Hamilton-Jacobi-Bellman Equations And Approximate Dynamic Programming On Time Scales, 2018 Missouri University of Science and Technology

#### Hamilton-Jacobi-Bellman Equations And Approximate Dynamic Programming On Time Scales, John E. Seiffertt Iv, Suman Sanyal, Donald C. Wunsch

*Donald C. Wunsch*

The time scales calculus is a key emerging area of mathematics due to its potential use in a wide variety of multidisciplinary applications. We extend this calculus to approximate dynamic programming (ADP). The core backward induction algorithm of dynamic programming is extended from its traditional discrete case to all isolated time scales. Hamilton-Jacobi-Bellman equations, the solution of which is the fundamental problem in the field of dynamic programming, are motivated and proven on time scales. By drawing together the calculus of time scales and the applied area of stochastic control via ADP, we have connected two major fields of research.

Gene Regulatory Networks Inference With Recurrent Neural Network Models, 2018 Missouri University of Science and Technology

#### Gene Regulatory Networks Inference With Recurrent Neural Network Models, Rui Xu, Donald C. Wunsch

*Donald C. Wunsch*

Large-scale time series gene expression data generated from DNA microarray experiments provide us a new means to reveal fundamental cellular processes, investigate functions of genes, and understand their relations and interactions. To infer gene regulatory networks from these data with effective computational tools has attracted intensive efforts from artificial intelligence and machine learning. Here, we use a recurrent neural network (RNN), trained with particle swarm optimization (PSO), to investigate the behaviors of regulatory networks. The experimental results, on a synthetic data set and a real data set, show that the proposed model and algorithm can effectively capture the dynamics of ...

Fuzzy Pso: A Generalization Of Particle Swarm Optimization, 2018 Missouri University of Science and Technology

#### Fuzzy Pso: A Generalization Of Particle Swarm Optimization, S. Abdelshahid, Donald C. Wunsch, Ashraf M. Abdelbar

*Donald C. Wunsch*

In standard particle swarm optimization (PSO), the best particle in each neighborhood exerts its influence over other particles in the neighborhood. In this paper, we propose fuzzy PSO, a generalization which differs from standard PSO in the following respect: charisma is defined to be a fuzzy variable, and more than one particle in each neighborhood can have a non-zero degree of charisma, and, consequently, is allowed to influence others to a degree that depends on its charisma. We evaluate our model on the weighted maximum satisfiability (maxsat) problem, comparing performance to standard PSO and to Walk-Sat.

Feedback Linearization Based Power System Stabilizer Design With Control Limits, 2018 Missouri University of Science and Technology

#### Feedback Linearization Based Power System Stabilizer Design With Control Limits, Wenxin Liu, Ganesh K. Venayagamoorthy, Donald C. Wunsch, Jagannathan Sarangapani

*Donald C. Wunsch*

In power system controls, simplified analytical models are used to represent the dynamics of power system and controller designs are not rigorous with no stability analysis. One reason is because the power systems are complex nonlinear systems which pose difficulty for analysis. This paper presents a feedback linearization based power system stabilizer design for a single machine infinite bus power system. Since practical operating conditions require the magnitude of control signal to be within certain limits, the stability of the control system under control limits is also analyzed. Simulation results under different kinds of operating conditions show that the controller ...

Fed-Batch Dynamic Optimization Using Generalized Dual Heuristic Programming, 2018 Missouri University of Science and Technology

#### Fed-Batch Dynamic Optimization Using Generalized Dual Heuristic Programming, Donald C. Wunsch, M. S. Iyer

*Donald C. Wunsch*

Traditionally fed-batch biochemical process optimization and control uses complicated theoretical off-line optimizers, with no online model adaptation or re-optimization. This study demonstrates the applicability, effectiveness, and economic potential of a simple phenomenological model for modeling, and an adaptive critic design, generalized dual heuristic programming, for online re-optimization and control of an aerobic fed-batch fermentor. The results are compared with those obtained using a heuristic random optimizer

Experiments On Adaptive Techniques For Host-Based Intrusion Detection, 2018 Missouri University of Science and Technology

