Search

Abstract The virtual synchronous generator (SG) (VSG) can not only enhance the inertia of the grid, but also introduce the oscillation characteristics of SG, which is easy to interact with the power angle of SG in the grid, and even produce low‐frequency oscillation (LFO). The authors first construct a two‐machine interconnected power system model containing VSG and traditional SG. The model is linearised to construct the state space equations to obtain the Phillips–Heffron model with VSG. The LFO path of action between VSG and SG is analysed. To reduce the negative damping torque provided by VSG to SG through this path, a virtual power system stabiliser (VPSS) is proposed and the controller parameters are adjusted according to the phase compensation method. Finally, the effectiveness of VPSS is verified by modal analysis and simulation comparison.

Abstract Self‐assembled monolayers (SAMs) employed in inverted perovskite solar cells (PSCs) have achieved groundbreaking progress in device efficiency and stability for both single‐junction and tandem configurations, owing to their distinctive and versatile ability to manipulate chemical and physical interface properties. In this regard, we present a comprehensive review of recent research advancements concerning SAMs in inverted perovskite single‐junction and tandem solar cells, where the prevailing challenges and future development prospects in the applications of SAMs are emphasized. We thoroughly examine the mechanistic roles of diverse SAMs in energy‐level regulation, interface modification, defect passivation, and charge transportation. This is achieved by understanding how interfacial molecular interactions can be finely tuned to mitigate charge recombination losses in inverted PSCs. Through this comprehensive review, we aim to provide valuable insights and references for further investigation and utilization of SAMs in inverted perovskite single‐junction and tandem solar cells.

With the development of 5G technology and increasing chip integration, traditional active cooling methods struggle to meet the growing thermal demands of chips. Thermoelectric coolers (TECs) have garnered great attention due to their rapid response, significant cooling differentials, strong compatibility, high stability and controllable device dimensions. In this review, starting from the fundamental principles of thermoelectric cooling and device design, high-performance thermoelectric cooling materials are summarized, and the progress of advanced on-chip TECs is comprehensively reviewed. Finally, the paper outlines the challenges and opportunities in TEC design, performance and applications, laying great emphasis on the critical role of thermoelectric cooling in addressing the evolving thermal management requirements in the era of emerging chip technologies.

Abstract Lead‐free polycrystalline manganese telluride holds great potential in the development of waste heat recovery due to its fascinating physical properties. However, the poor thermoelectric (TE) performance in the p‐type MnTe alloys always results from their inferior carrier concentration, leading to low power factor and high thermal conductivity which restrict the overall thermoelectric performance. In this work, the problem is solved by decoupling its electrical and thermal transports through the hole donor Ge‐deficiency in MnTe + x mol.% GeTe (0 ≤ x ≤ 4) compounds. Intrinsically, extra GeTe in MnTe + x mol.% GeTe compound offers free charge carriers due to a narrow bandgap comparatively, realizing not only a full assessment of stimulated electrical performance but also an enhanced power factor. Moreover, benefiting from the nano‐precipitates and tweed microstructures, the lattice thermal conductivity effectively reduces due to the intensive phonon scattering accordingly. Ultimately, a maximum ZT of ≈1.2 at 873 K in the 3 mol.% GeTe doped MnTe sample is realized.

VSG is a pivotal control strategy for GFM, designed to enhance the damping and inertia of power systems, thereby bolstering their stability. Despite its benefits, VSG can induce a QEO due to dynamic interactions between the GFM and the power system. This QEO, is different from control issues, contrasts with the EO associated with conventional SG, which is rooted in electromechanical factors. To address this phenomenon, the influence mechanism of VSG on QEO of the system is studied. First, a comprehensive control model for the VSG is established and its efficacy in providing damping and inertia is assessed. Subsequently, a small-signal model for the VSG when connected to the grid is developed. Utilizing this model, the system’s characteristic roots and the trajectories of principal state variables, elucidating the impact of VSG control parameters on QEO are analyzed. The validity of the models and the underlying mechanisms are confirmed through extensive simulations. Furthermore, for the key problems that need to be solved urgently in the application of VSG in power system with renewable energy, further thinkings are prospected.

Renewable distributed generations are associated with generation intermittency. Exacerbated by the consumption and demand uncertainty and their resulting mismatch, its energy trading suffers similar uncertainty. The situation is severe in the standalone distributed generations (SDG) for lacking transaction access to the utility grid. This paper proposes the energy transaction time determination and minimization algorithm for consumers in the SDG arena. First, blockchain technology is adopted for transaction enhancement and transaction data acquisition. The acquired blockchain data includes the hourly nodes (number of blockchain members), transaction sizes, and corresponding transaction durations. Next, the blockchain-recorded transaction data are fitted using the linear regression (LR) algorithm to obtain their fitting formula. The fitting formula was subsequently optimized in hourly intervals to obtain the optimal transaction time (energy delivery time) using particle swarm optimization (PSO). Finally, the optimal results are presented in a decision tree (DT) to energy consumers in the blockchain platform. Consequently, their transaction decision-making is guided by the result against the inherent transaction time uncertainty. Consumers can thus correctly adjust their transaction habits to suitably adapt to the transaction duration fluctuations in the energy trading arena. Energy transaction delays and transaction costs are consequently minimized leading to greater penetration of renewables and bridging the generation and consumption gap.

Since the high penetration of distributed energy sources complicates the dynamics of electrical power systems, accurate dynamic models are indispensable for study on the transient behavior of the microgrid (MG). In some practices, the lack of full detailed information results in failure of differential equation based dynamic modeling, which leads to a demand for a black-box MG modeling method. It is a critical challenge to maintain the effectiveness of the black-box model under a wide operating range and various fault conditions. In this paper, inspired by the mathematical equivalence between the recurrent neural network (RNN) and differential-algebraic equations (DAEs), a dynamic equivalent modeling method, using long short-term memory (LSTM), is presented to tackle this challenge. At first, the modeling equivalence and advantages of our basic idea are explained. Then, modeling procedures, including data preparation and design guidelines, are presented. Finally, the proposed method is applied to a multi-microgrid testing system for performance evaluation. The results, under various scenarios, reveal that the proposed modeling method has an adequate capability for representing the dynamic behaviors of a black-box MG under grid fault and operating point changing conditions.

