Optimising residential electric vehicle charging under renewable energy: Multi‐agent learning in software simulation and hardware‐in‐the‐loop evaluation

The integration of intermittent renewable energy sources coupled with the increasing demand of electric vehicles (EVs) poses new challenges to the electrical grid. To address this, many solutions based on demand response have been presented. These solutions are typically tested only in software‐based simulations. In this paper, we present the application in hardware‐in‐the‐loop (HIL) of a recently proposed algorithm for decentralised EV charging, prediction‐based multi‐agent reinforcement learning (P‐MARL), to the problem of optimal EV residential charging under intermittent wind power and variable household baseload demands. P‐MARL is an approach that can address EV charging objectives in a demand response aware manner, to avoid peak power usage while maximising the exploitation of renewable energy sources. We first train and test our algorithm in a residential neighbourhood scenario using GridLAB‐D, a software power network simulator. Once agents learn optimal behaviour for EV charging while avoiding peak power demand in the software simulator, we port our solution to HIL while emulating the same scenario, in order to decrease the effects of agent learning on power networks. Experimental results carried out in a laboratory microgrid show that our approach makes full use of the available wind power, and smooths grid demand while charging EVs for their next day's trip, achieving a peak‐to‐average ration of 1.67, down from 2.24 in the baseline case. We also provide an analysis of the additional demand response effects observed in HIL, such as voltage drops and transients, which can impact the grid and are not observable in the GridLAB‐D software simulation.     

FPGA-based reconfigurable control for switch fault tolerant operation of WECS with DFIG without redundancy

This paper deals with reconfigurable back-to-back converter topology and control orders in Wind Energy Conversion Systems (WECS). A typical WECS with Doubly Fed Induction Generator (DFIG) in balanced conditions is concerned. Based on the classical topology, a fault tolerant converter without any redundancy has been studied. The presented fault tolerant topology allows a “five-leg” structure with converters reconfiguration after switch failure detection. Furthermore, the control strategy for classical topology can no longer be applied after fault occurrence. Thus, a “five-leg” control strategy has also been proposed. The validation of the reconfigurable digital controller for the studied WECS with DFIG topology has been performed using a Hardware-in-the-Loop (HIL) reconfigurable platform including a Field Programmable Gate Array (FPGA) chip. HIL simulation results in both healthy and fault conditions have been presented to show simultaneously the viability of the studied converters topology and the reconfigurable control.     

Design and energy efficiency analysis of a pure fuel cell vehicle for Shell eco racer

The target of Shell Eco‐marathon competition of vehicle is to drive a fixed distance with the lowest quantity of fuel. To win the competition, the fuel cell‐powered propulsion system needs to be ultra efficient since the fuel cell system and transmission system are the key effects on the performance of the fuel cell‐powered propulsion system. In this study, a high‐efficiency fuel cell propulsion system has been designed and integrated in a prototype vehicle to participate the Shell Eco‐marathon Asia 2018 race. To achieve that, the vehicle dynamic is modeled to make the selection of the key components, and some experiments have been conducted to obtain the properly vehicle driving strategy. Based on the results of vehicle dynamic analysis, a high specific power proton‐exchange membrane fuel cell (PEMFC) stack with 1000 W and a high‐performance direct current (DC) brushless motor (1000 W) are selected to build the propulsion system of the Shell Eco‐marathon vehicle. Based on the experimental result, the racing time (1300‐1440 seconds) and varied range of racing speed (23‐27 km/h) are selected as the driving strategy. Finally, the efficiency of the fuel cell‐powered vehicle is analyzed. In the race at the year of 2018, the designed vehicle won the first place.     