#### Experiments On Adaptive Techniques For Host-Based Intrusion Detection, Timothy Draelos, Michael Collins, David Duggan, Edward Thomas, Donald C. Wunsch

*Donald C. Wunsch*

This research explores four experiments of adaptive host-based intrusion detection (ID) techniques in an attempt to develop systems that can detect novel exploits. The technique considered to have the most potential is adaptive critic designs (ACDs) because of their utilization of reinforcement learning, which allows learning exploits that are difficult to pinpoint in sensor data. Preliminary results of ID using an ACD, an Elman recurrent neural network, and a statistical anomaly detection technique demonstrate an ability to learn to distinguish between clean and exploit data. We used the Solaris Basic Security Module (BSM) as a data source and performed considerable ...

Experimental Verification Of Derivatives Adaptive Critic Based Neurocontroller Performance On Single Turbogenerators On The Electric Power Grid, 2018 Missouri University of Science and Technology

#### Experimental Verification Of Derivatives Adaptive Critic Based Neurocontroller Performance On Single Turbogenerators On The Electric Power Grid, Ganesh K. Venayagamoorthy, Donald C. Wunsch, Ronald G. Harley

*Donald C. Wunsch*

The design and real-time implementation of derivatives adaptive critic based neurocontrollers that replace the conventional automatic voltage regulators (AVRs) and turbine governors are presented in this paper. The feedback variables to the neurocontroller are completely based on local measurements from the turbogenerator. Experimental verification results are presented to show the superior performance of the derivatives adaptive critic based neurocontroller, compared to the conventional AVR and turbine governor controllers equipped with a power system stabilizer.

Excitation And Turbine Neurocontrol With Derivative Adaptive Critics Of Multiple Generators On The Power Grid, 2018 Missouri University of Science and Technology

#### Excitation And Turbine Neurocontrol With Derivative Adaptive Critics Of Multiple Generators On The Power Grid, Ganesh K. Venayagamoorthy, Donald C. Wunsch, Ronald G. Harley

*Donald C. Wunsch*

Based on derivative adaptive critics, neurocontrollers for excitation and turbine control of multiple generators on the electric power grid are presented. The feedback variables are completely based on local measurements. Simulations on a three-machine power system demonstrate that the neurocontrollers are much more effective than conventional PID controllers, the automatic voltage regulators and the governors, for improving the dynamic performance and stability under small and large disturbances

Experimental Studies With Continually Online Trained Artificial Neural Network Identifiers For Multiple Turbogenerators On The Electric Power Grid, 2018 Missouri University of Science and Technology

#### Experimental Studies With Continually Online Trained Artificial Neural Network Identifiers For Multiple Turbogenerators On The Electric Power Grid, Ganesh K. Venayagamoorthy, Donald C. Wunsch, Ronald G. Harley

*Donald C. Wunsch*

The increasing complexity of a modern power grid highlights the need for advanced system identification techniques for effective control of power systems. This paper provides a new method for nonlinear identification of turbogenerators in a 3-machine 6-bus power system using online trained feedforward neural networks. Each turbogenerator in the power system is equipped with a neuro-identifier, which is able to identify its particular turbogenerator and the rest of the network to which it is connected from moment to moment, based on only local measurements. Each neuro-identifier can then be used in the design of a nonlinear neurocontroller for each turbogenerator ...

Evolutionary Algorithms, Markov Decision Processes, Adaptive Critic Designs, And Clustering: Commonalities, Hybridization And Performance, 2018 Missouri University of Science and Technology

#### Evolutionary Algorithms, Markov Decision Processes, Adaptive Critic Designs, And Clustering: Commonalities, Hybridization And Performance, Donald C. Wunsch, Samuel A. Mulder

*Donald C. Wunsch*

We briefly review and compare the mathematical formulation of Markov decision processes (MDP) and evolutionary algorithms (EA). In so doing, we observe that the adaptive critic design (ACD) approach to MDP can be viewed as a special form of EA. This leads us to pose pertinent questions about possible expansions of the methodology of ACD. This expansive view of EA is not limited to ACD. We discuss how it is possible to consider the powerful chained Lin Kernighan (chained LK) algorithm for the traveling salesman problem (TSP) as a degenerate case of EA. Finally, we review some recent TSP results ...

