Model Identification Method Research Articles

Nearly Zero-Energy Building Load Forecasts through the Competition of Four Machine Learning Techniques

Heating, ventilation and air conditioning (HVAC) systems account for approximately 50% of the total energy consumption in buildings. Advanced control and optimal operation, seen as key technologies in reducing the energy consumption of HVAC systems, indispensably rely on an accurate prediction of the building’s heating/cooling load. Therefore, the goal of this research is to develop a model capable of making such accurate predictions. To streamline the process, this study employs sensitivity and correlation analysis for feature selection, thereby eliminating redundant parameters, and addressing distortion problems caused by multicollinearity among input parameters. Four model identification methods including multivariate polynomial regression (MPR), support vector regression (SVR), multilayer perceptron (MLP), and extreme gradient boosting (XGBoost) are implemented in parallel to extract value from diverse building datasets. These models are trained and selected autonomously based on statistical performance criteria. The prediction models were deployed in a nearly zero-energy office building, and the impacts of feature selection, training set size, and real-world uncertainty factors were analyzed and compared. The results showed that feature selection considerably improved prediction accuracy while reducing model dimensionality. The research also recognized that prediction accuracy during model deployment can be influenced significantly by factors like personnel mobility during holidays and weather forecast uncertainties. Additionally, for nearly zero-energy buildings, the thermal inertia of the building itself can considerably impact prediction accuracy in certain scenarios.

Optimal Fully Actuated System Approach-Based Trajectory Tracking Control for Robot Manipulators.

In this article, a trajectory tracking control strategy is proposed for robot manipulators via a fully actuated system (FAS) approach, which has shown its simplicity and flexibility for most of the nonlinear controller design. However, the motion control for robot manipulators is more complicated since unknown dynamical model, external disturbances, friction forces, and various physical constraints are required to be considered. Therefore, the FAS approach cannot be straightforwardly applied. To address these challenges, the dynamic model of robot manipulators is established via model identification methods. Furthermore, based on the identified model, an FAS composite control strategy with simple structure is designed, which is achieved by integrating a high-order disturbance observer (HODO) in the inner loop, with an FAS trajectory tracking controller in the outer loop. Specifically, the HODO is utilized for handling the uncertain dynamics and external disturbances. Moreover, the controller gains are optimized using a gradient-based optimal parameter tuning method (OPTM). By imposing joint angle constraints, joint angular velocity constraints, and input torque limits into the formulation, the OPTM also ensures the satisfaction of these physical constraints. Numerical simulations and experiments are provided to validate the performance of the proposed controller.

Analytical fractional reduced-order model identification method for processes with overdamped and underdamped response

In this paper, an analytical procedure to identify a fractional-first order plus dead time (FFOPDT) model is presented. This procedure is applied to processes with overdamped or underdamped step response, i.e., fractional order values in the range 0.5 ≤ α ≤ 2.0 can be estimated. The required process information is obtained through a simple open-loop step experiment and is based on fitting three arbitrary points (x1-x2-x3%) on the process reaction curve. Some numerical examples are proposed to show the effectiveness and applicability of this procedure for the identification of fractional-order models. The main feature of this approach is that it estimates the FFOPDT model of overdamped and underdamped processes.

Dynamic modelling and parametric analysis of a bolted joint rotor-bearing system considering stick-slip behavior at mating interface

Bolted joint structure is widely utilized in the aeroengine compress rotor systems with convenience in assembly and maintenance. Due to the inherent structure discontinuity and stick-slip behavior at the mating interface, stiffness softening and frictional energy dissipation would be generated at the mating interface, resulting in complex vibration features of the rotor system. In this work, to describe the stick-slip behavior at the mating interface of the rotor system, an Iwan-based model is introduced to describe the contact state of asperities at the multi-scale level. The corresponding model identification method is proposed based on the partial hysteresis curve of the bolted joint. After that, a dynamic model of the bolted joint rotor system is established, and its dynamic response is obtained using the Runge–Kutta method. The effect of preload and friction coefficient on the nonlinear vibration of the bolted joint rotor is evaluated in detail based on the dynamic model. The result indicates that the stiffness softening would lead to the critical speed decreasing and aggravating the vibration response of the rotor system, while the frictional energy dissipation would restrain the response amplitude and induce the self-excited vibration within the super-critical speed range. The higher preload or friction coefficient has little effect on the motion state of the rotor system within the sub-critical speed range and makes system vibration stable between the 1st- and 2nd-order critical speed while unstable at the super-critical speed range. This research helps understand the vibration features of bolted joint rotor systems under the condition of stick-slip behaviors at the mating interface.