Linear individual pitch control design for two‐bladed wind turbines

In this article, the conventional individual pitch control (IPC) strategy for wind turbines is reviewed, and a linear IPC strategy for two‐bladed wind turbines is proposed. The typical approach of IPC for three‐bladed rotors involves a multi‐blade coordinate (MBC) transformation, which transforms measured blade load signals, i.e., signals measured in a rotating frame of reference, to signals in a fixed non‐rotating frame of reference. The fixed non‐rotating signals, in the so‐called yaw and tilt direction, are decoupled by the MBC transformation, such that single‐input single‐output (SISO) control design is possible. Then, SISO controllers designed for the yaw and tilt directions provide pitch signals in the non‐rotating frame of reference, which are then reverse transformed to the rotating frame of reference so as to obtain the desired pitch actuator signals. For three‐bladed rotors, the aforementioned method is a proven strategy to significantly reduce fatigue loadings on pitch controlled wind turbines. The same MBC transformation and approach can be applied to two‐bladed rotors, which also results in significant load reductions. However, for two‐bladed rotors, this MBC transformation is singular and therefore, not uniquely defined. For that reason, a linear non‐singular coordinate transformation is proposed for IPC of two‐bladed wind turbines. This transformation only requires a single control loop to reduce the once‐per‐revolution rotating blade loads (‘1P’ loads). Moreover, all harmonics (2P, 3P, etc.) in the rotating blade loads can be accounted for with only two control loops. As in the case of the MBC transformation, also the linear coordinate transformation decouples the control loops to allow for SISO control design. High fidelity simulation studies on a two‐bladed wind turbine without a teetering hub prove the effectiveness of the concept. The simulation study indicates that IPC based on the linear coordinate transformation provides similar load reductions and requires similar pitch actuation compared with the conventional IPC approach. Copyright © 2014 John Wiley & Sons, Ltd.     

A grey‐box model of next‐day building thermal load prediction for energy‐efficient control

Accurate building thermal load prediction is essential to many building energy control strategies. To get reliable prediction of the hourly building load of the next day, air temperature/relative humidity and solar radiation prediction modules are integrated with a grey‐box model. The regressive solar radiation module predicts the solar radiation using the forecasted cloud amount, sky condition and extreme temperatures from on‐line weather stations, while the forecasted sky condition is used to correct the cloud amount forecast. The temperature/relative humidity prediction module uses a dynamic grey model (GM), which is specialized in the grey system with incomplete information. Both weather prediction modules are integrated into a building thermal load model for the on‐line prediction of the building thermal load in the next day. The validation of both weather prediction modules and the on‐line building thermal load prediction model are presented. Copyright © 2008 John Wiley & Sons, Ltd.     

Mirror rearrangement optimization for uniform flux distribution on the cavity receiver of solar parabolic dish concentrator system

The nonuniform and high‐gradient solar radiation flux on the absorber surface of solar dish concentrator/cavity receiver (SDCR) system will affect its operational reliability and service lifetime. Therefore, homogenization of the flux distribution is critical and important. In this paper, 2 mirror rearrangement strategies and its optimization method by combining a novel ray tracing method and the genetic algorithm are proposed to optimize the parabolic dish concentrator (PDC) so as to realize the uniform flux distribution on the absorber surface inside the cavity receiver of SDCR system. The mirror rearrangement strategy includes a mirror rotation strategy and mirror translation strategy, which rotate and translate (along the focal axis) each mirror unit of the PDC to achieve multipoint aiming, respectively. Firstly, a correlation model between the focus spot radius and mirror rearrangement parameters is derived as constraint model to optimize the PDC. Secondly, a novel method named motion accumulation ray‐tracing method is proposed to reduce the optical simulation time. The optical model by motion accumulation ray‐tracing method and optimization model of SDCR system are established in detailed, and then, an optimization program by combining a ray‐tracing code and genetic algorithm code in C++ is developed and verified. Finally, 3 typical cavity receivers, namely, cylindrical, conical, and spherical, are taken as examples to fully verify the effectiveness of these proposed methods. The results show that the optimized PDC by mirror rearrangement strategies can not only greatly improve the flux uniformity (ie, reduce the nonuniformity factor) and reduce the peak local concentration ratio of the absorber surface but also obtain excellent optical efficiency and direct useful energy ratio. A better optimization results when the PDC is optimized by mirror rotation strategy at aperture radius of 7.0 m, focal length of 6.00 m, and ring number of 6; the nonuniform factor of the cylindrical, conical, and spherical cavity receivers is greatly reduced from 0.63, 0.67, and 0.45 to 0.18, 0.17, and 0.26, respectively; the peak local concentration ratio is reduced from 1140.00, 1399.00, and 633.30 to 709.10, 794.00, and 505.90, respectively; and the optical efficiency of SDCR system is as high as 92.01%, 92.13%, and 92.71%, respectively. These results also show that the dish concentrator with same focal length can match different cavity receivers by mirror rearrangement and it can obtain excellent flux uniformity.     

