This work introduces a novel multi-agent reinforcement learning framework with periodic adversarial training to enhance coordinated control and power quality in multi-microgrid systems facing false data injection attacks. The architecture combines global state sharing for improved coordination with adversarial training for resilience against cyber threats. Evaluations on a modified IEEE 33-bus system with three microgrids demonstrate the enhanced multi-agent reinforcement learning (EMARL) method's superiority, achieving a median voltage deviation of 0.003 per unit during normal operations—91.9% better than uncontrolled systems, 81.3% better than multi-agent soft actor-critic (MASAC), and 50.0% better than multi-agent deep deterministic policy gradient (MADDPG). Under random attacks, the median deviation of EMARL remains within 0.01 per unit, outperforming MASAC (0.018 per unit) and MADDPG (0.02 per unit). The framework consistently maintained voltages within the 0.95-1.05 per unit range.

In this paper, a novel dual adaptive control strategy is proposed for LCL-filtered grid-connected inverters operating under different grid impedance conditions ranging from strong to weak grids. The proposed approach combines two adaptive mechanisms: a hybrid virtual impedance damping controller that adaptively adjusts the virtual resistance, inductance, and capacitance based on real-time grid impedance detection, and an event-triggered second-order phase lead compensator that is activated upon detection of a degraded signal quality. The DSOGI-FLL-based (double generalized second order integrator phase-locked loop) grid impedance estimator continuously monitors the grid condition, while the intelligent event-triggered switching mechanism seamlessly chooses between control strategies based on signal quality analysis. Results show that the methodology tests and realizes the stable operation of the system over the 0-20 mH equivalent inductive grid impedance range, which is much better than the conventional PI control and traditional phase lead compensation. It has excellent dynamic response when the grid impedance changes. Compared with the traditional method, the proposed dual adaptive strategy effectively extends the stable operation range while maintaining high power quality and fast transient response.

The multi-engine parallel inverter system in an is-landed microgrid faces complex and variable load perturbations, and the traditional control strategy is difficult to balance the steady-state accuracy and dynamic response performance of the system. In this paper, an enhanced fourth-order multi-extended state observer (ESO) intelligent regulation of the adaptive virtual impedance (AVI) system is proposed to realize the accurate identification and adaptive suppression of disturbances. The method innovatively constructs a multi-ESO structure acting on frequency, active power, and reactive power, and introduces a dedicated ESO coordinator to realize multi-objective co-optimization, designs a disturbance frequency analysis mechanism based on Fast Fourier Transform (FFT), and intelligently adjusts the ESO bandwidth and prediction model according to the perturbation characteristics, and develops a targeted anti-oscillation mechanism and an adaptive virtual impedance synergistic strategy to enhance the robustness of the system. The simulation results show that the proposed control strategy significantly reduces the voltage overshoot and recovery time during sudden load changes, improves the accuracy of the power distribution, effectively suppresses the circulating current, and especially shows obvious advantages under different disturbance conditions.

This paper presents a cockatiel-inspired hierarchical battery equalization strategy for multi-microgrid energy storage systems. The bio-inspired approach incorporates priority-based SOC processing, environmental perception mechanisms, and intelligent learning capabilities. Compared to traditional methods, the proposed strategy achieves 8.5% faster equalization completion (259.253s vs 283.023s), 55% better voltage regulation (1.6% vs 3.6% deviation), and 74% faster current response (0.931s vs 3.569s). Battery current stress is reduced by 50% (24.2A vs 47.8A), effectively extending battery lifetime while maintaining multi-microgrid stability. The biological coordination paradigm demonstrates superior adaptability and performance for distributed energy storage management.

In this paper, we propose an innovative multimicrogrid control method that organically combines a fast response consensus algorithm with a battery state-of-charge (SOC) equalization strategy. The proposed framework enables collaborative coordination across microgrid networks through strategically weighted communication structures. During scenarios involving rapid load variations and battery state transitions, the system achieves a remarkable response time of 197.22ms, demonstrating over 70% performance enhancement compared to conventional approaches. To mitigate control instabilities encountered during SOC balancing procedures, the methodology incorporates an intersection identification scheme coupled with hierarchical signal conditioning techniques, integrating linear SOC equalization logic and cyclic charging/discharging strategy, which ensures the healthy operation of the batteries while realizing fast response. This unified control approach successfully prolongs energy storage system operational life while preserving voltage regulation and rapid power distribution capabilities, providing a comprehensive solution for efficient cooperative control of distributed microgrid systems.

This paper proposes an active balancing architecture for parallel-connected battery modules in electric vehicle (EV) systems using dual active bridge (DAB) converters combined with MOSFET matrix switching networks to achieve state-of-charge (SOC) balancing without relying on a multi-transformer configuration. Meanwhile, an adaptive unscented Kalman filter (AUKF) is developed to solve the problem of inaccurate SOC estimation under nonlinear module dynamics. The effectiveness of the architecture is demonstrated by the MATLAB/Simulink simulation results: compared with the conventional passive balancing or multi-transformer-based solutions, the proposed DAB-MOSFET switch matrix topology utilizes a single DAB converter and reconfigurable MOSFET switching matrix to achieve a balancing efficiency of 96.4%, which greatly reduces the number of magnetic components and improves the system reliability, while the use of the AUKF estimation algorithm well reduces the impact of measurement noise on the SOC estimation. This work establishes a simulation-proven co-design framework that combines advanced power electronics with adaptive estimation algorithms to provide a scalable and cost-effective solution for large-scale EV battery systems.

This paper proposes a Vehicle-to-Grid (V2G) deployment strategy based on voltage sensitivity analysis to enhance voltage support and system stability in distribution networks with high electric vehicle (EV) integration. A stochastic model captures EV charging variability, and voltage sensitivity at each bus of the IEEE 33-node system guides the design of three spatial deployment strategies: centralized, random, and sensitivity-based distributed allocation. The simulation results show that the sensitivity-based strategy provides the most balanced voltage improvement, eliminating violations across the network. The centralized strategy offers strong but localized support, while the random strategy performs poorly overall. Further analysis under partial V2G node failure scenarios reveals that distributed deployment strategies retain significantly better voltage regulation compared to centralized deployment. These findings confirm that sensitivity-guided distributed V2G allocation offers a practical and robust solution for improving voltage stability and planning future EV integration.

