Local Model Networks Research Articles

Experimental parameter identification of scalable white and grey box jounce bumper models for ride comfort simulation

The design of a passenger car’s ride and handling goals requires an enormous amount of hardware and prototypes. To avoid time-consuming testing activities, one crucial trend is the virtualisation of suspension development. Another goal of the virtualisation is the determination of vehicle properties at an early stage of the development process. Highly accurate suspension models are required to transfer the full vehicle’s characteristics into simulation tools. Since the jounce bumper is an important component of the suspension system, modelling its functional behaviour is essential for virtualised chassis development. Therefore, two dynamic modelling approaches are derived and compared in this contribution. One approach is based on local linear model networks and thus can be seen as a grey box model. The other approach consists of white box rheological elements, which allow the physical interpretability of the model parameters. The comparison of the approaches is based on experimental parameter identification, which includes synthetic and real excitations recorded on representative roads. The advantages of the proposed approaches will be discussed in terms of their suitability for the virtual design of the driving behaviour.

Latin hypercubes for constrained design of experiments for data-driven models

Abstract The quality of data used for data-driven modeling affects the model performance significantly. Thus, design of experiments (DoE) is an important part during model development. The design space is constrained in many applications. In this work, the constrained case is investigated. An Latin hypercube based approach is applied and analyzed for strongly constrained design spaces. Contrary to commonly used optimization techniques, an incremental procedure is proposed. In every step, new data are added to the design. Each new point is selected by a distance-based criterion. The performance of the created designs is evaluated by the quality of the trained models. For different constraints, artificial data sets are created with a function generator. The performance of local model networks and Gaussian process regression models trained with those designs is evaluated and compared to models trained on data sets based on Sobol’ sequences.

System modeling and temperature control for a fuel cell system based on local model networks

ABSTRACT Fuel cells have been studied for use in stationary power generation and vehicle propulsion systems. The fuel cell thermal management subsystem is coupled and nonlinear, posing challenges for modeling and temperature control. This paper aims to integrate the physical models of the fuel cell stack, pump, thermostat, and other components combined with intelligent algorithms into an efficient system-level thermal management model framework and develop a model predictive controller to solve the temperature control problem. First, a physics-based nonlinear model of the fuel cell system is developed and used as a basis to identify the linearized model for different operating points. Then, the global model is obtained by fusing the local models with Gaussian validity functions using the local linear model tree method. Third, a multi-step prediction model is derived based on the local model networks, and a parameterized linear state space form is obtained and used for controller design. Furthermore, an online correction method is developed to reduce the model discrepancy. Finally, the accuracy of the system model and the performance of the proposed controller are verified by open-loop experimental data and a series of closed-loop simulation cases.

Retracted: Human Behavior Recognition in Outdoor Sports Based on the Local Error Model and Convolutional Neural Network.

[This retracts the article DOI: 10.1155/2022/6988525.].

An Outlier-Robust Growing Local Model Network for Recursive System Identification

An Outlier-Robust Growing Local Model Network for Recursive System Identification

Human Behavior Recognition in Outdoor Sports Based on the Local Error Model and Convolutional Neural Network.

With the rapid development of the Internet, various electronic products based on computer vision play an increasingly important role in people's daily lives. As one of the important topics of computer vision, human action recognition has become the main research hotspot in this field in recent years. The human motion recognition algorithm based on the convolutional neural network can realize the automatic extraction and learning of human motion features and achieve good classification performance. However, deep convolutional neural networks usually have a large number of layers, a large number of parameters, and a large memory footprint, while embedded wearable devices have limited memory space. Based on the traditional cross-entropy error-based training mode, the parameters of all hidden layers must be kept in memory and cannot be released until the end of forward and reverse error propagation. As a result, the memory used to store the parameters of the hidden layer cannot be released and reused, and the memory utilization efficiency is low, which leads to the backhaul locking problem, limiting the deployment and execution of deep convolutional neural networks on wearable sensor devices. Based on this, this topic designs a local error convolutional neural network model for human motion recognition tasks. Compared with the traditional global error, the local error constructed in this paper can train the convolutional neural network layer by layer, and the parameters of each layer can be trained independently according to the local error and does not depend on the gradient propagation of adjacent upper and lower layers. As a result, the memory used to store all hidden layer parameters can be released in advance without waiting for the end of forward and backward propagation, avoiding the problem of backhaul locking, and improving the memory utilization of convolutional neural networks deployed on embedded wearable devices.