Neural Network-Based Hammerstein Model Identification of a Lab-Scale Batch Reactor.

This paper focuses on two types of neural network-based Hammerstein model identification methods for the acrylamide polymerization reaction of a batch reactor process. The first neural-based identification type formulates the weights of the multilayer network directly as parameters of the nonlinear static and linear dynamic blocks of the Hammerstein model and trains the weights using a gradient-based backpropagation algorithm. In the second identification type, the nonlinear static block of the Hammerstein model is framed as a single hidden-layer feedforward network and both nonlinear and linear block parameters are trained using an extreme learning machine, where the training procedure is exempted from gradient calculation. The primary focus of the paper is neural-based model identification of a complex nonlinear system, which facilitates ease of linear/nonlinear controller design with good learning speed and less computations. A future work toward the machine learning-based nonlinear model predictive controller implementation using the Jetson Orin Nano board is also described.

Building thermal dynamics modeling with deep transfer learning using a large residential smart thermostat dataset

Understanding thermal dynamics and obtaining the computational model of residential buildings enable its scaled application in energy retrofits, control optimization and decarbonization. In this paper, we present a deep learning approach to model building thermal dynamics with smart thermostat data collected from residential buildings, with the goal to investigate model generalizability. In the first stage, we developed and compared different Deep Learning architectures including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) models and CNN-LSTM to predict indoor air temperature in a multi-step time horizon. In the second stage, we implemented a Transfer Learning (TL) process, which aims to improve the prediction performance on a new set of buildings (targets), exploiting the knowledge of related or similar buildings (sources). Different TL strategies and source model identification methods were investigated. The study showed that the CNN-LSTM performed the best among the architectures compared, with an average Mean Absolute Error (MAE) of 0.26 °C for one-hour-ahead (twelve 5-min future steps) predictions. Furthermore, the results showed that freezing the LSTM layer and fine-tuning the other layers of the CNN-LSTM achieved the best performance among four TL strategies, which further improved the performance with respect to a machine learning approach by 10%, and proving the effectiveness and generalizability of the proposed approach. A comparison of three different source model identification methods showed that randomly selecting source models constrained by similar building characteristics can provide good TL performance while retaining simplicity comparing with other quantitative source identification methods.

An improved dynamic model identification method for small unmanned helicopter

PurposeThe purpose of this paper is to introduce an improved system identification method for small unmanned helicopters combining adaptive ant colony optimization algorithm and Levy’s method and to solve the problem of low model prediction accuracy caused by low-frequency domain curve fitting in the small unmanned helicopter frequency domain parameter identification method.Design/methodology/approachThis method uses the Levy method to obtain the initial parameters of the fitting model, uses the global optimization characteristics of the adaptive ant colony algorithm and the advantages of avoiding the “premature” phenomenon to optimize the initial parameters and finally obtains a small unmanned helicopter through computational optimization Kinetic models under lateral channel and longitudinal channel.FindingsThe algorithm is verified by flight test data. The verification results show that the established dynamic model has high identification accuracy and can accurately reflect the dynamic characteristics of small unmanned helicopter flight.Originality/valueThis paper presents a novel and improved frequency domain identification method for small unmanned helicopters. Compared with the conventional method, this method improves the identification accuracy and reduces the identification error.

An Accurate Dynamic Model Identification Method of an Industrial Robot Based on Double-Encoder Compensation

Aiming at the challenges to accurately simulate complex friction models, link dynamics, and part uncertainty for high-precision robot-based manufacturing considering mechanical deformation and resonance, this study proposes a high-precision dynamic identification method with a double encoder. Considering the influence of the dynamic model of the manipulator on its control accuracy, a three-iterative global parameter identification method based on the least square method and GMM (Gaussian Mixture Model) under the optimized excitation trajectory is proposed. Firstly, a bidirectional friction model is constructed to avoid using residual torque to reduce the identification accuracy. Secondly, the condition number of the block regression matrix is used as the optimization objective. Finally, the joint torque is theoretically identified with the weighted least squares method. A nonlinear model distinguishing between high and low speeds was established to fit the nonlinear friction of the robot. By converting the position and velocity of the motor-side encoder to the linkage side using the deceleration ratio, the deformation quantity could be calculated based on the discrepancy between theoretical and actual values. The GMM algorithm is used to compensate the uncertainty torque that was caused by model inaccuracy. The effectiveness of the proposed method is verified by a simulation and experiment on a 6-DoF industrial robot. Results prove that the proposed method can enhance the online torque estimation performance by up to 20%.

How to Make the Stress Relaxation Experiment for Polymers More Informative.