Object-aware systems such as Deep Shape Prior SLAM (DSP-SLAM) provide a feasible technique for creating environment sparse maps while representing scene objects as complete 3D models. Such systems provide a compelling solution for improving the intelligence of care robots and enriching the user experience in augmented reality (AR) applications. However, owing to the abrupt and unpredictable movements exhibited by users during AR engagements and the need for real-time robot responses to changes, it is imperative that sensor data is processed robustly and speedily. DSP-SLAM suffers from a low-performance speed of 10–15 fps, although it is based on ORB-SLAM2 which can run at 30 fps. This is mainly because its instance segmentation approach has an average latency of 53ms(18.86fps). To improve tracking robustness, keyframes must be processed at a fast rate. We use a state-of-the-art one-stage deep learning detector, which significantly reduces the wait time for detection-based data association during keyframe creation, and finally present Robust Deep Shape Prior SLAM (RDSP-SLAM). The results show that segmentation was performed at 20ms (50fps), while the object 3D reconstruction quality was the same as that of DSP-SLAM. RDSP-SLAM accepts RGB sequential images at 30fps and tracks them at a mean latency of 38fps.

Standalone microgrids (MGs) and distributed generations (DGs) suffer power generation intermittencies. This leads to a frequent deficit in the supply to consumers. This is worsened by the further delay experienced in the transaction process of the generated energy. Critical loads face the greatest risk of the resulting power shortage. Using the blockchain-integrated data envelopment analysis (DEA), this paper proposes a customer-centered approach to minimizing these inherent issues. Consequently, generation capacity reliability is maximized while the energy supply latency is minimized. First, considering the varying generation reliabilities of standalone MGs, a simulation method is presented using the DEA algorithm for identifying and selecting the most reliable MG in real time for supply. That is, the microgrid with the maximum generation capacity. Next, to address MG’s energy supply latency, a blockchain-integrated energy trading platform is proposed (for prosumers’ DGs) to enhance the peer-to-peer electricity transaction experience. Thus, considering the inherent transaction delay in blockchain-integrated energy trading platforms, a DEA algorithm is proposed for determining and adopting the fastest blockchain transaction. Finally, an algorithm is developed to streamline the integration of the two-step enhancements. It was observed that, as the input data (efficiencies of MGs and blockchains) are changing, maximum power delivery and the fastest electricity transaction rate are achieved in the standalone generation arena at minimal cost. This addresses the characteristic supply deficits and delays, thus minimizing the risk of shedding critical loads.

In the electricity spot market (ESM), thermal power plants (TPP) are quoted based on fuel costs and carbon costs, which impose carbon costs from the generation side to the load-side. The cost of load-side participation in the ESM will increase further. In addition, the traditional ESM centralized bidding model is based on the marginal clearing price, without considering the line transmission constraints, which may affect the security of the power system. To solve this challenge, firstly, an ESM centralized bidding clearing model including the generation side and the load-side is established based on the transaction mechanism of centralized bidding in the ESM. In this model, the marginal clearing prices of two possible bidding scenarios in the ESM are considered. Secondly, the marginal clearing price of the ESM is designed based on the security constraint of the power flow, which can avoid power transmission exceeding its capacity. Finally, a carbon emission cost settlement mechanism is proposed based on the carbon emission flow (CEF) method, which can reduce the carbon emission cost paid by the load-side. The effectiveness of the method is verified by simulations. Compared with a traditional load-side settlement model, the settlement cost on the load-side is reduced by 30.45 %.

In the microgrid, the battery energy storage system (BESS) is used in combination with renewable energy to solve the problem of renewable energy accommodation. Meanwhile, power sources with the BESS can participate in valley filling and reduce electric generation costs. However, the BESS cannot contribute sufficient capability to the microgrid and reduce costs due to its limited capabilities. To tackle this challenge, the output power model of the combined cooling, heating and power microgrid (CCHPM) with the BESS is proposed in this paper. The model takes the micro-gas turbine, the photovoltaic unit, the wind turbine, the electric refrigerator and the BESS as a whole CCHPM. Meanwhile, a capacity optimization method of the BESS is developed to select the appropriate capacity and promote renewable energy accommodation. The minimum cost of the objective function is established. The electric power balance of the CCHPM, charge state constraints of the BESS, capacity constraints and ramp-up/down constraints of the units are derived to optimize the output power of units and the capacity of the BESS. The effectiveness of the proposed method is verified by the system simulation of YALMIP and GUROBI. Compared with the traditional method, the total cost of the microgrid is decreased by 1.4%, and the microgrid’s renewable energy accommodation is enhanced by 12.2%. This capacity optimization method can enhance the applicability of BESS in CCHPM.

Since the high penetration of renewable energy complicates the dynamic characteristics of the AC power electronic system (ACPES), it is essential to establish an accurate dynamic model to obtain its dynamic behavior for ensure the safe and stable operation of the system. However, due to the no or limited internal control details, the state-space modeling method cannot be realized. It leads to the ACPES system becoming a black-box dynamic system. The dynamic modeling method based on deep neural network can simulate the dynamic behavior using port data without obtaining internal control details. However, deep neural network modeling methods are rarely systematically evaluated. In practice, the construction of neural network faces the selection of massive data and various network structure parameters. However, different sample distributions make the trained network performance quite different. Different network structure hyperparameters also mean different convergence time. Due to the lack of systematic evaluation and targeted suggestions, neural network modeling with high precision and high training speed cannot be realized quickly and conveniently in practical engineering applications. To fill this gap, this paper systematically evaluates the deep neural network from sample distribution and structural hyperparameter selection. The influence on modeling accuracy is analyzed in detail, then some modeling suggestions are presented. Simulation results under multiple operating points verify the effectiveness of the proposed method.

