Featured Researches

Systems And Control

A Heuristic for Dynamic Output Predictive Control Design for Uncertain Nonlinear Systems

In this paper, a simple heuristic is proposed for the design of uncertainty aware predictive controllers for nonlinear models involving uncertain parameters. The method relies on Machine Learning-based approximation of ideal deterministic MPC solutions with perfectly known parameters. An efficient construction of the learning data set from these off-line solutions is proposed in which each solution provides many samples in the learning data. This enables a drastic reduction of the required number of Non Linear Programming problems to be solved off-line while explicitly exploiting the statistics of the parameters dispersion. The learning data is then used to design a fast on-line output dynamic feedback that explicitly incorporate information of the statistics of the parameters dispersion. An example is provided to illustrate the efficiency and the relevance of the proposed framework. It is in particular shown that the proposed solution recovers up to 78\% of the expected advantage of having a perfect knowledge of the parameters compared to nominal design.

Read more
Systems And Control

A Hybrid Deep Learning-Based State Forecasting Method for Smart Power Grids

Smart power grids are one of the most complex cyber-physical systems, delivering electricity from power generation stations to consumers. It is critically important to know exactly the current state of the system as well as its state variation tendency; consequently, state estimation and state forecasting are widely used in smart power grids. Given that state forecasting predicts the system state ahead of time, it can enhance state estimation because state estimation is highly sensitive to measurement corruption due to the bad data or communication failures. In this paper, a hybrid deep learningbased method is proposed for power system state forecasting. The proposed method leverages Convolutional Neural Network (CNN) for predicting voltage magnitudes and a Deep Recurrent Neural Network (RNN) for predicting phase angels. The proposed CNN-RNN model is evaluated on the IEEE 118-bus benchmark. The results demonstrate that the proposed CNNRNN model achieves better results than the existing techniques in the literature by reducing the normalized Root Mean Squared Error (RMSE) of predicted voltages by 10%. The results also show a 65% and 35% decrease in the average and maximum absolute error of voltage magnitude forecasting.

Read more
Systems And Control

A Learning-based Stochastic Driving Model for Autonomous Vehicle Testing

In the simulation-based testing and evaluation of autonomous vehicles (AVs), how background vehicles (BVs) drive directly influences the AV's driving behavior and further impacts the testing result. Existing simulation platforms use either pre-determined trajectories or deterministic driving models to model the BVs' behaviors. However, pre-determined BV trajectories can not react to the AV's maneuvers, and deterministic models are different from real human drivers due to the lack of stochastic components and errors. Both methods lead to unrealistic traffic scenarios. This paper presents a learning-based stochastic driving model that meets the unique needs of AV testing, i.e. interactive and human-like. The model is built based on the long-short-term-memory (LSTM) architecture. By incorporating the concept of quantile-regression to the loss function of the model, the stochastic behaviors are reproduced without any prior assumption of human drivers. The model is trained with the large-scale naturalistic driving data (NDD) from the Safety Pilot Model Deployment(SPMD) project and then compared with a stochastic intelligent driving model (IDM). Analysis of individual trajectories shows that the proposed model can reproduce more similar trajectories to human drivers than IDM. To validate the ability of the proposed model in generating a naturalistic driving environment, traffic simulation experiments are implemented. The results show that the traffic flow parameters such as speed, range, and headway distribution match closely with the NDD, which is of significant importance for AV testing and evaluation.

Read more
Systems And Control

A Low-Order Dynamic Model of Counterflow Heat Exchangers for the Purpose of Monitoring Transient and Steady-State Operating Phases

We present a model-based real-time method to monitor a counterflow heat exchanger's thermal performance for all operating conditions. A first principle reference model that describes the reference counterflow process in an accurate manner is derived first. Real gas behavior is taken into account. Without simplifications, the respective equations must be solved in an iterative, computationally expensive manner, which prohibits their use for real-time monitoring purposes. Therefore, we propose one-step-solvable model equations, resulting in an approximate but quick model, which is able to track an important thermal property reliably. The monitoring, i.e., the online estimation of the thermal properties, is achieved via a nonlinear Kalman-Filter. Due to the low-order dynamic model formulation, the overall monitoring scheme is accompanied by an acceptable computational burden. Moreover, it is easy to deploy and to adapt in industrial practice. Monitoring results, where the reference model replaces a real process with supercritical carbon dioxide, are given and discussed herein.