Different viscoelastic models and characteristics are commonly used to describe, analyze, compare and improve the mechanical properties of polymers. A time-dependent linear relaxation modulus next to frequency-domain storage and loss moduli are the basic rheological material functions of polymers. The exponential Maxwell model and the exponential stretched Kohlrausch-Williams-Watts model are, probably, the most known linear rheological models of polymers. There are different identification methods for such models, some of which are dedicated to specific models, while others are general in nature. However, the identification result, i.e., the best model, always depends on the specific experimental data on the basis of which it was determined. When the rheological stress relaxation test is performed, the data are composed of the sampling instants used in the test and on the measurements of the relaxation modulus of the real material. To build a relaxation modulus model that does not depend on sampling instants is a fundamental concern. The problem of weighted least-squares approximation of the real relaxation modulus is discussed when only the noise-corrupted time-measurements of the relaxation modulus are accessible for identification. A wide class of models, that are continuous, differentiable and Lipschitz with respect to parameters, is considered for the relaxation modulus approximation. The main results concern the models that are selected asymptotically as the number of measurements tends to infinity. It is shown that even when the true relaxation modulus description is completely unknown, the approximate optimal model parameters can be derived from the measurement data that are obtained for sampling instants that are selected randomly due to the appropriate randomization introduced whenever certain conditions regarding the adopted class of models are satisfied. It is shown that the most commonly used stress relaxation models, the Maxwell and Kohlrausch-Williams-Watts models, satisfy these conditions. Since the practical problems of the identification of relaxation modulus models are usually ill posed, Tikhonov regularization is applied to guarantee the stability of the regularized solutions. The approximate optimal model is a strongly consistent estimate of the regularized model that is optimal in the sense of the deterministic integral weighted square error. An identification algorithm leading to the best regularized model is presented. The stochastic-type convergence analysis is conducted for noise-corrupted relaxation modulus measurements, and the exponential convergence rate is proved. Numerical studies for different models of the relaxation modulus used in the polymer rheology are presented for the material described by a bimodal Gauss-like relaxation spectrum. Numerical studies have shown that if appropriate randomization is introduced in the selection of sampling instants, then optimal regularized models of the relaxation modulus being asymptotically independent of these time instants can be recovered from the stress relaxation experiment data. The robustness of the identification algorithm to measurement noises was demonstrated both by analytical and numerical analyses.

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.

An empirical parameter identification method considering hysteresis effects for LiFePO4 battery's electrochemical model

Compared with other types of lithium-ion batteries, there is an obvious hysteresis effect in LiFePO4 battery, which leads to inaccurate parameter identification for its mechanical modelling. This article proposes an empirical parameter identification method for the P2D model of LiFePO4 battery, which takes the hysteresis effect into consideration. The proposed method adds a modified zero-state model to describe hysteresis effects during the identification process of lithium intercalation intervals. Simulation results prove that errors of identified parameters under various working conditions are relatively small, and the average relative error is within 0.5 %. The proposed method to compensate for hysteresis effect is relatively simple and easy to implement, and can improve the accuracy and generalization ability of the electrochemical model of LiFePO4 batteries.

Identification of reaction wheel assembly disturbances based on variable modal frequencies

Micro-vibrations of the reaction wheel assembly (RWA) can have a negative impact on spacecrafts’ attitude control, surveillance, and measurement. Evaluating the micro-vibration magnitude during the design stage is possible by establishing an accurate dynamic model of the RWA disturbance characteristics and identifying the relevant parameters. Doing so can provide references for spacecraft design and development, ultimately leading to more efficient and effective spacecraft operations. However, the conventional methods of modeling and identifying the RWA and of predicting the distribution and magnitude of the disturbing force/torque are limited, as it relies on stable and constant modal frequencies. In the development mission of a satellite, a specific scenario is found that the modal frequency of the RWA changes with speed. This study proposes a novel modeling and parameter identification method for the RWA disturbances based on variable modal frequencies and is advantageous in this scenario. The RWA’s dynamic model is considered a covariant system related to the rotational speed. Additionally, frequency-domain and time-domain models have been established for the RWA’s translational and swinging motions. Simultaneously, the identification method of the model parameters is developed, which involves identifying the harmonic factors, the variable modal frequencies, and the amplitude coefficients. Finally, the method is validated by numerical simulations and experiments. The results indicate that the proposed method offers excellent identification performance, surpassing conventional methods in model accuracy and identification performance. This achievement holds significant reference value in the micro-vibration suppression and optimization design of spacecraft.