With the increasing penetration of wind power, large-scale integrated wind turbine brings stability and security risks to the power grid. For the aggregated modeling of large wind farms, it is crucial to consider low voltage ride-through (LVRT) characteristics. However, in aggregation methods, the approximate neglect behavior is essential, which leads to inevitable errors in the aggregation process. Moreover, the lack of parameters in practice brings new challenges to the modeling of a wind farm. To address these issues, a novel cyber-physical modeling method is proposed. This method not only overcomes the aggregation problem under the black-box wind farm but also accurately realizes the aggregation error fitting according to the operation data. The simulation results reveal that the proposed method can accurately simulate the dynamic behaviors of the wind farm in various scenarios, whether in LVRT mode or normal mode.

In recent years, the problematic circumstances of the constant cropping problem in facility crops have become increasingly serious. Compared to chemical disinfection, soil steam disinfestation offers the benefits of environmental protection and being pollution-free, which can effectively reduce the problem of constant cropping in crops. However, during the steam disinfection procedure, a large quantity of liquid water is formed due to the condensation of high-temperature steam, which causes soil pore blockage, seriously affecting the mass and heat transfer efficacy of steam and, thus, affecting the disinfection efficiency. Therefore, to solve this problem, this paper proposes the use of hot air dehumidification to remove excess water from soil pores and achieve the goal of dredging the pores. However, further exploration is needed on how to efficiently remove excess water from different pore structures through hot air applications. Therefore, this paper first used CFD simulation technology to simulate and analyze the hot air flow field, mass, and heat transfer in soil aggregates of different sizes (8 mm). Then, based on the soil hot air heating experimental platform, research was conducted on the mass and heat transfer mechanism of hot air under diverse soil pore conditions. The results show that as the soil particle size increases from 8 mm, the number of soil macropores also increases, which makes the soil prone to the formation of macropore thermal currents, and the efficiency of hot air heating for dehumidification first increases and then decreases. Among them, the 4–6 mm treatment has the best dehumidification effect through hot air heating, with a deep soil temperature of up to 90 °C and a water content reduction of 6%. The 4–6 mm treatment has a high-temperature heating and dehumidification area of 15–20 cm deep. The above results lay the theoretical foundations for the parameters of hot air heating and dehumidification operations, as well as the placement of the hot air pipe. This paper aims to combine hot air dehumidification technology, for the removal of excess water from soil, and dredging soil pores, ultimately achieving the goal of improving soil steam disinfection efficiency.

Faced with the increasingly severe global aging population with fewer children, the research, development, and application of elderly-care robots are expected to provide some technical means to solve the problems of elderly care, disability and semi-disability nursing, and rehabilitation. Elderly-care robots involve biomechanics, computer science, automatic control, ethics, and other fields of knowledge, which is one of the most challenging and most concerned research fields of robotics. Unlike other robots, elderly-care robots work for the frail elderly. There is information exchange and energy exchange between people and robots, and the safe human-robot interaction methods are the research core and key technology. The states of the art of elderly-care robots and their various nursing modes and safe interaction methods are introduced and discussed in this paper. To conclude, considering the disparity between current elderly care robots and their anticipated objectives, we offer a comprehensive overview of the critical technologies and research trends that impact and enhance the feasibility and acceptance of elderly care robots. These areas encompass the collaborative assistance of diverse assistive robots, the establishment of a novel smart home care model for elderly individuals using sensor networks, the optimization of robot design for improved flexibility, and the enhancement of robot acceptability.

Abstract The authors propose a multifunctional intelligent bed (MIB) that integrates multiple modes of interaction to improve the welfare of mobility‐impaired users and reduce the workload of medical personnel. The MIB features independent autonomous omnidirectional movement, position adjustment, multi‐degree‐of‐freedom (DOF) movement regulation and posture memory functions to facilitate comfortable and convenient interaction for mobility‐impaired users. In particular, an integrated “MIB‐state perception‐interaction interfaces” system is established, and a bed fall risk detection algorithm and assisted get‐up‐transfer algorithm is proposed. By recognising and sharing human body state characteristics, nursing collaboration can be achieved with caregivers or other nursing robots. Comprehensive experiments demonstrate that the MIB is a novel MIB that is highly adaptable to the environment, convenient to interact with and safe. By integrating the proposed algorithms, daily safety monitoring, assisted get‐up and defecation tasks can be effectively accomplished. This technology demonstrates excellent applicability and promising prospects for implementation in hospitals, nursing centres and homes catering to elderly and disabled individuals with mobility impairments.

With the development of new energy, the ancillary service demand of power grid is increasing. The published literature focuses on building a blockchain ancillary service market to ensure the safety and efficiency of ancillary service transactions. However, due to the high threshold and poor scalability, the existing blockchain market is hard to meet the demand of the response rate of ancillary services. Therefore, this paper proposes an energy storage capacity competition-based demand response method in blockchain ancillary service market. Firstly, a new market structure is proposed, which uses the Proof of Capacity (POC) as the consensus mechanism. The structure uses POC to realize the competition among users’ energy storage capacity. Secondly, the demand response process of ancillary service market is designed. During the transaction processes, the probability of users participating in the transaction is changed according to the users’ storage capacity. Dealing with different demands, the market chooses the user with the largest capacity or the combination of users to respond. The simulation shows that, compared with the traditional ancillary service market, the capacity competition-based market demand response rate is increased by 20%, the time to fulling all users’ capacity in the market is reduced by 2 h, and the call rate of user’s capacity is increased by 10%.

