The wind farm power loss due to wake interaction becomes more pronounced with a more compact farm layout. The steady-state wake model, for example, the Gaussian wake model, can help optimize the wake deflection and reduce the power loss. However, accurately estimating the farm-level ambient wind, like wind speed, direction, and turbulence intensity, is challenging, especially under time-varying inflow conditions. An artificial neural network-based (ANN) wind farm control strategy is proposed for real-time wake steering control. This algorithm only requires the measurement data of each turbine without estimating the farm’s ambient wind. Firstly, numerous steady-state simulation scenarios are designed by setting different ambient winds and yaw deflection. Then, these simulation measurement data and optimal yaw offsets are used to train the neural network as the farm control strategy. This method utilizes the nonlinear mapping capabilities and knowledge compression characteristics of neural networks. Finally, the effectiveness of the control strategy is validated in a dynamic simulation setting, which includes fluctuating wind direction in the inflow.

This article presents a solution to the output-feedback control problem of dc-dc converters with unknown independent output loads. The available measurements are limited to the voltages across the capacitors and the input sources. A general lossless model of the converter is considered, which eliminates the need for knowledge of parasitic elements such as the input side parasitic resistance. However, this poses a challenge for observer-based design due to the lack of detectability. To address this challenge, a dynamic full-information controller is designed, taking into account the control signal saturation of the system. To handle the presence of parametric uncertainty in the output dynamics, a constructive filtered transformation is proposed to convert the dynamics into a proper adaptive observer form. By utilizing the immersion and invariance technique, a reduced-order filter is derived, leading to a reduced-order observer. The parametric convergence is achieved with a weaker condition compared to the usual persistency of excitation requirement for converters with a single unknown load. The dynamic controller is reinforced in conjunction with the observer to overcome the detectability obstacle and uniform global asymptotic stability of the closed-loop system is guaranteed. The computation of filter dynamics is canceled in the control algorithm by redesigning the observer and updating the adaptation law with feedback from the regulation error. The performance of the proposed control algorithms is experimentally evaluated on a dc-dc Buck converter with unknown parasitic and output resistances.

This paper presents a design of a bidirectional DC- DC power electronic converter system enabling cyclic operation for a Reversible Solid Oxide Electrolysis Cell (RSO EC) stack for steam electrolysis. The cyclic operation of the RSOEC stack is investigated, and two different equivalent circuit models are presented for the mathematical representation of the stack's electrical dynamics: The well-established steady-state Resistive model and a novel Voigt model. From these, two combined mathematical models of the bidirectional Buck-Boost converter supplying the RSOEC stack are derived using the small-signal averaging technique. For tracking the cyclic output current reference to an RSOEC stack, two Proportional Integral Derivative with derivative Filter (PIDF) controllers are designed using the two combined mathematical models derived. Finally, the performances of the two PIDF controllers for the bidirectional DC-DC converter systems are compared and validated through simulations. The simulation results confirm the bidirectional Buck-Boost converter's ability to deliver cyclic bidirectional output current and demonstrate that the control tuned based on the Voigt mathematical representation of the RSOEC stack yields superior closed-loop performance in accordance with the control design requirements.

Finding an optimal schedule(s) for the buildings' energy equipment cluster is vital to realize sustainable development and energy-saving goals. However, high-impact uncertainties pose critical challenges in this regard. To relax these challenges, this paper develops an optimization model for operating the buildings' energy equipment cluster under uncertain predictions whose principal elements are the energy hub (EH) concept and the two-point estimate (TPE) method. The EH concept is used to find levels of the degree of freedom for optimization by modeling efficiently how multi-carrier energy resources and demands can be connected through this cluster of converters, conditioners, storage, and others. The TPE method is, however, used to improve the reliability and robustness of the model's predictions, leading to better decision-making under uncertainty. The TPE method integrates high-impact uncertainties related to multi-carrier energy prices and demands and the production capacity of renewable energy resources in optimization. The proposed optimization model has been applied to an industrial building, and its sufficiency and profitableness are examined in different scenarios.

This study presents a non-inverting transformer-less Boost converter with a higher voltage conversion ratio, resulting in quadratic growth of the output voltage with modest increments in the duty cycle rate. The continuous input current of this converter simplifies the input filter design and enhances fuel cells' lifetime and stability. Additionally, compared to existing architectures, this converter reduces the voltage and current stresses on semiconductor components, making it a promising alternative to other quadratic Boost converters. As a result, this topology appears to be a viable alternative for the remaining quadratic Boost converters. The inclusion of common-ground switches, which eliminate the need for additional components for gate driver isolation, is one of the main features that make this topology distinct. As a result, the converter fabrication cost and size are reduced. The proposed converter's steady-state analysis is thoroughly explained. Stresses placed on semiconductor devices. Then, a meaningful comparison between the proposed circuit and its counterparts is provided to understand this converter's unique attributes.