The increasing penetration of electric vehicles (EVs) introduces new challenges to power distribution networks, particularly due to the stochastic and clustered nature of EV charging behaviors. Conventional scheduling approaches often neglect user-level heterogeneity, resulting in inefficiencies in both grid operations and user satisfaction. This paper proposes a user-centric multi-objective scheduling framework based on a novel charging Urgency Index (UI), which quantifies the flexibility of individual EVs. Flexible charging demands are identified via a self-adaptive threshold, upon which a multi-objective optimization model is developed to balance grid performance with user experience. The proposed framework utilizes the Multi-objective Grey Wolf Optimization (MOGWO) to explore Pareto-optimal solutions considering five conflicting objectives: peak-valley gap, load fluctuation, user delay, load correlation, and charging cost. Simulation results validate the effectiveness of the approach in improving grid load profiles and economic outcomes, while preserving fairness and flexibility in EV user scheduling.

Hybrid energy storage system (HESS) in microgrid applications is controlled to balance the power between generation and load sides. However, power loss of converting and model parameter mismatch would affect the control performance. To this end, a deadbeat control algorithm for HESS combined with deep reinforcement learning is proposed in this article. In the proposed method, the variation of optimal HESS current reference caused by nonlinear power loss and model mismatch is regarded as a centralized disturbance that can be compensated by a deep deterministic policy gradient agent, and a deadbeat control generates an optimal duty cycle based on a precise reference current to eliminate system steady-state error and improve dynamic response speed. The effectiveness of the proposed algorithm is verified through simulation and hardware experiments. Results demonstrate that the steady-state error can be maintained within 1%. Compared to conventional deadbeat control methods, the proposed method reduces the bus voltage spike and settling time by 34.24%–44.44% and 16.66%–40.00%, respectively.

This paper investigates the impact of different earthing methods on overvoltage within modular multilevel Voltage Source Converter (MMC VSC) based Direct Current (DC) transmission and distribution grids, particularly in meshed multi-terminal (MTDC) configurations. A bipolar MTDC grid is modelled and simulated in PSCAD/EMTDC to investigate the influence of earthing on midpoint overvoltages. The transient response of the grid under DC fault scenarios is analysed through various case-studies, including solid earthing placement, variations in earthing impedance, and the application of surge arresters. The findings show the critical role of earthing in mitigating overvoltage, with low-resistive earthing options proving to be the most effective trade-off.

In the DC fault event, DC link capacitors immediately discharge a colossal current of high frequency. These large currents of high frequencies can instantly damage the components attached to the DC transmission system. Therefore, a rapid and robust DC fault detection topology is necessary, facilitating protection coordination in forthcoming Multi-Terminal Direct Current (MTDC) systems. This study introduces a novel fault detection technique using the time derivative of the High Pass Filter (HPF) response to identify fault currents in just a few microseconds (µs). This topology enables protection coordination for modern MTDC systems. Furthermore, this technique can identify different types of DC faults and remains unaffected by AC faults and load variation.

The DC link capacitor discharge in the event of a DC fault is rapid and contains high frequencies. This rapid discharge of high current and frequency interface with DC bus, Voltage Source Converter (VSC), and AC source. This interface damages the equipment and possibly living beings in proximity. Therefore, it is necessary to develop a topology to detect and identify fault types promptly, along with the isolation and restoration of the system. This study introduces an improved novel fault detection technique using the Highpass Chebyshev type 2 filter due to its flatter pass-band response. Further, the polarities of peaks obtained from the fault detection calculation are used again for another proposed novel fault location method. It identifies fault types to locate and isolate the faulty system using only polarities of amplitude response peaks. In the simulation, both methods were fast and accurate in detecting, locating, and identifying appropriate fault types only by High Pass Filter (HPF) amplitude response peaks and their polarities.

The imperative to transition towards renewable energy sources has catalyzed global interest in harnessing offshore wind energy, owing to its immense potential to boost sustainable electricity generation. However, integrating offshore wind farms into existing power grids poses formidable challenges, encompassing grid stability, power quality, and fault management. Nonetheless, the integration of offshore wind farms (OWF) with the power grid presents a significant opportunity for advancing renewable energy generation on a large scale. This study embarks on an exploration of the intricate dynamics and performance metrics inherent in the integration of OWF with onshore grids (OnG), particularly focusing on the interaction between a 480 MW Offshore high-voltage direct current (HVDC) system and the modified IEEE 39-bus network. The study incorporates a comprehensive wind generation profile from a 10-minute to 24-hourly timescale synthetic wind speed dataset for actual or proposed New Zealand wind farm sites, as well as load profiles, to provide a robust basis for analysis. Utilizing detailed modeling and analysis, this research delves into the nuanced aspects of offshore wind farm integration, shedding light on key factors such as grid stability, power flow dynamics, and system reliability.

In voltage source converter-based low-voltage dc systems (LVDC), the fault current of the dc-link capacitor is considerably high and destructive to system infrastructure. It is necessary to develop effective fault detection methods with sufficient sensitivity and accuracy. To achieve these targets, a complex domain analysis-based fault detection method is proposed in this article. Specifically, the proposed method first fits the transient current into a linear combination of exponential functions, which is solved in the Z-domain-based on the Padé approximation. Second, exponents of the fitted function are projected into the complex plane. A state circle centered on the origin is defined on the complex plane to detect dc faults according to the position of projection points relative to the state circle. The proposed method can differentiate several typical situations via theoretical analysis, including fault line transients, healthy line transients, and load switching. The performance of proposed method is validated with an experimental multiterminal LVDC system to reveal its effective performance compared with present frequency domain based methods, including the wavelet transform, the short-time Fourier transform, the S transform, and the Hilbert–Huang transform.

This article introduces a way to optimize the power output of a photovoltaic (PV) system by implementing a maximum power point tracking (MPPT) technique. The algorithm is based on adaptive particle swarm optimization (APSO) under partial shading conditions (PSC). APSO aims to solve the issue that conventional MPPT algorithms cannot track the optimal global solution when the photovoltaic array's power-voltage (P-V) curve shows multiple peaks under PSC. In APSO, the algorithm adaptively adjusts the learning factor and inertia weight to optimize convergence speed and precision. The simulation results prove that APSO outperforms the conventional particle swarm optimization (PSO) algorithm by rapidly and precisely tracking the maximum power point (MPP) under uniform illumination and static or dynamic PSC. Moreover, the APSO exhibits fewer power fluctuations during the tracking process.