The excretion care robot’s (ECR) accurate recognition of transfer-assisted actions is crucial during its usage. However, transfer action recognition is a challenging task, especially since the differentiation of actions seriously affects its recognition speed, robustness, and generalization ability. We propose a novel approach for transfer action recognition assisted by a bidirectional long- and short-term memory (Bi-LSTM) network combined with a multi-head attention mechanism. Firstly, we utilize posture sensors to detect human movements and establish a lightweight three-dimensional (3D) model of the lower limbs. In particular, we adopt a discrete extended Kalman filter (DEKF) to improve the accuracy and foresight of pose solving. Then, we construct an action prediction model that incorporates a fused Bi-LSTM with Multi-head attention (MHA Bi-LSTM). The MHA extracts key information related to differentiated movements from different dimensions and assigns varying weights. Utilizing the Bi-LSTM network effectively combines past and future information to enhance the prediction results of differentiated actions. Finally, comparisons were made by three subjects in the proposed method and with two other time series based neural network models. The reliability of the MHA Bi-LSTM method was verified. These experimental results show that the introduced MHA Bi-LSTM model has a higher accuracy in predicting posture sensor-based excretory care actions. Our method provides a promising approach for handling transfer-assisted action individual differentiation in excretion care tasks.

In the task of system analysis for VSG cluster, aggregation modeling method is widely used for simplification. However, there are inevitable errors occur from the process of cluster aggregation. To improve the accuracy of VSG cluster modeling, a data-physical driven modeling method is presented. At first, the equivalence between aggregation error and black box modeling issue is analyzed. Secondly, a hybrid model structure is proposed, which consists of single machine aggregation model and deep neural network based aggregated-error model. Then, to illustrate the modeling procedure, test cases are studied under large disturbance and multi-operating points conditions. The simulation results confirm that the proposed method can provide satisfactory modeling accuracy.

With the increasing penetration of electric vehicles (EVs), the orderly charging–discharging (C–D) strategy of EVs can effectively alleviate the impact of large-scale grid-connected EVs. However, the schedule of the C–D system of EV clusters encounters the curse of dimensionality problem. In addition, the performance of the C–D control strategy faces great challenges of environmental uncertainty of user demand and electricity price. This paper proposes an EV cluster scheduling strategy considering real-time electricity prices based on deep reinforcement learning. Firstly, we establish a distributed real-time optimal scheduling structure according to the real-time price signals of distribution system operators (DSO). Furthermore, to alleviate the curse of dimensionality, we propose a C–D model of a single EV according to the C–D characteristics of EV, and we establish the C–D control model of EVs as a Markov decision process (MDP). Finally, to adapt to the uncertainty of the learning environment, we propose a model-based deep reinforcement learning to optimize the C–D behavior of EVs. After day-ahead training and parameter saving of the proposed model, the C–D scheduling strategy is generated for the real-time system state at each moment of the day. The simulation results of the C–D scheduling strategy for cost-oriented EV charging show that the proposed scheduling strategy effectively reduces the user charging cost by 133.7 dollars and the load peak–valley difference, stabilizes the load fluctuation, and achieves the win–win situation between the power grid and EV users.

The participation of incentive demand response (IDR) can improve power grid flexibility, and reduce peak shaving pressure. However, the further development of incentive demand response services will be limited by the uncertainty of user response behavior. To tackle this problem, this paper proposes a data-driven load forecasting method for IDR, which considers consumer behavior. Firstly, we describe the power market auxiliary service operation mechanism for load aggregators, through which the ability of demand-side resources to respond to auxiliary services is improved. Then, add an attention layer to improve the traditional long and short memory (LSTM) model, and propose an IDR load forecasting method based on this model, which considers the model’s learning of historical data of user behavior. Finally, establish a simulation model to verify the effectiveness of the improved LSTM model forecasting. The proposed method is simulated and compared with the traditional LSTM method and k-nearest neighbor prediction method. The results show that compared with the other two methods, the mean error, root means square error, and mean absolute percentage error in the improved LSTM method are reduced by 7.31 MW, 8.32 MW, 2.06% and 10.62 MW, 13.46 MW and 3.09%, respectively, which effectively improves the accuracy of forecasting and increase the participation of demand-side resources in the power auxiliary service market.

To handle the multiple-objective problem including carbon emission reduction and economic operation of integrated energy system (IES), this paper proposes an optimal scheduling model for IES considering combined weights for low-carbon and economic operation. In the dispatch model of IES, we mainly use the combined cold heat and power (CCHP) to optimize the energy distribution, and use the charge and discharge power of electric vehicles (EVs) to enhance the regulation capability of CCHP. Furthermore, aiming at the lowest carbon dioxide emission and economic cost, and the highest renewable energy accommodation, a multi-objective function considering combined weights is proposed, which combines subjective weights with objective weights. The balance constraints of electricity, heat and cold for the IES are established. Finally, the particle swarm optimization algorithm is used to optimize the multi-objective model based on the energy balance and unit constraints of IES. The proposed method is verified by the case studies. The results show that the proposed model can improve the accommodation of renewable energy and reduce carbon dioxide emissions.

In practical applications of distributed generation, accurate phase and frequency detection of grid voltages is the critical aspect for the control of power converter. Thanks for its similarity and robustness, phase-locked loop (PLL) has been widespread. However, under distorted grid conditions with polluted harmonics, the accuracy of phase detection may be deteriorated. It is still a challenge to maintain the accuracy of phase detection and dynamic performance at same time. In this paper, to tackle this issue, a novel quasi-type-1 (QT1) structure based PLL is proposed. QT1-PLL incorporated with complex filter can meet the requirement of filtering capability, dynamic response and robustness. The utilization of complex filter based on frequency-fixed structure and feed-forward phase compensation path can eliminate the phase error oriented from frequency deviation. The results of comparative experiments confirm the effectiveness of the proposed PLL.

In the collection and transmission of power big data, the problem of data missing exists. In response to this problem, this paper proposes a power data detection and repair method based on SOM-LSTM. Firstly, a large amount of collected power data is analyzed and the type of missing data is determined. Then, the SOM is used to classify the power data. The LSTM is trained according to the characteristic values of different users to complete the detection and repair of different types of missing power data. Finally, the analysis is based on actual data in some regional loads of China. Experimental results show that, compared with the extreme learning machine (ELM) and LSTM, the proposed SOM-LSTM model reduces the mean absolute error (MAE) by 0.2498 and 0.3425, and the root mean square error (RMSE) by 0.1048 and 0.1469, respectively.