Performance enhancement of direct methanol fuel cell using multi‐zone narrow flow fields

New serpentine and spiral flow field configurations were developed to enhance the performance of direct methanol fuel cells (DMFCs). The new configurations are based on two primary concepts, namely, narrowing the flow field and partitioning the total active area of the fuel cell. Three flow channel heights of 0.8, 0.4, and 0.2 mm were investigated in serpentine and spiral flow fields. The main active area is considered a single zone and is partitioned into two‐ and four‐zone designs while maintaining the total inlet mass flow rate of the reactant and oxidant. To determine the performance parameters of the newly proposed designs, a three‐dimensional single‐phase isothermal model was developed, numerically simulated, and validated through experimental measurements. The findings of the current study indicate that a serpentine flow field configuration with a channel height of 0.2 mm and two zones attains an enhancement of the net power density of 37% compared to a conventional single‐zone design with a flow channel height of 0.8 mm. Similarly, for a spiral flow field design, the maximum net power density increased by 26% using a two‐zone configuration with a channel height of 0.2 mm, in comparison to the conventional design of a single‐zone and a flow channel height of 0.8 mm. The newly developed designs utilize the lower height of the flow fields to decrease the dimensions of the fuel cell stacks and reduce the material costs required.     

Optimization of the coupling of pressurized water nuclear reactors and multistage flash desalination plant by evolutionary algorithms and thermoeconomic method

Thermodynamic simulation programs are widely used for designing complex thermal systems, but most of them do not incorporate second law optimization techniques. In this study, an efficient optimization strategy is presented, which integrates three optimization techniques with a professional power plant and a cogeneration simulator so as to perform exergoeconomic optimization of complex thermal systems and generate combined pinch and exergy representations. This paper deals with the application of an evolutionary algorithm based on NSGA‐II to multi‐objective thermoeconomic optimization of coupling desalination plant with pressurized water reactor (PWR). In addition, one‐objective thermoeconomic optimization through genetic algorithm and mixed integer non‐linear mathematical programming methods has been applied for evaluation of multi‐objective optimization. The thermodynamic simulation of this plant has been performed in the THERMOFLEX simulator. An Excel Add‐in called THERMOFLEX link has been developed to calculate the exergy of each stream from THERMOFLEX simulation results. In addition, a computer code has been developed for thermoeconomic and improved combined pinch–exergy analysis in the MATLAB environment. Also, multi‐objective and one‐objective evolutionary algorithm optimization has been performed in MATLAB and one‐objective mathematical programming has been performed in LINGO software. Both the design configuration and the process variables are optimized simultaneously. The optimization algorithm can choose among several design options included in a superstructure of the feed water heaters and multistage flash desalination in a dual‐purpose plant. For the assumptions and simplifications made in this study, a 3000 MWh PWR power plant similar to Bushehr power plant has been considered. Copyright © 2008 John Wiley & Sons, Ltd.     