Enhancement of the HILOMOT Algorithm with Modified EM and Modified PSO Algorithms for Nonlinear Systems Identification

Developing a mathematical model has become an inevitable need in studies of all disciplines. With advancements in technology, there is an emerging need to develop complex mathematical models. System identification is a popular way of constructing mathematical models of highly complex processes when an analytical model is not feasible. One of the many model architectures of system identification is to utilize a Local Model Network (LMN). Hierarchical Local Model Tree (HILOMOT) is an iterative LMN training algorithm that uses the axis-oblique split method to divide the input space hierarchically. The split positions of the local models directly influence the accuracy of the entire model. However, finding the best split positions of the local models presents a nonlinear optimization problem. This paper presents an optimized HILOMOT algorithm with enhanced Expectation–Maximization (EM) and Particle Swarm Optimization (PSO) algorithms which includes the normalization parameter and utilizes the reduced-parameter vector. Finally, the performance of the improved HILOMOT algorithm is compared with the existing algorithm by modeling the NOx emission model of a gas turbine and multiple nonlinear test functions of different orders and structures.

Automated synthesis of a local model network based nonlinear model predictive controller applied to the engine air path

The efficient and emission reducing control of the air path of an internal combustion engine is a challenging task due to its nonlinear and multivariate nature. By applying the well-known local model network approach to describe the nonlinear process in terms of linear operating point approximations, a fast and efficient model generation through data-driven system identification can be achieved. In this paper it is demonstrated how a nonlinear multivariate model predictive controller can be synthesised from the identified model directly by exploiting its representation in flatness coordinates. For the proposed controller, a compactly formulated quadratic programme results. Because of the uniform representation of all local models in controllability canonical form, a state observer is rendered unnecessary. Additionally, input and output constraints can be taken into account in the optimisation directly. The effectiveness of the control scheme is demonstrated successfully in jointly controlling the exhaust manifold pressure and the engine out NOx concentration for a heavy-duty engine on the testbed.

Adaptive Intrusion Detection in the Networking of Large-Scale LANs With Segmented Federated Learning

Predominant network intrusion detection systems (NIDS) aim to identify malicious traffic patterns based on a handcrafted dataset of rules. Recently, the application of machine learning in NIDS helps alleviate the enormous effort of human observation. Federated learning (FL) is a collaborative learning scheme concerning distributed data. Instead of sharing raw data, it allows a participant to share only a trained local model. Despite the success of existing FL solutions, in NIDS, a network's traffic data distribution does not always fit into the single global model of FL; some networks have similarities with each other but other networks do not. We propose Segmented-Federated Learning (Segmented-FL), where by employing periodic local model evaluation and network segmentation, we aim to bring similar network environments to the same group. A comparison between FL and our method was conducted against a range of metrics including the weighted precision, recall, and F1 score, using a collected dataset from 20 massively distributed networks within 60 days. By studying the optimized hyperparameters of Segmented-FL and employing three evaluation methods, it shows that Segmented-FL has better performance in all three types of intrusion detection tasks, achieving validation weighted F1 scores of 0.964, 0.803, and 0.912 with Method A, Method B, and Method C respectively. For each method, this scheme shows a gain of 0.1%, 4.0% and 1.1% in performance compared with FL.

Development, Validation, and Application of an Optimization Scheme for Impellers of Centrifugal Fans Using Computational Fluid Dynamics-Trained Metamodels

Abstract A quick method for the design of efficiency-optimal centrifugal fan impellers is presented. It is based on an evolutionary optimization algorithm that identifies the optimal geometrical parameters for a given aerodynamic objective function. The range of the geometrical parameters considered allows covering aerodynamic design points appropriate for the complete class of centrifugal fans. The quickness of the method stems from evaluating the objective function using metamodels. In total, four metamodels, based on local model networks (LMN) and multi-layer perceptrons (MLP), were trained and eventually aggregated to reduce the variance (stochastic) error. The training data consist of approximately 4000 characteristic curves obtained from automated numerical steady-state Reynolds-averaged Navier–Stokes (RANS) flow simulations. The computational domain as well as the number of grid nodes and their distribution in the domain were optimized in a pre-study. For verification, a grid independence study was carried out. In addition, two criteria were defined to detect aerodynamic operating points associated with non-physical performance predictions. Finally, validation was secured with experimental data from three exemplary impeller designs. The proposed optimization scheme requires a costly initial one-time computational fluid dynamics (CFD) effort, but then allows a quick design of centrifugal fan impellers for arbitrary design points. The search for an optimal centrifugal impeller requires less than 1 min on a standard personal computer, while allowing up to 105 objective function evaluations for one search. Moreover, predicted performance curves that always come along with each design were found to be very reliable in comparison with experiments.