The increase in the penetration rate of renewable energy (RE) and the number of electric vehicles (EVs) has created problems for the balance of energy supply and demand of the power system. To solve this problem, it is necessary to evaluate the flexibility of the power system and optimize scheduling flexible resources to enhance the flexibility of the system. This paper proposes a power system flexibility optimization scheduling strategy considering RE and EVs. First, the models of RE and EVs charging load are established. Second, it analyzes the flexibility of the power system and proposes system flexibility evaluation indexes taking into account RE and EVs charging load. Finally, an optimal scheduling model of the power system is established. The improved particle swarm optimization (PSO) algorithm is used to solve it. The optimization results show that the proposed strategy can effectively improve the flexibility of the power system.

The rapid development of electric vehicles (EVs) makes the load of electric vehicle charging stations (EVCSs) affect the power grid. Aiming at the low accuracy of charging station load forecasting caused by the number of EVs, temperature and electricity price, and other factors, this paper proposes a load forecasting method of EVCSs based on a combination of multivariable residual correction grey model (EMGM) and long short-term memory (LSTM) network. Firstly, load influencing factors are analysed, and the grey theory is introduced into the load forecast of EVCSs. The role of EMGM in taking into account the effects of multiple factors and eliminating cumulative errors is analysed. Then, the EMGM and LSTM networks are combined to establish a mapping from the influencing factor data to the forecast, reducing the load forecast error of EVCs. Simulation and experimental results show that the accuracy of EVCSs’ load forecasting can be improved by this method.

With the large-scale deployment of smart meters and wide application of various machine learning algorithms, non-intrusive load monitoring (NILM) has attracted the attention of academia and industry. However, machine learning algorithms often suffer from high variability in load identification performance due to different features. In view of the poor generalization ability and low accuracy of load identification using an individual feature or classifier model, this paper proposes a novel NILM method that accomplishes deep feature fusion and classifier model fusion by an improved Dempster–Shafer (D–S) evidence theory. Firstly, we extract the power features, current harmonic features, and voltage–current (V–I) trajectory features from the input signals. Then, K-nearest neighbor (KNN), random forest (RF), and convolutional neural networks (CNN) are employed to identify load appliances using three individual features. Finally, the probability estimates of each classifier are transmitted to the aggregator for aggregation to obtain the final identification results by the improved D–S evidence theory. The experimental results on the PLAID dataset show that the proposed method can significantly improve identification performances.

At present, autonomous mobile robots are widely used in industrial and commercial fields. However, although the global path searched by existing mobile robot path planning methods has no collision with obstacles, there is a problem in that the path is close to obstacles and is not smooth, and there is a collision safety risk when the robot is actually moving. To solve the above problems, this paper proposes a global path planning method based on minimizing the risk index. Firstly, the distance calculation method of the heuristic function of the traditional graph search algorithm is improved to reduce the number of nodes in the search space. Additionally, by selecting the appropriate search neighborhood, the search efficiency and path smoothness of the algorithm are improved. Thirdly, to increase the distance between the original search path and obstacles, the risk index path search strategy is proposed. Finally, the minimized snap trajectory smoothing method with a safe corridor is used to smooth the original waypoint. Both simulation and real robot experimental results show that the minimum distance between waypoints and obstacles is increased by 43.72% on average, and the number of trajectory inflection points are reduced by 75.12% on average after optimization. As such, the proposed method can fully guarantee safety and generate smooth mobile robot paths in global trajectory planning tasks.

The internal parameters and topology of the power converter are unknown in some practical cases. Existing modeling methods based on impedance frequency scanning method can only guarantee that the dynamic modeling is effective at a single working point. To make the established dynamic model effective in a wide range, an equivalent modeling method for power converter based LSTM Neural Network is presented. At first, the equivalence of black-box modeling problem and deep loop neural network is studied. Then, dynamic modeling method black-box power converter on wide operating range by using LSTM neural network is proposed. Finally, the simulation results under large disturbance and multi-operating points show that the proposed method is effective under wide operation range.

One of the major concerns of the utility companies is to ensure that the generation capacity (GC) is maintained above the load growth. The demand for assessment of the GC at non-distant time intervals is thus crucial. In general, the rising load is linearly proportional to the generation deficiency. A linear regression approach had been successfully developed to predictively accommodate this demand to avert the issue of deficit supply arising from an unforeseen delay. The Monte Carlo (MC) technique was used in the generation system (GS) modeling. However, due to the inherent stochasticity associated with the MC algorithm, the emerging graphical relationship between the load and the generation deficiency is generally linear but always maintains nonlinearity at various intervals along the gradient curve. This paper integrates the artificial neural network (ANN) nonlinear feature using the Levenberg–Marquardt training algorithm with the MC simulation to accommodate the MC-associated nonlinearities to improve the generation system reliability (GSR) prediction. The generalization performance of the prediction obtained on the test data was found to have been greatly improved.

In this paper, the model of active curtailed wind accommodation based on heat storage electric boiler heating was studied and the curtailed wind accommodation strategy of combined heat storage electric boilers and coal-fired boilers heating was proposed. This is to maximize the curtailed wind accommodation for the ever-increasing curtailed wind problem and environmental problem caused by traditional energy heating in northwest China, northcentral China, and northeast China. Different from the heat storage electric boilers with constant power operation in the past, the multi-level input/exit operation mode of the heat storage electric boilers proposed could accommodate more wind power. The simulation analysis result of a wind power heating project in northeastern China shows that the strategy proposed for active curtailed wind accommodation of heat storage electric boiler heating can effectively improve the wind power accommodation capacity.