Effect of air gap on the performance of building-integrated photovoltaics

Ventilation of photovoltaic (PV) modules installed over or beside a building envelope can reduce the module temperature and increase the electrical conversion efficiency. A computational fluid dynamics (CFD) method has been used to assess the effect of the size of air gap between PV modules and the building envelope on the PV performance in terms of cell temperature for a range of roof pitches and panel lengths and to determine the minimum air gap that is required to minimise PV overheating. It has been found that the mean PV temperature and the maximum PV temperature associated with hot spots decrease with the increase in pitch angle and air gap. The mean PV temperature also decreases with increasing panel length for air gaps greater than or equal to 0.08 m whereas the maximum PV temperature generally increases with panel length. To reduce possible overheating of PV modules and hot spots near the top of modules requires a minimum air gap of 0.12–0.15 m for multiple module installation and 0.14–0.16 m for single module installation depending on roof pitches.     

A 500 W low-temperature thermoelectric generator: Design and experimental study

Most of the current thermal power-generation technologies must first convert thermal energy to mechanical work before producing electricity. In this study, a direct heat to electricity (DHE) technology using the thermoelectric effect, without the need to change through mechanical energy, was applied to harvest low-enthalpy thermal work. Such a power generation system has been designed and built using thermoelectric generator (TEG) modules. Experiments have been conducted to measure the output power at different conditions: different inlet temperature and temperature differences between hot and cold sides. TEG modules manufactured with different materials have also been tested. The power generator assembled with 96 TEG modules had an installed power of 500 W at a temperature difference of around 200 °C. An output power of over 160 W has been generated with a temperature difference of 80 °C. The power generated by the thermoelectric system is almost directly proportional to the temperature difference between the hot and the cold sides. The cost of the DHE power generator is lower than that of photovoltaics (PV) in terms of equivalent energy generated.     

Master‐slave game optimization method of smart energy systems considering the uncertainty of renewable energy

As the uncertainty of renewable energy output brings more and more risks to the day‐ahead dispatch of the power grid, an optimization scheduling strategy of a smart energy system based on improved master‐slave game model is proposed. Risk factors related to the uncertainties of renewable energy are introduced into the master‐slave game model. Taking the smart energy system as the leader and the end users as the follower, an optimized operation model of the smart energy system based on the improved master‐slave game model is established, which is transformed into a single‐layer linear programming model according to the Karush‐Kuhn‐Tucher conditions and the duality theorem. The benefits of the system and electric vehicle users in four application scenarios are obtained by the YALMIP algorithm and the sensitivity affecting the economics of the smart energy system is analyzed. The validity of the model is verified by a simulation analysis of actual operation data from the smart energy system in China. The simulation results show that the method proposed in this paper can increase the revenue of the smart energy system by 7%, reduce the risk cost and charging cost of electric vehicle users by 63.92% and 48.34%.     

Energy analysis of a low‐temperature heat pump heating system in a single‐family house

Energy costs and environmental concerns have made energy optimisation a viable option for buildings. Energy‐efficient heating systems together with an effective use of buildings thermal mass and tightness have a significant impact on the energy requirement and on the possibility for sizeable running cost savings. In this study we use the simulation tool TRNSYS‐EES to model and analyse the performance of a residential house and the low‐temperature heating system that serves its thermal needs. The building is a single‐family house with controlled ventilation and the chosen heating system is a hydronic floor heating system connected to an exhaust air heat pump. The aim of the simulation is to study the performance of the building, the heating system and the controls in an integrated manner. Overall, the results indicate that the energy efficiency issue implicates system design and system thinking concerns as well as techno‐economic difficulties. The controls and the choice of the operation mode are of a great importance. Copyright © 2004 John Wiley & Sons, Ltd.     

共轨柴油机仿真系统及ECU的硬件在环仿真

针对共轨柴油机的特点设计了油轨压力与转速输出的简单数学模型,并且建立了基于PowerPC的共轨柴油机仿真系统代替柴油机进行硬件在环仿真,从而辅助发动机控制单元(ECU)的设计。共轨柴油机仿真系统直接测量ECU的喷油输出脉冲宽度和油轨供油提前角,据此计算发动机转速和油轨油压,输出曲轴和凸轮信号。使用共轨柴油机仿真系统和ECU控制器进行联合仿真,可以确保控制策略的有效性。     

Design and real time implementation of type-2 fuzzy vector control for DFIG based wind generators

