proposed universal method for identifying the parameters of photoconverter models can be used to build models of a wide range of photovoltaic modules and solar cells. The model current-voltage characteristics of the considered photovoltaic modules are shown to be in good agreement with experimental characteristics across a wide range of temperature and insolation conditions corresponding to their actual operation. Higher accuracy is achieved when using the proposed identification method to determine the electrical parameters of mathematical models of photovoltaic converters than when using other well-known analytical and numerical methods. The result is a simple and effective method for constructing mathematical models of photovoltaic converters, which does not require the use of programming or the development of specialized numerical algorithms, allowing it to be used to solve many technical problems related to the design and operation of photovoltaic systems.

An industrial case study on the combined identification and offset-free control of a chemical process

For three decades, model predictive control (MPC) has been the flagship advanced control method in the chemical process industries. However, most implementations still use heuristic methods for designing MPC estimators, especially for offset-free MPC implementations. In this paper, we present a recently developed maximum likelihood-based method for the identification of linear augmented disturbance models for use in offset-free MPC. This method provides noise covariances that are used to derive Kalman filters and moving horizon estimators, forgoing the need for manual design and tuning of the estimator. The method is extended to handle closed-loop plant data. The proposed identification method and estimator design are evaluated in industrial-scale, real-world case study of a process at Eastman Chemical’s Kingsport plant. Using this identified model, we reduced the mean stage cost by 38% compared to the performance of the existing, hand-tuned MPC model.

Ship Diesel Engine Fault Diagnosis Using Data Science and Machine Learning

One of the most important elements of the reliability structure of a motor vessel is its power subsystem, with the most crucial component being the engine. An engine failure excludes the ship from operation or significantly limits its operation. Therefore, accurate fault diagnosis should be a crucial issue for modern maintenance strategies. In mechanical engineering, the vibration and acoustic signals recorded during the operation of the device are the most meaningful data used to identify the reliability state. In this paper, a novel system-oriented method of reliability state identification is proposed. The method consists of the analysis of the vibration and noise signals collected on each of the engine cylinders using supervised machine learning. The main novelty of this method is data augmentation application and SVM classifier implementation. Due to these aspects, the method becomes robust in the case of poor-quality data or a limited and incomplete learning dataset. The quality of the proposed identification method was evaluated by addressing a new industrial issue (Sulzer 6AL20/24 marine engine reliability state identification). During the tests, the efficiency of the method was analyzed in the case of a complete learning data set (all types of inability states were presented in the learning data set) and an incomplete learning data set (in the testing data set, there were new types of inability states). As a result, in both cases, a very high (100%) identification accuracy of the reliability state and the type of inability state was obtained. This is a significant increase in accuracy (4.6% for the complete and 22% for the incomplete learning data set) in comparison to the efficiency of the same method without the use of machine learning and data science.

Mutual Inductance and Load Identification of LCC-S IPT System Considering Equivalent Inductance of Rectifier Load

The variation of mutual inductance and load parameters will affect the transmission power and efficiency of the inductive power transfer (IPT) system. The identification of mutual inductance and load parameters is an essential part of establishing a stable and reliable IPT system. This paper presents a joint identification method of load and mutual inductance for the LCC-S IPT system, which does not require the establishment of primary and secondary communication and related control. Firstly, the resistance-inductance characteristics of the equivalent load of the rectifier are analyzed by simulation, and then the rectifier and system load are equivalent to the circuit model of resistance and inductance in series. Secondly, the characteristics of the reflected impedance are analyzed, and the functional relationship between the transmitter impedance and the rectifier impedance is established by using the ratio of the real part to the imaginary part of the reflected impedance, which realizes the decoupling of the load and the mutual inductance. Thirdly, the functional relationship between the equivalent impedance of the rectifier and the load resistance of the system is obtained by data fitting. Then, the equations of the above two functional relationships are combined. By measuring the voltage of the parallel compensation capacitor at the transmitting side, the current of the transmitting coil and the phase difference between the two, the battery load can be solved first, and then the mutual inductance can be calculated, so that the high-precision identification of the load and mutual inductance can be realized. Finally, an experimental platform of the LCC-S IPT system is built for experimental verification. The experimental results show that the maximum identification errors of mutual inductance and load are 5.20% and 5.53%, respectively, which proves that the proposed identification method can achieve high precision identification.

Bayesian damage identification of an unsymmetrical frame structure with an improved PSO algorithm

With the development of artificial intelligence technology, intelligent algorithms are widely utilized in the field of probabilistic model updating to solve the optimization problems. In this study, the opposition-based learning based on population centroid and boundary rebound strategy was incorporated into PSO to improve the optimization performance on structural damage detection, and the damage identification of an unsymmetrical frame based on Bayesian model updating with the improved PSO algorithm is proposed for the first time. Six benchmark functions were firstly employed to test the optimization performance of the improved PSO algorithm, which overcomes the premature convergence of the original PSO algorithm in which trapped particles often fall into local optimum. Then three damage scenarios of a laboratory unsymmetrical frame were utilized to validate the effectiveness of the proposed method in structural damage identification. The encouraging experimental results exhibit that the proposed method not only can successfully identify the damage locations and extents, but the associated uncertainties of the identified results can be quantified by calculating the posterior probability density functions of the uncertain model parameters. In addition, the effects of the quantities of measured information on the posterior uncertainties of the model parameters were investigated, and the uncertainty analysis results reveal that the posterior uncertainties associated with the identified model parameters are sensitive to the amount of measurement used in model updating. The proposed method is expected to help in damage identification and condition assessment of the practical engineering structures.

A Hybrid Genetic Algorithm-Based Parameter Identification Method for Nonlinear Hysteretic System with Experimental Verification

The Bouc–Wen nonlinear hysteretic model is widely used in nonlinear systems, and it is difficult to identify the parameters of Bouc–Wen model because of its high nonlinearity. In the paper, a hybrid genetic algorithm combined with sequential quadratic programming is proposed to solve the problem about the parameter identification of Bouc–Wen model. Sequential quadratic programming method has good local convergence and superlinear convergence rate. Therefore, compared with the genetic algorithm, the proposed identification method can accelerate the convergence speed in parameter recognition of Bouc–Wen model and achieve excellent recognition accuracy. Through the function test and the numerical simulation of the shear structure, it is proved that the proposed method has high convergence speed and accurate identification. In addition, the theory and method proposed in this paper are combined with Bouc–Wen model to identify parameters of overall restoring force model of the energy-dissipation structures, the identification results are in good agreement with the experimental results, verifying reliability and accuracy of the model and robustness of the proposed identification method.