This paper uses the neural network to realize a photovoltaic (PV) system's maximum power point tracking. After Bayesian optimization of hyperparameters, the model can accurately determine the maximum power point voltage according to solar irradiance and temperature. The optimized PV system not only adapts to different operating conditions but also achieves the optimal power output under uniform irradiance, static irradiance shading, and dynamic shading, ensuring the maximum efficiency of the PV system. Moreover, the optimized model is superior to the unoptimized model and the traditional perturbation and observation method in terms of accuracy. The improved model enhances the practicality and reliability of the PV system, helping to improve PV efficiency and reduce energy loss.

In this paper, an MPC-based faster joint control method is proposed for hybrid energy storage system (HESS), which consists of battery and supercapacitor in photovoltaic dc-microgrid. The proposed method utilizes the uncompensated power from the battery to improve the dc-link restoration and decrease overshoot. Simulations are conducted to validate the robustness and rapidity of the proposed method, and the results are compared with traditional double-loop PI controllers. The comparison results demonstrate that the new controller has higher dynamic performance and better robustness.

This paper presents a controller, with heuristic intelligence, which focuses on optimizing the occupants’ thermal comfort in a building, by judiciously adjusting the temperature set-points during HVAC system operation. The selection of temperature set-points is formulated as an optimization problem, using suitable cost functions, wherein occupant specified comfort parameters are taken as constraints. The performance of the controller is examined by implementing the control algorithm on a thermal model of a real building, located in New Zealand, developed in EnergyPlus™. Simulation results are compared with that of a PI controller during two seasonal extremes of New Zealand, to further demonstrate the efficacy of proposed controller. The results of investigation show a consistent overall performance of the controller. An annual energy saving of 212.9MW h is achieved without compromising the occupants’ thermal comfort.

This article proposes an alternate approach following command-filtered backstepping (CFBS) principle to control the terminal voltage and stabilize the speed of a diesel generator through the excitation system. This controller eliminates the errors introduced due to differentiation of input signal and is, therefore, better compared to the conventional backstepping control technique. The global stability conditions of the overall closed-loop system are established using Lyapunov criteria. The parameters of the controller are optimized using a comprehensive learning particle swarm optimization technique. The performance of the proposed controller is compared with two other controllers, which include a classical proportional integral derivative (PID) and ΔΣ-based PID controllers. The performance comparison using percentage overshoot and settling time demonstrate that the proposed controller is superior and could effectively control the terminal voltage and stabilize the speed under various fault conditions.

In this research paper, the Multi Agent System (MAS) is introduced to Multi Micro- Grid (MMG) mesh system with the consensus control to share the power between arbitrary inverters to meet the load demand. Each microgrid consists with the Hybrid Energy Storage (HESS) which include the battery and Supercapacitor (SC) to supply/absorb the energy according to the load demand. The consensus based droop characteristics are used with MAS topology to share the power between different microgrids. The overall system consists with five microgrids and they interconnected as meshed network. The implemented control architecture achieve the DC voltage stability among all the microgrids.The system's stability being analysed mathematically using graph theory. MATLAB/Simulink virtual environment is used to simulate the overall system. The Java Agent Development Framework (JADE) is used as the platform to see the status of the agents. The overall simulation results substantiate in different modes of operation and compared with the conventional control method.

This paper investigates some issues associated with the power sharing in the Hybrid Energy Storage System (HESS) in the Multi Micro- Grid System (MMGS) to meet the load demands. To address this problem in isolated microgrids, which often arises in emergency situations, the present study proposes an efficient energy management system (EMS) which operates the battery and supercapacitor (SC) based on their State of Charge (SOC) level using fuzzy logic based control algorithm. The optimal amount of charging/discharging of the battery and SC is decided by the fuzzy inference system (FIS) . Simulations are carried out by creating a microgrid test bench for fuzzy logic system (FLS) in MATLAB/Simulink environment. The performance of the proposed approach is validated considering different modes of operation and loading conditions and found to be satisfactory.

This paper presents a solar-PV (photo-voltaic) system with wireless power transfer (WPT) and battery storage backup in various state changes throughout the day. A 378W PV module is connected with high-frequency converters at 20 kHz switching frequency and an LCL compensation network for the wireless power transfer across the primary and the pick-up coil. For the power backup, a 48V Li-ion battery is used for the storage. The voltage and current characteristics of the battery and IPT (Inductive Power Transfer) interface are investigated when the PV module is at STC(Standard Test Conditions). The simulation result shows that a complete analysis of wireless power transfer of PV power with battery storage can be achieved to form a robust and reliable system for low voltage(LV) applications. This paper also studies the effect of a variable load on the battery current and voltage respectively.

Several frequency-domain based fault detection methods were developed in the high voltage (HVDC) system in recent years. In this paper, a modified empirical mode decomposition (EMD) designated as median mode decomposition (MMD) is introduced, and a new fault detection method is developed based on that. Compared with the traditional EMD, MMD is able to obtain the intrinsic median mode function (IMMF) with less sifting iterations, leading to less computation burden in the actual use. This characteristic is useful for dc fault detection because the detection delay of dc fault should be limited within 3 ms. By obtaining the Hilbert spectrum of the IMMF after MMD, the fault transient is able to be differentiated within 0.35 ms detection delay with reliable discrimination. This method is verified by simulation results in a 4-bus ring multi-terminal dc (MTDC) system in PSCAD. It is also compared with other fault detection methods to test its validity.

Although the capability of digital microprocessors has been developed rapidly, the heavy computation burden is a challenge for the model predictive control on the applications of power electronics. This paper presents a zero iteration model predictive control (MPC) for a hybrid energy storage system (HESS) with dual batteries. A PI controller is used to predicting the closed-form battery current and control signals, which can be used for relax the constraints and find the active constants in order to obtain the solutions without iteration. The dual battery can share the power to ensure the charging/discharging behaviours are always limited to the specified active constraint.

Agent-based models (ABM) are a kind of micro scale model that imitate the simultaneous operations and interactions of multiple agents in an attempt to re-create and predict the appearance of complex process. Netlogo is a real time simulation software tool to design this model with the help of programming and coding. This paper identifies decision variables based on electric vehicles (EVs) charging statistics and the heuristic decisions in EVs charging at public charging stations, commercial place and offices are converted into constraints of (ABM). This unique model is the version of real time charging scenario at the charging stations. With the help of programmed model in Netlogo, the behaviour of EVs user under different real life scenarios are observed and recorded. The proposed system is implemented and designed in Netlogo to test the results.