Doubly fed induction generator is very sensitive to voltage variations in the grid, which pose limitation for wind power plants during the grid integrated operation. Handling the uncertainity in wind speed and grid faults is a major challenge to fulfill the modern grid code requirements. This paper proposes a new control strategy for Rotor side converter using Interval type-2 fuzzy sets which can model and handle uncertainties in the system parameters. The presence of third dimension in the membership function, offers an additional degree of freedom in the design of the controller to counter the effects of fluctuations in wind speed and low voltage during severe grid fault conditions. A 2 MW DFIG connected to the grid is modelled in simulation software RSCAD and interfaced with Real time digital simulator (RTDS) to perform the simulations in real-time. The RTDS platform is considered by many research laboratories as real-time testing module for controller prototyping and also for hardware in the loop (HIL) applications. The controller performance is evaluated in HIL configuration, by performing the real-time simulations under various parameter uncertainties. The proposed controller can improve the low voltage ride through capability of DFIG compared to that of PI and type-1 fuzzy controller.     

Real time optimal energy management strategy targeting at minimizing daily operation cost for a plug-in fuel cell city bus

This paper proposes an optimal real-time energy management strategy targeting at daily operation optimization for a plug in proton exchange membrane fuel cell electric vehicle (PFCEV) for public transportations. A novel real-time optimal energy management strategy based on the determined dynamic programming (DDP) strategy is proposed, namely the DBSD (charge Depleting – Blended – Sustaining – Depleting) strategy. A simulation model is set up to compare the DDP strategy, the DBSD strategy and the CDCS (Charge Depleting and Charge Sustaining) strategies. Compared to the CDCS strategy, the daily operating cost can be reduced by 6.4% with the DBSD strategy, and it can be reduced by 9.5% with the DDP strategy. On-road testing with the DBSD strategy shows that, the daily operation cost is 510.2 Sig. $ (100 km)⁻¹. The electric energy consumption in pure battery driven mode is about 1.68 kWh km⁻¹, and the equivalent hydrogen consumption in hybrid driven mode is about 0.14 kg km⁻¹.     

Optimizing 100%‐renewable grids through shifting residential water‐heater load

A low‐carbon electricity supply for Australia was simulated, and the installed capacity of the electrical grid was optimized by shifting the electricity demand of residential electric water heaters (EWHs). The load‐shifting potential of Australia was estimated for each hour of the simulation period using a nationwide aggregate EWH load model on a 90 × 110 raster grid. The electricity demand of water heaters was shifted from periods of low renewable resource and high demand to periods of high renewable resource and low demand, enabling us to effectively reduce the installed capacity requirements of a 100%‐renewable electricity grid. It was found that by shifting the EWH load by just 1 hour, the electricity demand of Australia could be met using purely renewable electricity at an installed capacity of 145 GW with a capacity factor of 30%, an electricity spillage of 20%, and a generation cost of 15.2 ¢/kWh. A breakdown of the primary energy sources used in our scenario is as follows: 43% wind, 29% concentrated solar thermal power, and 20% utility photovoltaic. Sensitivity analysis suggested that further reduction in installed capacity is possible by increasing the load‐shifting duration as well as the volume and insulation level of the EWH tank.     

Conceptual study of an accelerator‐driven ceramic fast reactor with long‐term operation