Data-driven modelling and dynamic analysis of the multistable energy harvester with non-Gaussian Lévy noise

In engineering, due to the complex structural characteristics of system and the non-Gaussian properties of random excitation, it is difficult to establish an accurate stochastic dynamic model for the strongly nonlinear multistable vibration energy harvester (VEH), especially for these driven by non-Gaussian Lévy noise. From the view of machine learning, a data-driven model identification method is devised to extract the non-Gaussian governing laws of the multistable VEH with the aid of the observed sample trajectory data. Based on the Nonlocal Kramers-Moyal formulas, the Lévy, drift and diffusion terms can be approximately expressed by the sample trajectories of the system. By implementing the least square method and the stepwise sparse regressor algorithm, the optimal drift and diffusion coefficients can be identified, and then the non-Gaussian stochastic differential equation of VEH is extracted. Two examples are utilized to verify the feasibility and effectiveness of the data-driven modelling method in VEH, which indicates that the identified results agree well with the original system. Finally, the stochastic dynamic behaviors induced by non-Gaussian Lévy noise are explored based on the data-driven penta-stable VEH. The proposed method can provide the theoretical guidance for the modelling and dynamics research of VEH in engineering.

Modeling of Eyld2000-2d Anisotropic Yield Criterion Considering Strength Differential Effect and Analysis of Optimal Calibration Strategy.

Sheet metals usually experience various loading paths such as uniaxial tension, uniaxial compression, biaxial tension, and simple shear during the forming process. However, the existing constitutive models cannot always accurately describe blanks' anisotropic yield and plastic flow behavior of blanks under all typical stress states. Given this, this paper improves the Eyld2000-2d yield criterion by introducing hydrostatic pressure to the A-Eyld2000-2d yield criterion that can describe the strength differential effect of materials. Meanwhile, to control the curvature of the yield surface more effectively, the near-plane strain yield stresses were added in the parameter identification process to calibrate the exponent m, so that the exponent is no longer considered as a constant value. Taking the widely used AA6016-T4, AA5754-O, DP980, and QP980 blanks in the automotive stamping industry as an example, the effectiveness of the new model and different parameter identification methods was verified by predicting experimental data under various simple and complex loading paths. Subsequently, the new model employing the optimal parameter identification strategy was compared with four widely used asymmetric yield criteria under associated and non-associated flow rules, including CPB06, LHY2013, S-Y2004, and Hu & Yoon2021, to further verify the accuracy of the proposed constitutive model. The results indicate that parameter identification strategy with variable exponent can significantly improve the flexibility of the yield criterion in describing the plastic anisotropy of blanks. Compared to the other yield criteria examined in this work, the new model provides the best prediction accuracy for the yield stresses and plastic flows of all blanks, especially in the near-plane strain and simple shear stress states. Modeling under the concept of anisotropic hardening can more accurately capture the evolving plastic behavior of blanks than isotropic hardening.

Classification of Lithium-Ion Batteries Based on Impedance Spectrum Features and an Improved K-Means Algorithm

This article presents a classification method that utilizes impedance spectrum features and an enhanced K-means algorithm for Lithium-ion batteries. Additionally, a parameter identification method for the fractional order model is proposed, which is based on the flow direction algorithm (FDA). In order to reduce the dimensionality of battery features, the Pearson correlation coefficient is employed to analyze the correlation between impedance spectrum features. The battery classification is carried out using the improved K-means algorithm, which incorporates the optimization of the initial clustering center using the grey wolf optimization (GWO) algorithm. The experimental results demonstrate the effectiveness of this method in accurately classifying batteries and its high level of accuracy and robustness. Consequently, this method can be relied upon to provide robust support for battery performance evaluation and fault diagnosis.

Multi-Scale Recursive Identification of Urban Functional Areas Based on Multi-Source Data

The study of urban functional area identification is of great significance for urban function cognition, spatial planning, and economic development. In the identification of urban functional areas, most studies considered only a single data source and a single division scale, the research results have problems such as low update frequency or incomplete information in a single data set, and overfitting or underfitting in a single spatial resolution. Aiming at the above problems, this paper proposes a multi-scale recursive recognition method based on interactive validation for urban functional areas using taxi trajectory data and point of interest (POI) data as the main data sources. First, the dynamic time warping (DTW) algorithm generates a time series similarity matrix, a CA-RFM model combining the clustering algorithm and random forest model is constructed. The model extracts significant feature regions as inputs through a K-medoid clustering algorithm, which are imported into the random forest model for urban functional zone (UFZ) identification. Then, to overcome the shortcomings of a single scale in expressing urban structural characteristics, a recursive model of different levels of urban road networks is established to classify multi-scale functional areas. Finally, cross-validation using the CA-RFM model and POI quantitative identification method obtains the final identification results of urban functional areas. This paper selects Shenzhen as the study area, the results show that the combination of clustering algorithm and random forest model greatly reduces the error of manual selection of training samples. In addition, the study demonstrates the superiority of the proposed method in two aspects, namely, faster delineation and improved accuracy in urban functional area identification.