There are numerous prospects for automated unmanned forklifts in the fields of intelligent logistics and intelligent factories. However, existing unmanned forklifts often operate according to offline path planning first followed by path tracking to move materials. This process does not meet the needs of flexible production in intelligent logistics. To solve this problem, we proposed an optimized online motion planner based on the output of the state grid as the original path. Constraints such as vehicle kinematics; dynamics; turning restriction at the end of the path; spatial safety envelope; and the position and orientation at the starting point and the ending point were considered during path optimization, generating a precise and smooth trajectory for industrial forklifts that satisfied non-holonomic vehicle constraints. In addition, a new rapid algorithm for calculating the spatial safety envelope was proposed in this article, which can be used for collision avoidance and as a turning-angle constraint term for path smoothing. Finally, a simulation experiment and real-world tray-insertion task experiment were carried out. The experiments showed that the proposal was effective and accurate via online motion planning and the tracking of automated unmanned forklifts in a complicated environment and that the proposal fully satisfied the needs of industrial navigation accuracy.

The optimal scheduling of integrated energy system (IES) can improve energy efficiency and economic operation. However, the existing scheduling methods cannot accurately handle the dynamic changes of supply and demand sides in the electricity-gas-heat IES thanks to their power uncertainties. To tackle this problem, an optimal dispatch framework based on the asynchronous advantage actor-critic (A3C) method of IES is proposed. Firstly, we describe the dispatch problem of IES with multiple uncertainties as Markov decision process (MDP) according to the corresponding mathematical models and constraints. Then, the dispatch framework based on A3C is developed to optimize the control decision for supply and demand sides by asynchronous learning of agents and to reduce the relevance of parameters update of neural networks by the multi-agent utilization of the central processing unit (CPU) multi-threading function. Finally, our proposed methods are verified by the simulations. Compared with the previous optimization algorithms, the training time of the proposed methods is shortened by 37% and 30%, and the daily average operating cost is reduced by 3%, 5.2% and 8.7%.

The dynamic behaviors of microgrid become more complicated due to the increasing implementation of distributed energy generation, energy storage. It is significant to study the dynamic response for power planning, analysis, and control of microgrid. Hence, an accurate dynamic equivalent model is essential, since it can help to evaluate the performance by simulation to avoid the loss and danger in practical test. However, the dynamic model based on differential equation cannot be established because of the lack of information in most of time. To build dynamic model when microgrid is a black-box system, a gated recurrent unit based neural network is proposed in this paper. The proposed neural network can be treated as a black-box differential–algebraic equations. The structure design and model training procedure are presented in detail. Study cases are implemented to evaluate the performance of modeling method. The comparison results show that the proposed dynamic modeling method can precisely estimate the dynamic response of microgrid.

In the power sector, microgrids play a supportive role in bridging the adequacy gap in the conventional electricity supply. Trading of the generated energy has recently been improved by blockchain technology which offers a new cheap, secure, and decentralized transaction approach. Its operation is however associated with an undesired inherent delay during energy transactions initiated by the prosumers, thus, failure to timely attend to incidences of urgent demand could end up in catastrophe at the consumer's side. This article thus proposes a cyber-enhanced transactive microgrid model using blockchain technology with optimized participants' permission protocol to ameliorate this challenge. It is demonstrated that the optimized blockchain participants' permission model leads to improved transaction speed and greater convenience. The transaction speed simulation is thereafter performed and it was also demonstrated that the node population has a greater effect than the transaction block size on the transaction speed improvement.

Renewable energy is usually connected to the DC micro-grid by a large number of power electronic devices, which have the advantages of a fast system response, but the disadvantage to reduce the inertia of the system, which makes the stability of the system worse. It is necessary to increase the inertia of DC micro-grid so that it can recover and stabilize well when it receives a disturbance. In this paper, a small-signal model of DC micro-grid with constant power load (CPL) is established, and a flexible virtual inertial (FVI) control method based on DC bus voltage real-time variation is proposed, by controlling the DC/DC converter of the energy storage system, the problem of system oscillation caused by introducing voltage differential link to the system is solved. Compared with the droop control method, the FVI control method can increase the inertia of DC micro-grid system, reduce the influence of small disturbances, and improve the stability of the system. Finally, the validity of the FVI control method based on small signal model is verified in dSPACE.

The virtual synchronous generator (VSG) technique is used to simulate the external characteristics of a synchronous generator (SG) to provide certain damping and inertia to power systems. However, it may easily cause low-frequency oscillation of the power system. We studied the small-signal stability of a grid-connected virtual synchronous generator. Firstly, the small-signal models of single-VSG and multi-VSG grid-connected systems were established. Subsequently, the system eigenvalues were obtained by solving the state matrix, and the system oscillation modes were analyzed. The eigenvalue analysis method was used to analyze the impacts of parameter changes, such as virtual moment of inertia, virtual damping coefficient, line resistance, and line inductance, on system stability. Finally, our conclusions were verified by numerous simulation models.

This work develops intelligent walking assist robot and gait rehabilitation robot for people with walking disabilities, where distinguishing of abnormal gait is crucial during safe human-robot interaction. Although vision sensors and wearable inertial devices are widely used in fall detection system, shielding of robot mechanical structure brings instability for visual detection method, and utilizing a large number of wearable sensors brings discomfort and complexity for users. Meanwhile, the frequent drag-to-drop gait has never been considered, which is a kind of intermittent state between normal state and fall. It's a major challenge to achieve an abnormal gait recognition system with high stability and good comfort. In this paper, a novel non-contact gait recognition method is proposed, which can identify fall and drag-to-drop gait in a stable and comfortable way. Leg movement data is obtained by a built-in-robot camera, which accurately describes gait law through contrast test and error analysis. New-type ESMF algorithm is proposed to evaluate knee angle precisely and resolve the issue of marker point loss effectively. The comprehensive experiments have shown that this innovative approach calculates accurately knee angle in the walking process so as to recognize the phenomenon of drag-to-drop gait and fall. Additionally, this method having a sound theoretical basis and higher stability solves the issue of marker point loss, which could be applied to a walker with a similar structure for elderly and disabled in a comfortable way.