To improve both safe operation and high resource utilization in nuclear power, we propose and investigate the concept of an accelerator‐driven ceramic fast reactor (ADCFR). This reactor type has the potential to operate continuously throughout a 40‐year core life, without fuel shuffling or supplementation. The ADCFR consists of a high‐power superconducting linear accelerator, a gravity‐driven dense granular spallation‐target, and a ceramic fast reactor. The performance of the ADCFR was assessed by using a neutron‐physics simulation, thermal calculations, and a characteristic analysis. The results show that the peak position for the neutron spectrum in the ADCFR is at about 0.1 MeV. This means that it falls with the fast neutron spectrum, and it can convert loaded nuclear fertile material into fissile fuel. Using a burnup simulation, the ideal effective multiplication‐factor (K_eff) was calculated by using a combination of subcritical (accelerator‐driven) and critical modes. In 40 year of operation, K_eff is obtained from the initial 0.98 to the peak ~1.02 and then to ~0.99. Different granular coolant materials were selected to compare neutron performance. In breeding, the differences are relatively small. The thermal calculation indicates that heat transfer performance of granular makes it possible to meet the required specifications in theory. Finally, the corresponding characteristics, with regard to the 2‐phase coolant, ceramic materials, nuclear safety performance, operation modes, economics, and range of applications were analyzed. Accelerator‐driven ceramic fast reactors can achieve very high levels of inherent safety, good breeding performance, high power‐generation efficiency, and high flexibility in wide range of applications.     

Structure regulation and photocatalytic activity enhancement of Bi2WO6 by “double‐effect modification” mode

In this paper, the Cl‐doped hollow microspherical Bi₂WO₆ nano‐photocatalyst was self‐assembly synthesized creatively by one‐step solvothermal method using the surface charge regulation strategy for electrostatic adsorption effect of anion in KCl and cation in polyvinylpyrrolidone (PVP). The crystal structure, morphology, element distribution, chemical state, and optical characteristic were, respectively, characterized. The reasons for improving the photocatalytic activity were investigated by degrading tetracycline hydrochloride (TCH). Combined with characterization analysis and the density functional theory (DFT theory), the “double‐effect modification” mode under the surface charge control strategy can not only successfully dope the Cl^‐ to achieve functionalization of the product, but also achieve effective control of the product morphology through electrostatic self‐assembly process. The electron hole pair recombination was further inhibited, and the photocatalytic activity was improved. The proposed mode provides a simpler, more efficient and widely applicable method for the modification of Bi₂WO₆ to enhance its photocatalytic activity.     

Different diode models comparison using Lambert W function for extracting maximum power from BIPV modules

Perfect modeling of the building‐integrated photovoltaic (BIPV) module circuit equivalent is required for examining the operation of a BIPV system. Before designing the power electronic converters of an overall BIPV system, a perfect diode modeling is required that usually resembles the I‐V and P‐V characteristics of the BIPV modules. In this research article, different types of diode modeling of BIPV systems along with their comparative analysis based on the Lambert W function in MATLAB/Simulink environment is presented. The main aim of this research article is to analyze Mprime transparent M 115‐130P‐FI BIPV modules (115 Wp and 130 Wp) and compare the existing diode models in terms of accuracy and extraction of unknown parameters for the same. Simulation results for Lambert W function based comparison of the five‐parameter model (FPM), seven‐parameter model (SPM) and nine‐parameter model (NPM) power of 115 Wp and 130 Wp BIPV modules along with their percentage errors are well presented. Lambert W function based comparison of FPM, SPM, and NPM is further made at different values of irradiations and temperatures respectively.     

System analysis of a multifunctional PV/T hybrid solar window

The work presented in this article aims to investigate a PV/T hybrid solar window on a system level. A PV/T hybrid is an absorber on which solar cells have been laminated. The solar window is a PV/T hybrid collector with tiltable insulated reflectors integrated into a window. It simultaneously replaces thermal collectors, PV-modules and sunshade. The building integration lowers the total price of the construction since the collector utilizes the frame and the glazing in the window. When it is placed in the window a complex interaction takes place. On the positive side is the reduction of the thermal losses due to the insulated reflectors. On the negative side is the blocking of solar radiation that would otherwise heat the building passively. This limits the performance of the solar window since a photon can only be used once. To investigate the sum of such complex interaction a system analysis has to be performed. In this paper results are presented from such a system analysis showing both benefits and problems with the product. The building system with individual solar energy components, i.e. solar collector and PV modules, of the same size as the solar window, uses 1100 kW h less auxiliary energy than the system with a solar window. However, the solar window system uses 600 kW h less auxiliary energy than a system with no solar collector.