A topological perspective on distributed network algorithms

More than two decades ago, combinatorial topology was shown to be useful for analyzing distributed fault-tolerant algorithms in shared memory systems and in message passing systems. In this work, we show that combinatorial topology can also be useful for analyzing distributed algorithms in failure-free networks of arbitrary structure. To illustrate this, we analyze consensus, set-agreement, and approximate agreement in networks, and derive lower bounds for these problems under classical computational settings, such as the local model and dynamic networks.

Local Model Network Based Multi-Model Predictive Control for a Boiler - Turbine System

A controller-weighted multi-model predictive control (MMPC) strategy based on local model network (LMN) is proposed to address the nonlinearity and wide operating range of the boiler-turbine (B-T) system with constraints. The LMN model of the nonlinear plant is identified off-line based on data-driven modeling method. Since each local model is valid only in local regime, different local constraints are considered in designing local predictive controllers corresponding to different local models. The local controllers are run in parallel and each controller is assigned with a weight by the implicit scheduling unit. The weighted sum of the outputs of local controllers is taken as a global control signal and applied to the plant. The efficacy of the proposed MMPC is validated by simulations on a boiler-turbine system.

Efficient and accurate technology for aircraft loads estimation

In this paper, a novel high-precision aircraft loads estimator (named EAGLE) will be introduced. EAGLE is based on the hybridization of two state-of-the-art loads estimation methods (a Luenberger observer and the local model network approach) and combines the individual advantages of the underlying methods. The hybrid method features an accurate estimation of structural loads (such as bending moments, shear forces, and torsional moments) due to maneuvers and gusts even for high load events and can be set up for a given flight test database in shortest time. Therefore, the hybrid method provides a cost-efficient setup of an accurate aircraft loads monitoring system. The paper focuses on the introduction of the new loads observer method and its validation by means of an extensive flight test database containing high maneuver and gust load events, which has been acquired using the ultra-light test aircraft UW-9 Sprint.

Fighting the curse of dimensionality with local model networks

Abstract This thesis is settled in the field of data-based modeling (identification) and specifically focuses on the weakening of the effects of the curse of dimensionality with local model networks (LMNs). The methods for fighting the curse of dimensionality originate from the fields of input selection and design of experiments (DoE).

Nonlinear Predictive Control for a Boiler–Turbine Unit Based on a Local Model Network and Immune Genetic Algorithm

This paper proposes a nonlinear model predictive control (NMPC) strategy based on a local model network (LMN) and a heuristic optimization method to solve the control problem for a nonlinear boiler–turbine unit. First, the LMN model of the boiler–turbine unit is identified by using a data-driven modeling method and converted into a time-varying global predictor. Then, the nonlinear constrained optimization problem for the predictive control is solved online by a specially designed immune genetic algorithm (IGA), which calculates the optimal control law at each sampling instant. By introducing an adaptive terminal cost in the objective function and utilizing local fictitious controllers to improve the initial population of IGA, the proposed NMPC can guarantee the system stability while the computational complexity is reduced since a shorter prediction horizon can be adopted. The effectiveness of the proposed NMPC is validated by simulations on a 500 MW coal-fired boiler–turbine unit.

Incremental Fuzzy C-Regression Clustering From Streaming Data for Local-Model-Network Identification

In this paper, a new approach to evolving fuzzy model identification from streaming data is given. The structure of the model is given as a local model network in Takagi–Sugeno form, and the partitioning of the input–output space is based on metrics in which these local models are defined as prototypes of the clusters. This means that the clusters and the local models share the same parameters; therefore, the number of parameters of the evolving system is much lower in comparison to similar systems of comparable complexity, and the problems of parameter identifiability are not a particular issue. The algorithm adds the local models in an incremental fashion and recursively adapts the local model parameters. The proposed algorithm is tested on three examples to demonstrate the main features. The first example is a simple simulated example with intersecting clusters; the second is a very well-known benchmark that treats the Mackey–Glass time series; the third is an example that shows the classification of the data from a laser rangefinder. These examples show the great potential of the proposed approach in certain applications.