Cyber-physical systems (CPSs) are implemented in many industrial and embedded control applications. Where these systems are safety-critical, correct and safe behavior is of paramount importance. Malicious attacks on such CPSs can have far-reaching repercussions. For instance, if elements of a power grid behave erratically, physical damage and loss of life could occur. Currently, there is a trend toward increased complexity and connectivity of CPS. However, as this occurs, the potential attack vectors for these systems grow in number, increasing the risk that a given controller might become compromised. In this article, we examine how the dangers of compromised controllers can be mitigated. We propose a novel application of runtime enforcement that can secure the safety of real-world physical systems. Here, we synthesize enforcers to a new hardware architecture within programmable logic controller I/O modules to act as an effective line of defence between the cyber and the physical domains. Our enforcers prevent the physical damage that a compromised control system might be able to perform. To demonstrate the efficacy of our approach, we present several benchmarks, and show that the overhead for each system is extremely minimal.

The present study applies reinforcement learning strategy for controller design to improve the low voltage ride through (LVRT) capability of a hybrid power system through convertible static compensator (CSC). This article considers different configurations of CSC such as static synchronous series compensator (SSSC), one static synchronous compensator (STATCOM), two STATCOMs, and unified power flow controller (UPFC). Both Q-learning and dynamic fuzzy Q-learning (DFQL)-based controllers are designed and their performances were compared with classical proportional-integral derivative (PID) controller considering a 3-machine system consisting of 2-synchronous and 1-wind energy systems. The results of simulation show that the performance of DFQL-based controller is better compared to other 2-controllers in improving the LVRT capability. Further, it is shown that the UPFC and two STATCOMs configurations of CSC provide higher voltage support compared to other configurations.

Heating, ventilating and air-conditioning systems have a significant share in the energy consumed by buildings. Modeling of room temperature dynamics is the first-step in designing an efficient air-conditioning system. In this regard, this paper proposes a semi-nonlinear thermal model: ordinary differential equation, with parameters as nonlinear functions of ambient temperature and cooling air flow-rate. To validate the performance of the model, a three-roomed building, equipped with an air-conditioning system is modeled, and Navier-Stokes equations are solved to simulate the temporal evolution of temperatures for different ambient temperatures and cooling air flow-rates. The steady-state temperature and transient solution parameters of the thermal model are assumed to be polynomial functions of ambient temperature and flowrate, and are determined by minimizing the errors between the thermal model- and the computational fluid dynamics-based solutions of final and transient temperatures, respectively. The proposed thermal model of third-order and nonlinear type transient coefficients is shown to predict the temporal evolution of temperature accurately. Further increase in prediction accuracy is achieved by recursively updating the parameters online using extended Kalman filter. The high prediction accuracy of the proposed thermal model makes it a potential candidate for the design of an optimal temperature regulator.

Widespread adoption of electric vehicles (EVs) would significantly increase the overall electrical load demand in power distribution networks. Hence, there is a need for comprehensive planning of charging infrastructure in order to prevent power failures or scenarios where there is a considerable demand-supply mismatch. Accurately predicting the realistic charging demand of EVs is an essential part of the infrastructure planning. Charging demand of EVs is influenced by several factors, such as driver behavior, location of charging stations, electricity pricing, etc. In order to implement an optimal charging infrastructure, it is important to consider all the relevant factors that influence the charging demand of EVs. Several studies have modeled and simulated the charging demands of individual and groups of EVs. However, in many cases, the models do not consider factors related to the social characteristics of EV drivers. Other studies do not emphasize on economic elements. This paper aims at evaluating the effects of the above factors on EV charging demand using a simulation model. An agent-based approach using NetLogo is employed in this paper to closely mimic the human aggregate behavior and its influence on the load demand due to charging of EVs.

A significant portion (ca. 40%) of the world's total energy is utilized in buildings. Cooling of large buildings worldwide is done by centrifugal chiller systems, where the compressors, pumps, and fans are driven by ac motors. These motors are energy hungry parts of the chiller. All-variable speed chillers, taking into account load variations, reduced the energy demand significantly compared to the constant speed ones. Induction motors (IMs) are commonly used in modern chillers. However, as the efficiency of the IM decreases with the speed, the overall efficiency drops, as the chiller operates most of the time in part-load condition. In this paper, SimulationX software is used to quantitatively investigate about the energy efficiency improvement by using synchronous reluctance motor (SynRM) in all-variable speed centrifugal chillers. The system was tested for three distinct building load profiles. The results are judged in comparison with the state-of-the-art IE2 IMs. The results provide quantitative evidence that SynRM can significantly increase the energy efficiency in the chiller systems and has a big potential to substitute IMs in these applications.

DC marine vessels with medium-voltage compact dc power systems are dominated by a significant amount of active loads and a finite number of generation sources. In such scenarios, the network configuration of the dc power system is expected to get dynamically altered to fulfill the required generation and load demands for the desired marine mission. Such varying network configurations make the transient responses significantly different from the conventional ac grids and the prospective dc grids. In this regard, this paper performs systematic transient studies to devise fault management strategies for the dc marine vessels. A platform supply vessel (PSV) is taken as an example of the marine vessel, due to its complex operating scenarios and wider applicability in the marine industry. Pole-to-pole short-circuit faults are considered owing to its severity. A novel current-only directional protection for the dc PSV is proposed based on the directional zonal interlocking and short-time Fourier transform. The efficacy of the proposed method is substantiated by confirming against a range of fault impedances initiated at the generator terminals, load terminals, lines, and buses of the dc PSV. All the analyses are conducted in the real-time simulation model of the dc PSV.

The maximum current carrying capacity of a power cable (Ampacity) is determined by the thermal characteristics of the cable components and surrounding medium in which they are buried. Power cable ampacity calculations are based on typical standard tables defined with predetermined parameters. In realtime, the environment of installation plays significant role in cable current carrying capacity. This paper presents FEM approach to determine the maximum current capacity within the safe operating limits of the cable by modeling electro-magnetic heat transfer. The study includes complex thermo-electric coupling and heat transfer in different zones surrounding the cable and its effect on the conductor operating temperature. The results show the difference between the proposed approach and standard calculations and improvements to accurate rating of the cables.

With the rapid demand in the power system, the renewable energy sources (RES) have led to the higher penetration in the existing main grid. The proper control strategy for energy storage systems (ESS) have received a huge attention. The ESS is critical to maintain the correct power balance between the RESs and load demand. In this research, fuzzy logic control algorithm is implemented among other existing control algorithms to operate the battery and supercapacitor (SC) based hybrid energy storage system (HESS) in an optimal way. The battery is used to compensate the energy requirement during long duration, while the supercapacitor handles mainly the transient power fluctuations. This increases the battery life cycle due to limited stress on the battery. The proposed fuzzy inference system decides the optimal amount of charging/discharging of the HESS based on the state of charge (SOC) constraints of the battery and supercapacitor. MATLAB/Simulink is used to create and implement the microgrid test bench and fuzzy logic system (FLS). The simulation results substantiate the potential of FLS in the microgrid energy management in different modes of operation and load conditions.