In this paper, in order to reduce the burden of caring for individuals who are bedridden, we present a method of reasoning about the objects and actions that could satisfy their physiological needs. First, we introduce a method of representing knowledge about everyday objects in terms of their properties and functions. Based on this representation, we then propose a desire-driven reasoning approach that bridges the gap between physiological desires and robot actions. This can also deal with issues caused by uncontrolled domains, including incomplete knowledge and dynamic environments. Finally, we evaluate the proposed method by applying our newly developed KUT-PCR personal care robot to real household scenarios.

In the process of designing electromagnetic equipment how to obtain the satisfied distribution of magnetically dense precisely though the variation of permanent magnet structure is a big challenge. The existing methods only do optimization for parameters under a fixed or several fixed structures. However, no optimization is done concern both structure and parameters simultaneously. To address this, a structural expression code (SEC) method is proposed. It can encode the structure and parameters simultaneously, realizing the optimization of the real structural scheme and parameters at the same time. The SEC method can solve a large proportion of the optimal design of electromagnetic equipment with flexible structure. Of course, it can also be used in other mechanical structural optimal design. The SEC method is proved by optimizing the structure and parameters of a permanent magnet for permanent magnet synchronous motor (PMSM) combined with genetic operator and evolutionary framework. The simulation result obtained by finite element method shows the optimal designed permanent magnet shape produces an air gap flux density waveform whose sinusoidal distribution is 99.78%.

In a wind-diesel system with weak network architecture, many techniques are utilized to enhance its stability. However, the frequency oscillation between the grid-side converter and the diesel generator may be caused by their interactions, and there are only a few studies considering this issue, which is a major challenge for the stability. To tackle this challenge, the cause of frequency oscillation is studied, and an adaptive fuzzy PI controller based on the variable universe is proposed to suppress oscillation in this paper. The control parameter of the wind turbine generator (WTG) converter’s dc-link voltage loop and wind speed has an important influence on the oscillation. Our main idea is to suppress frequency oscillation by the optimizing control parameter. In order to analyze the interaction between WTG and diesel generator, we develop the small-signal models of the wind-diesel generator. Subsequently, the adaptive fuzzy PI controller based on the variable universe is proposed in the dc-link voltage loop to optimize control parameter instead of gain-scheduled PI controller, and its key feature is the innovative use of an extension factor with d-axis current in the output universe. The function of the extension factor is to adapt WTG to different wind speeds. The experimental results demonstrate that the proposed controller can significantly suppress frequency oscillation. The oscillation amplitude of isolated grid frequency is also significantly reduced by up to 0.175 Hz.

The excretion behavior in daily life for the elderly and the disabled is a high frequency, high physical load, and risky behavior. Therefore, we proposed an auxiliary lavatory robot (ALR) with autonomous movement and self-cleaning capability. When the nursing staff assists a user in transferring from a standing or lying state to sitting on the ALR, the ALR can follow the user according to their position and posture. Over the whole transfer process, the ALR always provides the user with the best transfer position and posture, which is an effective approach to reduce workload and physical load. However, confusion and occlusion of the lower limbs between the nursing staff and the user would affect the user’s posture recognition. First, in this paper, a method combined with object segmentation and shape constraint was proposed to extract the contour of the lower limbs of the user and the nursing staff. Then, depending on the position constraint and dynamic characteristics of the legs contour and back contour of the user, a dynamic posture recognition approach based on a two-level joint probabilistic data association algorithm (JPDA) was proposed. Finally, the leg target recognition experiment, path-tracking experiment, and auxiliary excretion transfer experiment were implemented to verify the effectiveness and robustness of our proposed algorithm. The experimental results showed that our proposed method improved the safety and convenience of the user, and it also reduced the workload and physical load of the nursing staff. The ALR, integrated with the proposed method, has a good universal property for the elderly and disabled with weak motion capability in hospitals, pension centers, and families.

Roboticized nursing technology is a significant means to implement efficient elderly care and improve their welfare. Introducing multi-heterogeneous robot systems (MHRS) and sensor networks into a smart home is a promising approach to improve the safety and acceptability of elderly care services in daily life. Among them, the energy consumption and task planning of MHRS determine nursing safety, which is particularly important in the real nursing process. Therefore, we established a novel smart home for elderly care based on seven heterogeneous nursing robots, and proposed a multi-robot task allocation (MRTA) algorithm, considering execution time and energy consumption. The whole system efficiency makes up for the functional limitations and service continuity of traditional MHRS. To realize efficiently conducted multitasks, we established an architecture with centralized task allocation center, robot alliance layer and distributed execution layer for the MHRS. The self-organizing architecture contributes to overall task allocation, communication and adaptive cooperative control between different robots. Then, to clearly describe the continuous nursing process with multiple simultaneous demands and emergency tasks, we modeled the whole nursing process with continuity, multi-priority, and interpretability. A novel MRTA algorithm with a dynamic bidding mechanism was proposed. Comprehensive experiments showed that the proposed algorithm could effectively solve the three key problems of multi-priority tasks, multi-robot safe and adaptive cooperation, and emergency task call in the scene of elderly care. The proposed architecture regarding the smart home could be applied in nursing centers, hospitals, and other places for elderly care.

It is a considerable challenge to realize the accurate, continuous detection of handgrip strength due to its complexity and uncertainty. To address this issue, a novel grip strength estimation method oriented toward the multi-wrist angle based on the development of a flexible deformation sensor is proposed. The flexible deformation sensor consists of a foaming sponge, a Hall sensor, an LED, and photoresistors (PRs), which can measure the deformation of muscles with grip strength. When the external deformation squeezes the foaming sponge, its density and light intensity change, which is detected by a light-sensitive resistor. The light-sensitive resistor extended to the internal foaming sponge with illuminance complies with the extrusion of muscle deformation to enable relative muscle deformation measurement. Furthermore, to achieve the speed, accuracy, and continuous detection of grip strength with different wrist angles, a new grip strength-arm muscle model is adopted and a one-dimensional convolutional neural network based on the dynamic window is proposed to recognize wrist joints. Finally, all the experimental results demonstrate that our proposed flexible deformation sensor can accurately detect the muscle deformation of the arm, and the designed muscle model and convolutional neural network can continuously predict hand grip at different wrist angles in real-time.

