Search

This paper has proposed a novel bearing fault detection method about adaptive Sparse-spike Deconvolution based on Curvelet Transform (CTSSD), where the novel technique about adaptive Sparse-spike Deconvolution names after ASSD. Its purpose is to recover the pulse sequence from a vibration signal including complex noise, and to evaluate the periodic pulse position and pulse amplitude. Firstly, in order to make the results sparse and improves the stability of the result, the L1 norm regularization method is proposed in this paper, which is used to constrain the signal pulse sequence sparsely. Secondly, considering that regularization parameters are not adaptive, the Quantum behavior Particle Swarm Optimization (QPSO) algorithm is proposed to determine the optimal regularization parameters, adaptively. Finally, considering that the periodic features of ASSD extraction are not continuous, the Curvelet transform is further introduced. The fault signal is transformed into the Curvelet domain, and the Curvelet coefficient is used to characterize the fault signal pulse sequence. This method proposed in this paper is applied to the simulation signal and the vibration signal of rolling bearing fault, and is compared with the ASSD and the minimum entropy deconvolution (MED) to verify the reliability and effectiveness.

Increased attention has been paid to the high-capacity variable-speed pumped storage to stabilize the large-scale renewable energy. Nevertheless, the optimum reference values of variable-speed units (i.e., the optimum speed and optimum gate opening) come from the actual operating data from power plants or the model pump-turbine test from manufacturers. It is generally a commercial secret, and no historical data is available for a new power plant. For general researchers, it is not easy to obtain relevant data. Therefore, this paper proposes an optimization method for the variable-speed unit and establishes a simplified steady-state model that can be used for medium and long-term power system research. First, based on the pump-turbine's performance curves, this paper presents an optimization method for the variable-speed unit in turbine and pump mode without additional tests, aiming to optimize the turbine's efficiency and track the input power of the pump. Second, the concept of “steady operation area in non-optimum condition” is proposed based on the non-linear characteristics of the optimum operating condition curve. Finally, simplified steady-state models of the variable-speed unit are established, and the model validation has been tested in typical scenario analysis. The results show that the variable-speed unit's steady-state operating characteristic can be described by this simplified linear model, based on the power, the head, and the optimum speed (or the gate opening). This simplified model's percentage error is less than 5%, which meets engineering accuracy requirements. The error mainly comes from the linear correlation coefficient of the optimum operating condition curve and the operating head's deviation from the design head. This model is available for power system planning and dispatch.

Semantic segmentation is a task that covers most of the perception needs of intelligent vehicles in an unified way. Recent studies witnessed that attention mechanisms achieve impressive performance in computer vision task. Current attention mechanisms based segmentation methods differ with each other in position and form of the attention mechanism, and perform differently in practice. This paper firstly introduces the effectiveness of multi-scale context features and attention mechanisms in segmentation tasks. We find that multi-scale and channel attention can play a vital role in constructing effective context features. Based on this analysis, this paper proposes an efficient attention pyramid network (EAPNet) for semantic segmentation. Specifically, to efficient handle the problem of segmenting objects at multiple scales, we design efficient channel attention pyramid (ECAP) which employ atrous convolution with channel attention in cascade or in parallel to capture multi-scale context by using multiple atrous rates. Furthermore, we propose a residual attention fusion block (RAFB), whose purpose is to simultaneously focus on meaningful low-level feature maps and spatial location information. At the same time, we will explore different channel attention modules and spatial attention modules, and describe their impact on network performance. We empirically evaluate our EAPNet on two semantic segmentation datasets, including PASCAL VOC 2012 and Cityscapes datasets. Experimental results show that without MS COCO pre-training and any post-processing, EAPNet achieved 81.7% mIoU on the PASCAL VOC 2012 validation set. With deeplabv3+ as the benchmark, EAPNet improve the model performance of more than 1.50% mIoU.

In this paper, we propose a new hardware architecture of a very high-speed finite impulse response (FIR) filter using fine-grained seamless pipelining. The proposed full-parallel pipeline FIR filter can produce an output sample in a few gate delays by placing the pipeline registers not only in between components, but also across the components. A precise critical path analysis at the gate level allows to create an appropriate pipelining strategy depending on the throughput requirement. This paper also presents two alternative architectures, each offering different trade-offs in terms of area and throughput rate. The proposed FIR filters are synthesized to measure the maximum throughput and the balance between complexity and speed. The synthesis results show that the proposed fully pipelined FIR filter supports up to throughput of 1.8 Giga samples per second and offers 73.5% less area-delay product (ADP) than the existing systolic designs. Also, the proposed single multiplier-accumulator (MAC) based FIR filter has 3 times higher throughput and 26.0% less area with 75.8% less ADP compared to the existing design.

Under the background of big data, the use of massive online data to improve the real-time characteristics and reliability of wind power prediction and to reduce the impact of wind farms on the power grid makes the power supply and demand balance important problems to solve. This paper provides a new solution for short-term wind power forecasting to address these problems. In this paper, an improved random forest short-term prediction model based on the hierarchical output power is proposed, and it is used to forecast the power output of a real wind farm located in Northwest China. First, a chi-square test is adopted to discretize the power data to divide the large-scale training data and remove abnormal data. The novelty of this study is the establishment of a classification model with the output wind power as the classification target and the use of Poisson re-sampling to replace the bootstrap method of the random forest, that is, to improve the training speed of the random forest algorithm. The results indicate that the proposed technique can estimate the output wind power with an MSE of 0.0232, and the comparison illustrates the effectiveness and superiority of the proposed method.