A new camera model identification method based on color correction features

A new camera model identification method based on color correction features

Development of a method for the scenario-based identification of calculation models (EMS)

Abstract. A key goal of final disposal is the safe confinement of the radioactive waste to protect people and the environment. In order to evaluate the safe confinement, inter alia the consideration and analysis of expected and deviating developments of the disposal system are required. The derivation and identification of potential safety-relevant developments is the task of scenario development. With regard to the safety assessment, the scenarios (developments) are to be treated within the framework of numerical model calculations. It is essential that the calculation cases cover the scenarios and comprehensively consider the essential developments. The focus of the EMS (Entwicklung einer Methode zur szenariengestützten Identifizierung von Berechnungsmodellen) project (Beuth et al., 2021) is the development and testing of a method for the scenario-based identification of calculation models and the associated calculation cases. The main aspect of the method is an intensive, iterative exchange between the disciplines of scenario development and modeling, in which calculation cases are gradually optimized or adapted with a view to representativeness up to the required degree of scenario coverage. The testing and further developing of the procedure shall help to identify difficulties in the development of scenario-based calculation cases. Transparency, traceability and possible gaps in justification for modeling decisions are also important issues. Up to now, only little research has been done on the question of when a scenario is covered by calculation cases. Experience from safety analyses, e.g., the preliminary safety analysis for Gorleben (Fischer-Appelt et al., 2013), confirm that the derivation of calculation cases from scenarios is difficult. Therefore, one goal of this study is to assist the development of the derivation from calculation cases based on scenario development. The outcome of the EMS project is intended to support the Federal Office for the Safety of Nuclear Waste Management (BASE) within the framework of supervision and approval in the evaluation process of whether scenarios have been sufficiently covered by calculation cases.

Regression-based identification and order reduction method for nonlinear dynamic structural models

In this paper, we propose a regression-based nonlinear reduced-order model for nonlinear structural dynamics problems, called the Nonlinear Identification and Dimension-Order Reduction (NLIDOR) algorithm. We evaluate the algorithm using a simple toy model, a chain of coupled oscillators and an actual three-dimensional flat plate. The results show that NLIDOR can accurately identify the natural frequencies and modes of the system and capture the nonlinear dynamical features, while the linear Dynamic Mode Decomposition (DMD) method can only capture linear features and is influenced by nonlinear terms. Compared with the full-order model (FOM), NLIDOR can effectively reduce computational cost, while compared with DMD, NLIDOR significantly improves computational accuracy. The results demonstrate the effectiveness and potential of NLIDOR for solving nonlinear dynamic problems in various applications.

Machine-Learning Assisted Identification of Battery Life Models

Predictive battery life models are commonly utilized to extrapolate degradation trends observed during accelerated aging tests for simulation of degradation in real-world applications. Thus, fitting accelerated aging data as accurately as possible and with low uncertainty is crucial for making believable projections of battery lifetime, but it is challenging to identify algebraic expressions that accurately fit multivariate degradation trends. A review of models published in literature reveal some common expressions for fitting calendar aging data, which is only dependent on temperature and state-of-charge, but no consistency across many models for fitting cycle aging data, indicating the need for a statistically rigorous data driven approach for developing empirical models.This talk will describe a machine-learning assisted method for identification of predictive battery life models utilizing bilevel optimization and symbolic regression. Bilevel optimization with cross-validation is used to statistically determine cell- and stress-dependent model parameters, while symbolic regression identifies both linear and multiplicative candidate expressions to predict stress-dependent degradation rates by selecting low-order subsets of features from a generated feature library. Because model expressions are identified empirically, it is crucial to ensure resulting models behave according to physical expectations, so the stability of models for interpolation or extrapolation is interrogated qualitatively through simulation and quantitatively through cross-validation and uncertainty quantification via bootstrap resampling. This model identification approach substantially improves upon models identified purely using expert judgement in terms of both accuracy and uncertainty.Model simulation and validation is then conducted by deriving a state-equation form of the predictive model, enabling simulation of battery aging under dynamic stresses. This enables validation of the predictive battery model on lab-based tests with varying conditions or on drive-cycle or application-cycle testing protocols. Parameter uncertainty can be carried forward into model simulation, giving lifetime estimates and confidence windows for cell- or system-level lifetime. The financial impact of battery model uncertainty can be estimated by incorporating uncertainty into a technoeconomic model.