Engine Data Classification With Simultaneous Recurrent Network Using A Hybrid Pso-Ea Algorithm, 2018 Missouri University of Science and Technology

#### Engine Data Classification With Simultaneous Recurrent Network Using A Hybrid Pso-Ea Algorithm, Xindi Cai, Donald C. Wunsch

*Donald C. Wunsch*

We applied an architecture which automates the design of simultaneous recurrent network (SRN) using a new evolutionary learning algorithm. This new evolutionary learning algorithm is based on a hybrid of particle swarm optimization (PSO) and evolutionary algorithm (EA). By combining the searching abilities of these two global optimization methods, the evolution of individuals is no longer restricted to be in the same generation, and better performed individuals may produce offspring to replace those with poor performance. The novel algorithm is then applied to the simultaneous recurrent network for the engine data classification. The experimental results show that our approach gives ...

Efficient Training Techniques For Classification With Vast Input Space, 2018 Missouri University of Science and Technology

#### Efficient Training Techniques For Classification With Vast Input Space, Donald C. Wunsch, Emad W. Saad, J. J. Choi, J. L. Vian

*Donald C. Wunsch*

Strategies to efficiently train a neural network for an aerospace problem with a large multidimensional input space are developed and demonstrated. The neural network provides classification for over 100,000,000 data points. A query-based strategy is used that initiates training using a small input set, and then augments the set in multiple stages to include important data around the network decision boundary. Neural network inversion and oracle query are used to generate the additional data, jitter is added to the query data to improve the results, and an extended Kalman filter algorithm is used for training. A causality index ...

Dynamic Re-Optimization Of A Fed-Batch Fermentor Using Heuristic Dynamic Programming, 2018 Missouri University of Science and Technology

#### Dynamic Re-Optimization Of A Fed-Batch Fermentor Using Heuristic Dynamic Programming, Donald C. Wunsch, M. S. Iyer

*Donald C. Wunsch*

Traditionally, fed-batch biochemical process optimization and control uses complicated theoretical off-line optimizers, with no online model adaptation or re-optimization. This study demonstrates the applicability, effectiveness, and economic potential of a simple phenomenological model for modeling, and an adaptive critic design, heuristic dynamic programming, for online re-optimization and control of an aerobic fed-batch fermentor. The results are compared with those obtained using a heuristic random optimizer

Divide And Conquer Evolutionary Tsp Solution For Vehicle Path Planning, 2018 Missouri University of Science and Technology

#### Divide And Conquer Evolutionary Tsp Solution For Vehicle Path Planning, Ryan J. Meuth, Donald C. Wunsch

*Donald C. Wunsch*

The problem of robotic area coverage is applicable to many domains, such as search, agriculture, cleaning, and machine tooling. The robotic area coverage task is concerned with moving a vehicle with an effector, or sensor, through the task space such that the sensor passes over every point in the space. For covering complex areas, back and forth paths are inadequate. This paper presents a real-time path planning architecture consisting of layers of a clustering method to divide and conquer the problem combined with a two layered, global and local optimization method. This architecture is able to optimize the execution of ...

Dual Heuristic Programming Excitation Neurocontrol For Generators In A Multimachine Power System, 2018 Missouri University of Science and Technology

#### Dual Heuristic Programming Excitation Neurocontrol For Generators In A Multimachine Power System, Ganesh K. Venayagamoorthy, Donald C. Wunsch, Ronald G. Harley

*Donald C. Wunsch*

The design of nonlinear optimal neurocontrollers that replace the conventional automatic voltage regulators for excitation control of turbogenerators in a multimachine power system is presented in this paper. The neurocontroller design is based on dual heuristic programming (DHP), a powerful adaptive critic technique. The feedback variables are completely based on local measurements from the generators. Simulations on a three-machine power system demonstrate that DHP-based neurocontrol is much more effective than the conventional proportional-integral-derivative control for improving dynamic performance and stability of the power grid under small and large disturbances. This paper also shows how to design optimal multiple neurocontrollers for ...