Read more
Systems And Control

A Meta-learning based Distribution System Load Forecasting Model Selection Framework

This paper presents a meta-learning based, automatic distribution system load forecasting model selection framework. The framework includes the following processes: feature extraction, candidate model labeling, offline training, and online model recommendation. Using user load forecasting needs as input features, multiple meta-learners are used to rank the available load forecast models based on their forecasting accuracy. Then, a scoring-voting mechanism weights recommendations from each meta-leaner to make the final recommendations. Heterogeneous load forecasting tasks with different temporal and technical requirements at different load aggregation levels are set up to train, validate, and test the performance of the proposed framework. Simulation results demonstrate that the performance of the meta-learning based approach is satisfactory in both seen and unseen forecasting tasks.

Read more
Systems And Control

A Methodology for the Development of RL-Based Adaptive Traffic Signal Controllers

This article proposes a methodology for the development of adaptive traffic signal controllers using reinforcement learning. Our methodology addresses the lack of standardization in the literature that renders the comparison of approaches in different works meaningless, due to differences in metrics, environments, and even experimental design and methodology. The proposed methodology thus comprises all the steps necessary to develop, deploy and evaluate an adaptive traffic signal controller -- from simulation setup to problem formulation and experimental design. We illustrate the proposed methodology in two simple scenarios, highlighting how its different steps address limitations found in the current literature.

Read more
Systems And Control

A Multi-Agent Deep Reinforcement Learning Approach for a Distributed Energy Marketplace in Smart Grids

This paper presents a Reinforcement Learning (RL) based energy market for a prosumer dominated microgrid. The proposed market model facilitates a real-time and demanddependent dynamic pricing environment, which reduces grid costs and improves the economic benefits for prosumers. Furthermore, this market model enables the grid operator to leverage prosumers storage capacity as a dispatchable asset for grid support applications. Simulation results based on the Deep QNetwork (DQN) framework demonstrate significant improvements of the 24-hour accumulative profit for both prosumers and the grid operator, as well as major reductions in grid reserve power utilization.

Read more
Systems And Control

A Non-Isolated High Step-Up Interleaved DC-DC Converter with Diode-Capacitor Multiplier Cells and Dual Coupled Inductors

In this paper, a non-isolated high step-up dc-dc converter is presented. The proposed converter is composed of an interleaved structure and diode-capacitor multiplier cells for interfacing low-voltage renewable energy sources to high-voltage distribution buses. The aforementioned topology can provide a very high voltage gain due to employing the coupled inductors and the diode-capacitor cells. The coupled inductors are connected to the diode-capacitor multiplier cells to achieve the interleaved energy storage in the output side. Furthermore, the proposed topology provides continuous input current with low voltage stress on the power devices. The reverse recovery problem of the diodes is reduced. This topology can be operated at a reduced duty cycle by adjusting the turn ratio of the coupled inductors. Moreover, the performance comparison between the proposed topology and other converters are introduced. The design considerations operation principle, steady-state analysis, simulation results, and experimental verifications are presented. Therefore, a 500-W hardware prototype with an input voltage of 30-V and an output voltage of 1000-V is built to verify the performance and the theoretical analysis.

Read more
Systems And Control

A Note on Order and Index Reduction for Descriptor Systems

We present order reduction results for linear time invariant descriptor systems. Results are given for both forced and unforced systems as well methods for constructing the reduced order systems. Our results establish a precise connection between classical and new results on this topic, and lead to an elementary construction of quasi-Weierstrass forms for a descriptor system. Examples are given to illustrate the usefulness of our results.

Read more
Systems And Control

A Novel Cluster Classify Regress Model Predictive Controller Formulation; CCR-MPC

In this work, we develop a novel data-driven model predictive controller using advanced techniques in the field of machine learning. The objective is to regulate control signals to adjust the desired internal room setpoint temperature, affected indirectly by the external weather states. The methodology involves developing a time-series machine learning model with either a Long Short Term Memory model (LSTM) or a Gradient Boosting Algorithm (XGboost), capable of forecasting this weather states for any desired time horizon and concurrently optimising the control signals to the desired set point. The supervised learning model for mapping the weather states together with the control signals to the room temperature is constructed using a previously developed methodology called Cluster Classify regress (CCR), which is similar in style but scales better to high dimensional dataset than the well-known Mixture-of-Experts. The overall method called CCR-MPC involves a combination of a time series model for weather states prediction, CCR for forwarding and any numerical optimisation method for solving the inverse problem. Forward uncertainty quantification (Forward-UQ) leans towards the regression model in the CCR and is attainable using a Bayesian deep neural network or a Gaussian process (GP). For this work, in the CCR modulation, we employ K-means clustering for Clustering, XGboost classifier for Classification and 5th order polynomial regression for Regression. Inverse UQ can also be obtained by using an I-ES approach for solving the inverse problem or even the well-known Markov chain Monte Carlo (MCMC) approach. The developed CCR-MPC is elegant, and as seen on the numerical experiments is able to optimise the controller to attain the desired setpoint temperature.

Read more

Ready to get started?

Join us today