The energy management system (EMS) of microgrid deals with lots of complex issues, both from generation and demand side. The intermittent nature of renewable energy sources and constantly fluctuating load demands require robust energy management. This paper introduces the structure of hierarchical control strategies and common control approaches in the microgrid. In this paper, a novel enhanced EMS in islanding mode is presented for the hybrid energy storage system (HESS), comprising of two batteries and supercapacitors. A coordination control scheme is introduced at the primary level for the HESS. This will reduce the power stress on batteries, while improving the power quality.

Modern and future substations are aimed to be more interconnected, leveraging communication standards like IEC 61850-9-2, and associated abstract data models and communication services like generic object oriented substation event, manufacturing message specification, and sampled measured value. Such interconnection would enable fast and secure data transfer, sharing of the analytics information for various purposes like wide area monitoring, faster outage recovery, blackout prevention, distributed state estimation, etc. This would require strong focus on communication security, both at system level as well as at embedded device level. Although communication level security is dealt in IEC 62351, implementation attack on the embedded system is not considered. Since the embedded system makes the core of the smart grid, in this paper, we take a deeper look into impact of implementation attacks on substation security. An overview of potential exploits is first provided. This is followed by a case study, where implementation attacks like malicious fault injection attacks and hardware Trojan are used to compromise a substation level intelligent electronic device. The studied scenario extends implementation attacks beyond its usual exploit of confidentiality to affect power grid integrity and availability.

Electric vehicle (EV) charging stations will play an important role in the smart city. Uncoordinated and statistical EV charging loads would further stress the distribution system. Photovoltaic (PV) systems, which can reduce this stress, also show variation due to weather conditions. In this paper, a hybrid optimization algorithm for energy storage management is proposed, which shifts its mode of operation between the deterministic and rule-based approaches depending on the electricity price band allocation. The cost degradation model of the energy storage system (ESS) along with the levelized cost of PV power is used in the case of EV charging stations. The algorithm comprises of three parts: categorization of real-time electricity price in different price bands, real-time calculation of PV power from solar irradiation data, and optimization for minimizing the operating cost of EV charging station integrated with PV and ESS. An extensive simulation study is carried out with an uncoordinated and statistical EV charging model in the context of Singapore to check effectiveness of this algorithm. Furthermore, detailed analysis of subsidy and incentive to be given by the government agencies for higher penetration of renewable energy is also presented. This work would aid in planning of adoption of PV-integrated EV charging stations, which would expectedly replace traditional gas stations in future.

The rapid discharge of a dc-link capacitor of the voltage-source converter (VSC)-based dc system is the primary indication of the fault condition. Apart from the time-domain analysis, frequency-domain analysis of the fault current could also be utilized for dc fault detection, as the rapidly rising fault current is expected to have high-frequency components. This paper proposes two fault detection methods and compares their performances with the wavelet transform. The first method is the time-domain analysis of the dc-link capacitor discharge and is termed as the capacitive discharge technique. The relationship between the dc line current and the behavior of the dc-link capacitor is measured in terms of a correlation coefficient, whose value can be used to establish a fault basis. The second method is the frequency-domain-based short-time Fourier transform, which is used for quantitative analysis of high-frequency components in the fault current. These methods are extensively analyzed and compared using a scaled-down VSC-based dc system experimental test setup. Comparison has been done based on fault detection time, sensitivity to fault parameters, influence of sampling frequency, and computation speed. Furthermore, the selectivity of the fault detection methods is studied on the multiterminal dc systems of two different topologies (ring and radial), modeled in PSCAD/EMTDC. The experimental and simulation results substantiate the applicability of all the methods to the dc system. Brief comparative analysis with the di/dt method is also presented to highlight the advantages of the proposed methods.

Large scale adoption of electric vehicles (EVs) would significantly increase the overall electricity demand of the power distribution networks. Hence, there is a need for comprehensive planning of charging infrastructure in order to prevent power failures or scenarios where there is a considerable demand-supply mismatch. Accurately predicting the realistic charging demand of energy storage systems (ESS) used in EVs is an essential part of the infrastructure planning. Charging demand of ESS used in EVs is affected by several factors such as driver behavior, location of charging stations and electricity pricing. In order to implement the optimal charging infrastructure, it is important to consider all the crucial factors that affect the charging demand of ESS in EVs. Several studies have modelled and simulated the charging demand of individual as well as group of EVs. However, in many cases the models did not include factors that deal with the social characteristics of EV drivers, while the others did not emphasise on the economic elements. This paper aims to evaluate the effects of above factors on the EV charging demand using a simulation model. Agent-based approach using NetLogo is employed in this study to closely mimic the human crowd behaviour and its influence on the load demand due to charging of ESS used in EVs.

Stationary Battery Systems (SBS) are becoming a critical component in power distribution network across the world. Penetration of renewable energy sources which are intermittent in nature is a huge influence on the requirement of SBS. Furthermore, SBS are used in other applications such as peak load management, load-shifting, voltage regulation and power quality improvement. With increase in penetration on SBS, the requirement for modeling charging characteristics considering capacity loss is also increasing drastically. Minimal resource requirement and capability to leverage on smart meter data are the important parameters that are to be focused while developing any model for such applications. In this paper, an analysis on different curve fitting approaches that can be used for predicting the charging profiles of SBS based on lithium iron phosphate batteries is presented.

This paper presents a new approach for the determination of underground cable ampacity that considers surrounding medium of the cable, and compares the results from the FEM model with standard approach. It shows how ampacity of a buried cable is affected by the extent of heat transfer from the cable to the surrounding soil and also by the heterogeneity of the soil and its thermal characteristics. Numerical and finite element model of steady-state thermal analysis and ampacity evaluation are presented in this paper. COMSOL software is used for the three dimensional simulation of a 44kV armored HVAC XLPE cable buried directly in native soil. The methodology includes mathematical solutions for heat transfer equations to calculate and obtain the temperature at the cable surface and results show the optimal acceptable ampacity. The results from the proposed method provide solutions for ampacity problems that require flexibility and dynamic approach for real life scenarios that are not present in the literature and previous works.