Detection Of Basal Cell Carcinoma Using Electrical Impedance And Neural Networks, 2018 Missouri University of Science and Technology

#### Detection Of Basal Cell Carcinoma Using Electrical Impedance And Neural Networks, Rohit Dua, Daryl G. Beetner, William V. Stoecker, Donald C. Wunsch

*Donald C. Wunsch*

Variations in electrical impedance over frequency might be used to distinguish basal cell carcinoma (BCC) from benign skin lesions, although the patterns that separate the two are nonobvious. Artificial neural networks (ANNs) may be good pattern classifiers for this application. A preliminary study to show the potential of neural networks to distinguish benign from malignant skin lesions using electrical impedance is presented. Electrical impedance was measured in vivo from 1 kHz to 1 MHz at five virtual depths on 18 BCC and 16 benign or premalignant lesions. A feed-forward neural network was trained using back propagation to classify these lesions ...

Detection And Identification Of Vehicles Based On Their Unintended Electromagnetic Emissions, 2018 Missouri University of Science and Technology

#### Detection And Identification Of Vehicles Based On Their Unintended Electromagnetic Emissions, Xiaopeng Dong, Haixiao Weng, Daryl G. Beetner, Todd H. Hubing, Donald C. Wunsch, Michael Noll, Huseyin Goksu, Benjamin Moss

*Donald C. Wunsch*

When running, vehicles with internal combustion engines radiate electromagnetic emissions that are characteristic of the vehicle. Emissions depend on the electronics, harness wiring, body type, and many other features. Since emissions are unique to each vehicle, these may be used for identification purposes. This paper investigates a procedure for detecting and identifying vehicles based on their RF emissions. Parameters like the average magnitude or standard deviation of magnitude within a frequency band were extracted from measured emission data. These parameters were used as inputs to an artificial neural network (ANN) that was trained to identify the vehicle that produced the ...

Detection And Classification Of Impact-Induced Damage In Composite Plates Using Neural Networks, 2018 Missouri University of Science and Technology

#### Detection And Classification Of Impact-Induced Damage In Composite Plates Using Neural Networks, Rohit Dua, Steve Eugene Watkins, Donald C. Wunsch, K. Chandrashekhara, Farhad Akhavan

*Donald C. Wunsch*

Artificial neutral networks (ANN) can be used as an online health monitoring systems (involving damage assessment, fatigue monitoring and delamination detection) for composite structures owing to their inherent fast computing speeds, parallel processing and ability to learn and adapt to the experimental data. The amount of impact-induced strain on a composite structure can be found using strain sensors attached to composite structures. Prior work has shown that strain-based ANN can characterize impact energy on composite plates and that strain signatures can be associated with damage types and severity. This paper reports the extension of this approach for damage classification using ...

Counterexample Of A Claim Pertaining To The Synthesis Of A Recurrent Neural Network, 2018 Missouri University of Science and Technology

#### Counterexample Of A Claim Pertaining To The Synthesis Of A Recurrent Neural Network, Xindi Cai, Donald C. Wunsch

*Donald C. Wunsch*

Recurrent neural networks have received much attention due to their nonlinear dynamic behavior. One such type of dynamic behavior is that of setting a fixed stable state. This paper shows a counterexample to the claim of A.N. Michel et al. (IEEE Control Systems Magazine, vol. 15, pp. 52-65, Jun. 1995), that "sparse constraints on the interconnecting structure for a given neural network are usually expressed as constraints which require that pre-determined elements of T [a real n×n matrix acting on a real n-vector valued function] be zero", for the synthesis of sparsely interconnected recurrent neural networks.

Coordinated Machine Learning And Decision Support For Situation Awareness, 2018 Missouri University of Science and Technology

#### Coordinated Machine Learning And Decision Support For Situation Awareness, Timothy Draelos, Pengchu Zhang, Donald C. Wunsch, John E. Seiffertt Iv, Gregory Conrad, Nathan Brannon

*Donald C. Wunsch*

For applications such as force protection, an effective decision maker needs to maintain an unambiguous grasp of the environment. Opportunities exist to leverage computational mechanisms for the adaptive fusion of diverse information sources. The current research employs neural networks and Markov chains to process information from sources including sensors, weather data, and law enforcement. Furthermore, the system operator's input is used as a point of reference for the machine learning algorithms. More detailed features of the approach are provided, along with an example force protection scenario.