Abstract Lead‐free chalcogenide SnTe has been demonstrated to be an efficient medium temperature thermoelectric (TE) material. However, high intrinsic Sn vacancies as well as high thermal conductivity devalue its performance. Here, β‐Zn4Sb3 is incorporated into the SnTe matrix to regulate the thermoelectric performance of SnTe. Sequential in situ reactions take place between the β‐Zn4Sb3 additive and SnTe matrix, and an interesting “core–shell” microstructure (Sb@ZnTe) is obtained; the composition of SnTe matrix is also tuned and thus Sn vacancies are compensated effectively. Benefitting from the synergistic effect of the in situ reactions, an ultralow κlat ≈0.48 W m−1 K−1 at 873 K is obtained and the carrier concentrations and electrical properties are also improved successfully. Finally, a maximum ZT ≈1.32, which increases by ≈220% over the pristine SnTe, is achieved in the SnTe‐1.5% β‐Zn4Sb3 sample at 873 K. This work provides a new strategy to regulate the TE performance of SnTe and also offers a new insight to other related thermoelectric materials.

This work develops an intelligent walk assist robot for people with walking disabilities, where the precise tracking of the user’s movement is crucial. The time-varying friction and load changes impose significant challenges on the tracking. Although the digital acceleration controller is effective to handle them, it is difficult and time consuming to manually tune the control parameters for optimal performance. This is why the automatic parameter optimization techniques are popular in the literature. Despite this, the existing parameter tuning algorithms suffer from sub-optimal performance and low computational efficiency. In this paper, an improved genetic algorithm (IGA) is proposed, which can quickly identify the optimal parameters. Our algorithm explores a few advanced genetic mechanisms including nonlinear ranking selection, arithmetic crossover operation method with competition and selection mechanisms among several crossover offspring, and adaptive change of mutation scaling. The simulation and experimental results have shown that this novel IGA algorithm can effectively reduce the tracking error from 18mm to 0.08mm and reduces the computational complexity.

Polycrystalline Mn1-xCuxTe (x = 0, 0.025, 0.05, 0.075) thermoelectric materials were prepared by a combined method of melt-quenching and hot press. The effect of Cu doping on the electrical resistivity, band gap, the Seebeck coefficient and thermal conductivity was investigated. The power factors of the Cu-doped samples increase greatly due to the decrease of electrical resistivity and the higher Seebeck coefficient in high temperatures. In addition, the thermal conductivities of the Cu-doped samples also reduce due to the extra phonon scattering from the point defects introduced by Cu doping. As a result, the thermoelectric performance of MnTe is greatly enhanced, and a maximum ZT value of ∼0.55 in the Mn0.925Cu0.075Te sample at 773 K is achieved, which is 35% greater than that of the pristine MnTe sample.

Due to increasing load and characteristic stagnation and fluctuations of existing generation systems capacity, the reliability assessment of generation systems is crucial to system adequacy. Furthermore, a rapid load increase could amount to a consequent sudden deficit in the generation supply before the next scheduled assessment. Hence, a reliability assessment is conducted at regular and close intervals to ensure adequacy. This study simulates and establishes the relationship between the load growth and generation capacity using the generation and load data of the IEEE reliability test system (IEEE RTS ‘96 standard). The generation capacity states and the risk model were obtained using the sequential Monte Carlo simulation (MCS) method. The load was gradually increased stepwise and is simulated against the constant generation capacity. In each case, the reliability index was recorded in terms of loss-of-load evaluation (LOLE). The recorded reliability index was thereafter fitted with the load-growth trend by the linear regression approach. A predictive assessment approach is thereafter proffered through the obtained fitting equation. In addition, a reliability threshold is effectively determined at a yield point for a reliability benchmark.

The volatility and uncertainty of high-penetration renewable energy (RE) challenge the stability of the power system. To tackle this challenge, an optimal dispatch of high-penetration RE based on flexible resources (FRs) is proposed to enhance the ability of the power system to cope with uncertain disturbances. Firstly, the flexibility of a high-penetration RE integrated power system is analyzed. The flexibility margin of power supply and flexible adaptability of RE are then introduced as the evaluation indices for optimal operation. Finally, a multi-objective optimal dispatch model for power system flexibility enhancement based on FRs under the constraint of flexibility indices is proposed. The simulation results show that the proposed optimal dispatch can effectively enhance the flexibility of the power system and the penetration of RE and reduce pollutant emissions. Compared with the conventional method, the daily average emissions of CO2, SO2, and NOx with the proposed method are reduced by about 83,600 kg, 870 kg, and 370 kg, respectively, the maximum allowable volatility of net load is increased by 7.63%, and the average volatility of net load is reduced by 2.67%.

With the increasing penetration of renewable energy, a weak grid with declining inertia and distorted voltage conditions becomes a significant problem for wind and solar energy integration. Grid frequency is prone to deviate from its nominal value. Grid voltages become more easily polluted by unbalanced and harmonic components. Grid synchronization technique, as a significant method used in wind and solar energy grid-connected converters, can easily become ineffective. As probably the most widespread grid synchronization technique, phase-locked loop (PLL) is required to detect the grid frequency and phase rapidly and precisely even under such undesired conditions. While the amount of filtering techniques can remove disturbances, they also deteriorate the dynamic performance of PLL, which may not meet the standard requirements of grid codes. The objective of this paper is to propose an effective PLL to tackle this challenge. The proposed PLL is based on quasi-type-1 PLL (QT1-PLL), which provides a good filtering capability by using a moving average filter (MAF). To accelerate the transient behavior when disturbance occurs, a modified delay signal cancellation (DSC) operator is proposed and incorporated into the filtering stage of QT1-PLL. By using modified DSCs and MAFs in a cascaded way, the settling time of the proposed method is reduced to around one cycle of grid fundamental frequency without degrading any disturbance rejection capability. To verify the performance, several test cases, which usually happen in high renewable penetrated power systems, are carried out to demonstrate the effectiveness of the proposed PLL.