In this paper, a new control approach is proposed for PV system with hybrid energy storage system (HESS) in isolated DC grid application. The proposed control approach solves the current controller conflict problem in HESS and provides faster DC link voltage restoration. In the proposed control approach a predictive term is used to control the battery current and the supercapacitor (SC) current. In the proposed control approach the voltage error term and the uncompensated power from the battery is added to the supercapacitor current reference to achieve faster DC link voltage restoration and less stress in the battery system. The system parameters design and closed loop system stability analysis of the proposed control approach are discussed in detail in the paper. The effectiveness of the proposed control approach is verified by simulation studies.

Integrated power systems (IPSs) with medium voltage direct current (MVDC) distribution are gaining importance in civil and defense marine vessels as they promise to provide cleaner, more reliable operation of vessels along with reduced fuel consumption. The integration of battery energy storage systems (BESSs) in the MVDC distribution system enables peak shaving of generators, optimal scheduling of generators, and near instantaneous power reserve. In this paper, an isolated bipolar dc-dc converter is presented which interfaces the BESS with the bipolar dc distribution bus. The proposed converter uses high frequency transformer isolation which helps in protecting the BESS against rapid discharge in the event of short circuit faults on the dc bus. Detailed steady-state analysis of the proposed converter is presented in this paper. A closed loop control system is developed in the MATLAB®/Simulink simulation platform which regulates the output of the isolated bipolar converter to its reference value. The dynamic performance of the converter is demonstrated with load variations on both the positive and negative buses. The battery discharge limiting capability is also verified.

In underground low-pressure gas pipelines, a leak may result in a complication more dangerous and difficult to detect, known as the water ingress problem. Groundwater enters the gas pipeline through a crack, and eventually blocks the gas flow. Vibration sensors are used to detect the water ingress problem. The sensor output indicates a marked increase in the occurrence of spikes from the vibration sensor once the water ingress starts. Since the signal is not much greater in magnitude than the baseline signal it is difficult to definitively detect. The estimation error from a Kalman filter is used to detect the occurrence of water ingress and with multiple sensors even the location of the water ingress can be detected.

From multiple industrial testing on chiller performance, refrigerant contaminants have been known as one of the main cause of efficiency loss. As the refrigerant is circulated throughout the refrigeration cycle, these efficiency loss causes great energy wastage and damage to equipment. This paper studies the possible impact of contaminants in a chiller system through the MATLAB simulations. The model block built by the external library namely, “Thermolib®“, the model consists of the compressor, evaporator, condenser and expansion valve. The refrigerant R134a was used throughout the simulations. The thermodynamic properties of the R134a such as enthalpy, heat transfer coefficient and pressure of the system were adjusted to simulate the presence of refrigerant contaminants in the system. The simulation model was able to produce the accurate results that corresponds with existing industrial standards.

Electricity price is essential factor in the deployment of electric vehicles (EVs) on large scale. In wholesale electricity market, EV charging stations(ECS) connected with suitably sized energy storage system (ESS) can save substantial amount of money by managing their time of utilisation (TOU). In this study, a real-time EV charging model at ECS along with ESS degradation model is considered to analyse effect of the ESS for TOU pricing benefits. The objective is to minimize the EVs charging cost. The proposed algorithm focuses on real-time energy management using combination of heuristic approach and deterministic approach. The algorithm analyses electricity pricing trend using historical and forecasted statistical data and controls the power imported from the grid, while feeding dynamic EV charging load. Significant amount of cost saving is seen in results due to deployment of ESS in utilizing TOU pricing benefits of wholesale electricity price.

This paper presents a phase shifted triangular carrier pulse width modulation technique for control of modular multilevel converter (MMC) based HVDC system. This switching modulation technique has reduced average switching frequency when compared to conventional method where switching state changes in each sampling period or control cycle. A dq reference frame based circulating current controller (CCC) is implemented to suppress the double line frequency arm circulating currents of MMC. Details regarding the outer loop power control and DC bus voltage control for transmitting desired amount of power through HVDC system are also discussed. A detailed matlab simulink model of eight submodules (SM) per arm MMC based HVDC system was developed and the corresponding simulation results are presented. Simulation studies with and without CCC are carried out and the corresponding variation in arm currents, SM capacitor voltages are also discussed.

A probabilistic method is proposed to detect and localize leaks in low-pressure gas distribution networks. These leakage events are estimated using flow and pressure information obtained from the steady state analysis of gas network. The approach provides an estimation of the leaked pipe section and the amount of gas outflow from the pipe section containing leaks. The reliability of the methodology is shown by analyzing the network in the presence of network modeling errors. These errors account for the variation in demand value of gas at the outlet nodes and the change in pipe roughness due to age. Even in the presence of large noise in the network, this methodology provides an accuracy of more than 80% in the localization of leak. The study aims to develop a real-time online monitoring system for low pressure gas distribution networks. Moreover, this technique is cost effective and can be easily integrated with the existing monitoring system.

Building energy management is a major R&D topic, as buildings account for about 32% of energy consumption worldwide. Design and implementation of controller for energy efficient operation in building is one of the methods to save the energy in the building. In this paper, a fuzzy logic based approach is used to design the energy management controller. Fuzzy logic has been widely used in the control engineering and many other fields of study. In the proposed approach, fuzzy logic is used to simulate the building load profile, and control the load of the appliance by flexibly choosing load scheduling. The controller would set a power threshold that cannot be exceeded, otherwise an on/off control of the home appliances would be invoked. In such a way, peak shaving of the power could be achieved. Using MATLAB-based graphical user interfaces, such fuzzy logic-based application could very well be implemented, reflecting on real-life scenarios.

Faults in a gas pipeline network is one of the major impairments towards the safe gas distribution among consumers. So, it is significant to detect and locate the faults in a pipeline network. In this work, a novel sensor based fault localization scheme, FLoGPN is proposed for detection and localization of faults in a gas distribution network. Here leak in gas pipeline network is considered as fault. This scheme uses Josang's Beta Reputation model to combine information from the sensors to select the nearest sensor to the fault. The aim is to identify the nearest sensor to the fault that indicates the faulty pipe segment in the whole network. The method is tested in the presence of noise to check its reliability and the simulation results show that the scheme can localize fault in a gas pipeline network accurately with the SNR value 20 dB or more.