A Data-Driven Based Voltage Control Strategy for DC-DC Converters: Application to DC Microgrid

This paper develops a data-driven strategy for identification and voltage control for DC-DC power converters. The proposed strategy does not require a pre-defined standard model of the power converters and only relies on power converter measurement data, including sampled output voltage and the duty ratio to identify a valid dynamic model for them over their operating regime. To derive the power converter model from the measurements, a local model network (LMN) is used, which is able to describe converter dynamics through some locally active linear sub-models, individually responsible for representing a particular operating regime of the power converters. Later, a local linear controller is established considering the identified LMN to generate the control signal (i.e., duty ratio) for the power converters. Simulation results for a stand-alone boost converter as well as a bidirectional converter in a test DC microgrid demonstrate merit and satisfactory performance of the proposed data-driven identification and control strategy. Moreover, comparisons to a conventional proportional-integral (PI) controllers demonstrate the merits of the proposed approach.

Intelligent identification of vehicle’s dynamics based on local model network

This paper proposes an intelligent approach for dynamic identification of the vehicles. The proposed approach is based on the data-driven identification and uses a high-performance local model network (LMN) for estimation of the vehicle’s longitudinal velocity, lateral acceleration and yaw rate. The proposed LMN requires no pre-defined standard vehicle model and uses measurement data to identify vehicle’s dynamics. The LMN is trained by hierarchical binary tree (HBT) learning algorithm, which results in a network with maximum generalizability and best linear or nonlinear structure. The proposed approach is applied to a measurement dataset, obtained from a Volvo V70 vehicle to estimate its longitudinal velocity, lateral acceleration and yaw rate. The results of identification revealed that the LMN can identify accurately the vehicle’s dynamics. Furthermore, comparison of LMN results and a multi-layer perceptron (MLP) neural network demonstrated the far-better performance of the proposed approach.

Investigating the Generalization Capability and Performance of Neural Networks and Neuro-Fuzzy Systems for Nonlinear Dynamics Modeling of Impaired Aircraft

One of the challenging problems in the case of aircraft failure is to determine the new altered dynamics of the impaired aircraft. Among various methods, neural networks and neuro-fuzzy systems can be used for high-fidelity modeling of the aircraft nonlinear dynamics with the aim of onboard applications in real time. However, the method with better generalization capability is more preferred specifically in the case of unpredicted aircraft failures. Generalization of a network is mainly dependent on the network's parameters, the employed training algorithm, and the amount of training data. In this paper, several neural networks and local model networks are trained using different training algorithms and different amounts of training data to model the nonlinear dynamics of an impaired aircraft with the damaged rudder. These networks are compared based on their generalizations to the new cases of rudder failure. The effect of using different amounts of training data on the generalization capability and performance of the networks has also been investigated. The results of this paper show that both network types have good performance but neural networks generalize better to the new failure cases than local model networks. Also based on the obtained results, a significant reduction in the number of training samples could be accomplished without a considerable decrease in the network's performance and generalization. Finally, a neural network-based sensitivity analysis method is proposed which utilizes the network's regression equation as an emulator for fast model evaluations and can be used as an advisory tool for choosing safer path planning strategies.

Implementation of cooperative Fuzzy model predictive control for an energy-efficient office building

The design and implementation of a nonlinear model predictive controller (MPC) scheme for an energy-efficient office building is presented. Because of the nonlinear building behavior local linear model networks are used to model the building dynamics. For each of the building's zones a dedicated Fuzzy model predictive controller (FMPC) is implemented. In addition to the FMPCs a global MPC is designed to control the coupling zone (concrete core activation). To coordinate the overactuated system a cooperative iteration-loop is designed, which leads to cooperative Fuzzy model predictive control (CFMPC). The CFMPC has been successfully implemented in a real office building, and the first results after commissioning are presented and discussed. Additional simulation studies demonstrate that the proposed CFMPC achieves higher comfort with less cost as compared to a rule-based controller.