This article presents a predictive home energy management system (HEMS) for a residential building with integration of a plug-in electric vehicle (PEV), a photovoltaic array, and a heat pump. A stochastic model predictive control (MPC) strategy is applied in the HEMS in order to minimize the home's electricity cost and reduce the PEV battery degradation cost. Moreover, the MPC ensures that home load demand, PEV battery charging requirements, and household thermal comfort conditions are met. The MPC operates in real-time and thus minimizes the effects of gap between the forecasted and real data on the HEMS performance by updating its control decisions and the forecast data as the stochastic parameters are realized in each time step. The obtained simulation results demonstrate that the proposed control strategy reaches 96% to 97% of ideal performance achieved by off-line optimization counterpart with perfect data.

An effective fault diagnosis scheme can improve system's safety and reliability. Artificial Intelligence (AI) provides a good framework to deal with this issue. Deep learning is a successful implementation of AI that its superior isolation performance find its way in fault diagnosis area. In this study, based on feature extraction abilities of Convolutional Neural Network (CNN), a deep network have been developed in order to isolate different kinds of faults in Tennessee Eastman process. This network has an end-to-end structure with 13 layers that takes raw sensor's data and has isolation performance of more than 98 percent. A comparison between our proposed method and a linear classifier that uses Principal Component Analysis(PCA) for feature extraction and a Neural Network (NN) with 2 hidden layers as nonlinear classifier have been conducted to show the performance of the proposed fault isolation scheme.

Obtaining an appropriate model is very crucial to develop an efficient energy management system for the smart home, including photovoltaic (PV) array, plug-in electric vehicle (PEV), home loads, and heat pump (HP). Stochastic modeling methods of smart homes explain random parameters and uncertainties of the aforementioned components. In this paper, a concise yet comprehensive analysis and comparison are presented for these techniques. First, modeling methods are implemented to find appropriate and precise forecasting models for PV, PEV, HP, and home load demand. Then, the accuracy of each model is validated by the real measured data. Finally, the pros and cons of each method are discussed and reviewed. The obtained results show the conditions under which the methods can provide a reliable and accurate description of smart home dynamics.

In order to reduce the levelised cost of energy (LCOE) of the offshore wind farm, many optimization works should be done to reduce the investment and increase the energy production. As one of the main expenses, the electrical system can take up more than 15% of the total investment while cable costs take a large proportion. In order to make a cost-effective wind farm, the cable connection layout should be optimized. This paper proposes a novel way for offshore wind farm cable connection layout design. The LCOE, which concerns three aspects: electrical power losses, power captured by wind turbines (WT) and investment, is selected to set up the objective function. Since all the optimization variables are integers, a heuristic algorithm, integer particle swarm optimization algorithm (IPSO), is adopted to find a near optimal solution. To improve the performance of the IPSO, an adaptive method for parameter control is used to help to find a better solution. Comparisons are made with results obtained by the Norwegian center for offshore wind energy (NORCOWE) reference wind farm and the presented method. From the simulation, it can be noticed that the presented approach can help to find a cable connection scheme which can reduce the LCOE by 1.75%.

In this paper, a novel single-stage single-phase transformer-less buck-boost inverter is proposed, in which a reduced number of passive components is used. The proposed inverter combines the conventional buck, boost, and buck-boost converters in one converter in order to obtain a sinusoidal output voltage. In the proposed inverter, the input DC source and the load or grid have the same ground. Therefore, the leakage current problem in photovoltaic (PV) systems is eliminated. Furthermore, the proposed inverter supports the bi-directional power flow capability and it can inject reactive power to the load or the grid. The analysis, principle of operation, and modulation of the proposed inverter are introduced in this paper, where simulation results using MATLAB/Simulink models in both islanded and grid-tied modes of operation are shown.

Due to the increasing size of offshore wind farm, the impact of the wake effect on energy yields become more and more evident. The seafloor topography would limit the layout of the wind farm so that irregular layout is usually adopted in large scale offshore wind farm. However, the calculation for the energy yields in irregular wind farm considering wake effect would be difficult. In this paper, a mathematical model which includes the impacts of the variation of both wind direction and velocity on wake effect is established. Based on the wake model, a binary matrix method is proposed for the energy yields calculation for irregular wind farms. The results show that the proposed wake model is effective in calculating the wind speed deficit. The calculation framework is applicable for energy yields calculation in irregular wind farms.

With the increasing integration of the wind power into the power system, wind farm are required to be controlled as a single unit and have all the same control tasks as conventional power plants. The wind farm central controller receives control orders from Transmission System Operator (TSO), then dispatch the wind power reference to each wind turbine. One of the most commonly used dispatch methods is to dispatch the wind power reference to each wind turbine proportional to each wind turbine's available wind power without the consideration of the wake effect. The wake which depends on the thrust efficient of upstream wind turbines in the wind farm influences the downstream wind speed which determines the available wind power of the downstream wind turbine. Optimize the wind power production of each wind turbine in the wind farm by the optimization of the pitch angle and tip-speed-ratio of each turbine can increase the total available wind power of the wind farm. This paper proposed a new dispatch method which dispatches the wind power reference to each wind turbine proportional to each wind turbine's optimal wind power to increase the available wind power of the whole wind farm. Particle Swarm Optimization (PSO) is used to obtain the optimal wind power for each wind turbine. A case study is carried out. The available wind power of the wind farm was compared between the traditional dispatch method and the proposed dispatch method with the consideration of the wake effect.