Leak detection is vital in oil and gas pipelines, as it can cause financial loss and impact the environment, prove fatal to human life and also affect the effective functioning of domestic household. Acoustic emission test is a non-intrusive technique for leak detection. Leakage in pipes creates vibrations which are transmitted along the pipe walls. These waves can be detected by using acoustic sensors or accelerometers installed on the pipe wall for the purpose of analyses. In this work, a simulation study is carried where the leak location is defined by an Incident Pressure Field and this source propagates waves in the frequency range of about 20-200Hz. The results are verified by experimental work by introducing synthetic leak and disturbance signals. The Fast Fourier Transform (FFT) was used as the mathematical tool to evaluate the dominant peak in the frequency spectrum during an event of leak or disturbance. The experimental results obtained are in accordance with the simulated results. The accuracy of determining an event occurrence can be improved with error less than 5%.

Gas pipelines are the most vital and commonly used method of transportation of fuel, water, gas. Any form of leak in the pipeline can be catastrophic to mankind, incurring huge financial losses. In this paper, an experimental study has been carried out with respect to a real-time monitoring on the pressure and the flow variations. A small section of the pipeline with valves is considered to generate a synthetic leak in a pipeline. The regulation of these valves can be used to monitor the effects of leak on pressure and flow parameters of the town gas. Town gas is widely used in household and industrial applications in Singapore. A systematic test has been performed to analyze the irregularities due to leak. The performance and the magnitude of the signals will help in defining the effects due to leak location and valve operation.

DC power systems are increasingly being considered for marine applications as they provide certain advantages. Proper modeling and control approach of the system components with suitable dc bus voltage control and power sharing among the generators are important aspects for smooth functioning of the marine dc power system. Conventional bus voltage control in ac systems is usually achieved by using automatic voltage regulator (AVR), which requires the measurement of the generator terminal voltage. Since generators interfaced with the voltage source converter (VSC) system are envisaged in the marine dc power system, determination of generator terminal voltage for voltage control becomes difficult as the measurement is comprised primarily of PWM waveforms. This paper implements the closed loop flux estimation of brushless synchronous generator (BLSG) for dc bus voltage control. This paper also covers the droop control for power sharing among the generators. The proposed method is implemented in Opal-RT which is a real-time simulation platform and the results are discussed in the paper. Apart from the results, the limitations of the closed loop flux control are also highlighted in the paper.

DC shipboard power systems are gaining popularity primarily because they allow each prime mover to be operated at an optimal speed with respect to fuel efficiency. Another notable advantage is the simplified operation of generators in parallel without concern for frequency synchronization. Careful design of the protection system for dc shipboard power systems is necessary as the continuity of electrical power is required to carry out critical marine missions. In this paper the protection requirements for MVDC shipboard power system with multiple parallel generation system are discussed and the `Directional Protection' method is presented as an attractive candidate. The backup protection algorithm is also developed for increased redundancy. The protection settings for main and backup protection system are discussed and the results are presented. The various challenges associated with the directional protection are also highlighted.

In this paper, a unified adaptive energy management scheme (EMS) is proposed for renewable-interfaced hybrid energy storage system (HESS) under grid connected/islanded conditions. A second harmonic based phased locked loop is employed for effective synchronization/resynchronization of the microgrid system under contingency conditions. The operation and management of the microgrid system under both these modes are accomplished by an efficient adaptive power management algorithm. A quantitative analysis on the HESS performance is provided in order to investigate the effectiveness of the proposed approach. Load curtailment and off-maximum power point tracking features are also accommodated in the proposed scheme. This approach address seamless transfer between the various sub-modes of the system along with additional services such as power quality enhancement and effective power dispatch between various sources. The effectiveness of the proposed scheme is verified by both simulation and experimental investigations.

Under the `Smart Grid' initiative, future medium and low voltage distribution grids would incorporate renewable sources like rooftop solar, etc. This would require bi-directional power flow between the generation and the loads. For optimal fault isolation and distribution automation, directional protection would be of great advantage. However, traditional directional overcurrent relays utilize the reference voltage phasor for estimating the direction of the fault, requiring measurement of both current and voltage using respective sensors. This makes the directional overcurrent relays prohibitively costly than the non-directional type for distribution side utilizations. Novel current-only directional protection algorithms have been proposed and demonstrated. However, as the current-only approach relies on the prefault current direction as the polarizing quantity, if the direction of the prefault current changes during normal condition, the direction definitions (forward or reverse) would be vice versa. This is a problem for the current-only approach. In this paper, a signal processing-based solution is proposed, leveraging the fact that change in direction in prefault condition is associated with a phase angle change of 180°.

Water ingress problem caused by gas leaks in distribution gas pipeline is a rather thorny problem, which the gas companies typically encounter. Replacing the existing faulty pipeline is rather costly and troublesome. For gas pipelines, when leak happens, gas escapes from the pipelines into the free space or surrounding environment, and results in cooling of the nearby area due to the Joule-Thomson effect. Commercial distributed temperature sensing (DTS) system with appropriate operation method could be a solution as it can achieve a resolution of 0.05°C. The background knowledge, characteristic and measuring principle of DTS technique is introduced. As one of the principal composition for the natural gas, methane is selected for study and the temperature change during leak has been evaluated. For the pipe's gauge pressure of 2 kPa, when methane leaks through narrow crack, the temperature drops about 0.69°C which could be detected by the DTS system.

Low voltage ride-through (LVRT) is mandatory required in many recent revised grid codes and technical rules for distributed resources, including diesel engine generator, connected to the power systems. Research works for LVRT compliance and capability enhancement have been largely concentrated on to wind power generator. However, not much research works have been published about LVRT assessment of a diesel engine generator. Diesel engine generators are increasingly becoming important for power system and smart grid. This paper presents the development of LVRT capability curve and the assessment of LVRT compliance for a diesel engine generator. The possible methods for LVRT capability enhancement are briefly discussed in the paper as well.

High voltage DC (HVDC) system is important transmission technology, with several advantages over the AC transmission system. Transient analysis plays an important role in the power systems, in particular for the multi-terminal DC (MTDC) system. In DC systems, the DC line fault can result in extremely fast rising current compared to its AC counterpart due to absence of the line inductance. Therefore, the analysis of fast transients is of great interest in the HVDC system, in particular for MTDC system. In this paper, the application of Short Time Fourier Transform (STFT) in transient analysis of current signals in the MTDC system has been investigated. Because of uncertainty principle with the STFT, finer time resolution has been chosen at the expense of frequency resolution, allowing for faster detection. The CIGRE B4 multi-terminal DC grid system has been modeled in PSCAD/EMTDC, and transient events are simulated. The steady-state and the transient currents in the different terminals are monitored and analyzed using the STFT.

Stator winding inter-turn fault is a common fault in induction machines. A substantial number of works have been developed for detection of this fault; however, there are only a few works on fault parameter estimation and faulty phase identification. In particular, there have been no works dealing with estimation of fault loop resistance. This paper presents a method for estimating the fraction of shorted turns and the fault loop resistance and identifying the faulty phase under steady-state condition. It provides a closed-form solution and hence is applicable to online fault detection and identification. Simulation results demonstrate the effectiveness of the proposed method.

Heating, Ventilating and Air-conditioning systems have a significant share in the energy consumed by buildings. Development of a highly accurate thermal model is the first step in designing an efficient air-conditioning system. In this regard, this paper proposes a semi-nonlinear thermal model — ordinary differential equation with parameters as nonlinear functions of ambient temperature and cooling air flow-rate — to approximate the temperature solutions simulated using computational fluid dynamics (CFD). A three-roomed building, equipped with an air-conditioning system is modeled, and Navier-Stokes equations are solved to simulate the temporal evolution of temperatures for different ambient temperatures and cooling air flow-rates. The steady-state temperatures and transient solution parameters of the thermal model are assumed to be polynomial functions of ambient temperature and flow-rate, and are determined by minimizing the error between the thermal model- and the CFD-based final and transient temperature solutions, respectively. The proposed thermal model of third-order, with nonlinear type transient coefficients is shown to predict the temporal evolution of temperature accurately. The high prediction accuracy of the proposed thermal model makes it to be a potential candidate for the design of an optimal temperature regulator.

Low autocorrelation binary sequences (LABS) play important role in communication and radar applications, due to low aperiodic autocorrelation property. The asymptotic merit factor of LABS has been conjectured by Golay to be 12.32. The minimum energy level of LABS has been proven to be y or, respectively for even or odd sequence length N. In this paper, following the theoretical minimum energy level analysis, a new asymptotic merit factor value of 10.23 is estimated. Sequences of length 4 to 60, found by exhaustive search, strongly support this value. Furthermore, best found sequences of higher lengths (61 to 304) also conform to the asymptotic value of 10.23.

Building energy efficiency has increasing focus worldwide. The heating, ventilation, air-conditioning (HVAC) devices account for highest energy consumption in buildings in Singapore. Multipoint sensing information in the various scenario is critical to optimize the HVAC operation, and increase the energy efficiency of the air-conditioning. A software called FloVENT® is used for simulation of room models in different scenario. It calculates the temperature, airflow speed, pressure and other parameters in the mathematical way, providing the solutions for the models. The first experiment is done to find the effect of the multipoint sensing system due to the moving heat source. A testing heat source, represented by 11 visitors, is located in 5 different positions for the sensitivity test of the monitors. Another experiment simulates the dynamic situation of visitors coming in and leaving the room. The experiment results are analyzed for identifying the most sensitive sensor and the most optimal sensing point. The parameters of temperature change from the most optimal monitor can be used in the control system for improving the energy efficiency of the motors.

With the increasing demand of high reliability and safety of modern electric machines, failure prognosis becomes more and more important since it is efficient to increase reliability and reduce downtime cost. In this work, a model-based remaining useful life (RUL) prediction method is developed for induction motor with stator winding short circuit fault. The induction motor model with stator winding short circuit fault is introduced based on reference frame transformation theory. The winding short circuit fault is characterized by the fraction of short turns and the fault loop resistance. In this paper, the motor life is defined as the stator winding insulation life due to thermal stresses because from a thermal point of view, the stator winding insulation is the weakest part of induction motors. A particle filter method is used to realize unknown parameter estimation and RUL prediction. Simulation results are provided to validate the proposed method.

Modern civilization depends heavily on the consumption of electrical energy for industrial, commercial, agricultural, domestic and social purposes. However, for the current HVDC system, proper protection devices and logic are not yet as mature as the AC counterpart. This paper presents the fault analysis for the protection of the HVDC (65-765 kV range) grid, using PSCAD. Faults in the DC transmission line are analyzed. This paper also looks into the response of the system to each kind of faults. It is observed that the AC and DC faults have different signatures allowing us to tell them apart. The rise time of the fault current is presented here. Analysis of load changes is also done comparatively with the fault cases.

HVDC system has become practically mature over the years but it is still met with some protection issues which should be discussed, for example, the circuit breaker (CB) should be selective to not trip if the transient is temporary, such as overcurrent due to load change. This paper addresses the problem with identifying the type of faults in a HVDC system using wavelet transform (WT). The wavelet transform is proven to be able to capture the distinctive feature of the fault pattern, specifically fault current rising time and oscillation pattern, which are helpful to form a basis for the tripping decision. Three phase-to-ground fault and DC fault are of concern in this paper as their effects are the most detrimental to the system. The point-to-point HVDC system is simulated using PSCAD, and the simulation result is subsequently processed in MATLAB to perform the wavelet transform.

In chemometrics, data from infrared or near-infrared (NIR) spectroscopy are often used to identify a compound or to analyze the composition of a material. This involves the calibration of models that predict the concentration of material constituents from the measured NIR spectrum. An interesting aspect of multivariate calibration is to achieve a particular accuracy level with a minimum number of training samples, as this reduces the number of laboratory tests and thus the cost of model building. In these chemometric models, the input refers to a proper representation of the spectra and the output to the concentrations of the sample constituents. The search for a most informative new calibration sample thus has to be performed in the output space of the model, rather than in the input space as in conventional modeling problems. In this paper, we propose to solve the corresponding inversion problem by utilizing the disagreements of an ensemble of neural networks to represent the prediction error in the unexplored component space. The next calibration sample is then chosen at a composition where the individual models of the ensemble disagree most. The results obtained for a realistic chemometric calibration example show that the proposed active learning can achieve a given calibration accuracy with less training samples than random sampling.

This paper presents the development and implementation of a generalized backpropagation multilayer perceptron (MLP) architecture described in VLSI hardware description language (VHDL). The development of hardware platforms has been complicated by the high hardware cost and quantity of the arithmetic operations required in online artificial neural networks (ANNs), i.e., general purpose ANNs with learning capability. Besides, there remains a dearth of hardware platforms for design space exploration, fast prototyping, and testing of these networks. Our general purpose architecture seeks to fill that gap and at the same time serve as a tool to gain a better understanding of issues unique to ANNs implemented in hardware, particularly using field programmable gate array (FPGA). The challenge is thus to find an architecture that minimizes hardware costs, while maximizing performance, accuracy, and parameterization. This work describes a platform that offers a high degree of parameterization, while maintaining generalized network design with performance comparable to other hardware-based MLP implementations. Application of the hardware implementation of ANN with backpropagation learning algorithm for a realistic application is also presented.