In this first of two closely related papers, we set the foundation for a new framework, called Response Surfaces (RSs), to address fundamental problems of analyzing, designing, and visualizing spiking neurons and networks. An RS is a plot of the direct transfer function between input and output firing times of a spiking neuron and shows all the patterns of input spike times that fire a neuron at a given time. Thus, an RS is a graphical tool for visualizing, analyzing, and designing spiking neural networks. In this paper, we develop a linear RS framework based on triangular, post-synaptic-potential waveforms and apply it to the following problems: graphing the transfer function of a linear spiking neuron, designing an efficient spiking-XOR gate, analyzing phase tracking, and calculating spike times of arbitrarily large recurrent spiking networks. As another fundamental result of the linear RS framework, we show that the output firing time of a spiking neuron is equivalent to the center-of-mass of input spike times (acting as positions) and weights (acting as masses) plus a fixed delay. In the second paper, we explore the possibilities of more biologically realistic post-synaptic-potential waveforms and create a nonlinear RS framework as an extension of the linear RS framework presented here. Although our application examples in both papers focus on design and analysis, we touch on why the RS framework is helpful for understanding the effects of weight changes in learning algorithms.

Recognizing the current weather conditions from a single image is of great theoretical significance. It also has potential practical value for daily life and traffic scheduling. To achieve that, typical weather recognition methods focus on learning a general weather description, e.g., sunny, cloudy, foggy, rainy and snowy etc, for the overall weather condition. However, it is far away from being sufficient for many tasks especially traffic management and control. To solve this key problem, this paper proposes a Global-Similarity Local-Salience Network (abbreviated as GSLSNet) for traffic weather recognition. Specifically, a simple but effective Global-Similarity Module (GSM) is proposed to recognize the overall weather condition and a Local-Salience Module (LSM) is presented to restrict the network to focus on road weather details. Besides, this paper also provides a new traffic weather dataset, named TWData, which is the first fine categorized dataset especially for highway weather recognition. Experimental results compared with state-of-the-art methods on both public datasets and TWData demonstrate the superiority of the proposed GSLSNet.

This paper reviews motor techniques for reducing the cost of electric vehicles (EVs) and improving their range. In recent years, environmental issues, energy crises and the greenhouse effect have urged the popularization of clean energy EVs. In order to achieve this goal, it is necessary to overcome technical difficulties in vehicle cost and range. As a key component of an EV, the motor occupies a large proportion of the overall vehicle cost, and its efficiency directly affects the mileage. In this context, this article discusses the merits and challenges of three mainstream EV motors: permanent magnet synchronous motor (PMSM), induction motor (IM), and switched reluctance motor (SRM) in terms of vehicle cost and range. Then this paper compares the advanced techniques of these motors in terms of topology, material applications and control strategies. Finally, the development trends and opportunities of the three motors in EVs are predicted.

Recent studies have made significant development in path planning for ships. However some studies blindly obey rules of COLREGS (International Regulations for Preventing Collisions at Sea) or only adopt turning to starboard ignoring actual practice, like, turning to port and considering deviation from the planned route when taking actions for collision avoidance. In view of the COLREGS and ordinary seamen practice, this paper proposes collision avoidance actions before encounter situation and collision avoidance actions in an encounter situation. Based on the different stages of the encounter situation, it will add more choices for ships when taking action. To specify different stages of encounter situation clearly and take proper collision avoidance actions, this paper makes a quantitative analysis of three primary encounter situations; velocity obstacles (VO) is employed to find allowed velocity space for own ship (OS); by making further analysis of the relationship among distance at the closest point of approach, bow cross, and COLREGS, the method gives a clear direction for OS to search the best velocity in allowed velocity space for three primary encounter situations; VO utility function is applied to search specific value of the best velocity and is useful for different encounter situations. Simulations show that the results are effective and deterministic for collision avoidance. This method not only prevents blindly obeying rules of COLREGS but also promotes reducing deviation from the planned route.

The objective of this paper is to propose bidirectional charging/discharging strategies for three-phase grid-tied converters. The bidirectional power flow control feature of the converter is able to realize both charging and discharging capability. Besides, in order to achieve high charging efficiency as well as extend the life of the battery, five charging strategies are adopted and developed: 1) the constant current (CC) charging, 2) the pulse-ripple-current (PRC) charging, 3) the sinusoidal-ripple-current (SRC) charging, 4) the bidirectional pulse-ripple-current (BPRC) charging and 5) the bidirectional sinusoidal-ripple-current (BSRC) charging. The direct quadrature (d-q) transformation is utilized for the converter to realize different charging methods. These methods can be achieved by the digital signal processor (DSP) without adding extra circuit components. In addition, the charging power differences between each strategy are considered and analyzed in this paper. Finally, both simulation and experimental results obtained from a 5-kW prototype circuit verify the performance and feasibility of the proposed bidirectional charger.

A distributed denial of service (DDoS) attack represents a major threat to service providers. More specifically, a DDoS attack aims to disrupt and deny services to legitimate users by overwhelming the target with a massive number of malicious requests. A cyberattack of this kind is likely to result in tremendous economic losses for businesses and service providers due to increasing both operating and financial costs. In recent years, machine learning (ML) techniques have been widely used to prevent DDoS attacks. Indeed, many defense systems have been transformed into smart and intelligent systems through the use of ML techniques, which allow them to defeat DDoS attacks. This paper analyzes recent studies concerning DDoS detection methods that have adapted single and hybrid ML approaches in modern networking environments. Additionally, the paper discusses different DDoS defense systems based on ML techniques that make use of a virtualized environment, including cloud computing, software-defined network, and network functions virtualization environments. As the development of the Internet of Things (IoT) has been the subject of significant research attention in recent years, the paper also discusses ML approaches as security solutions against DDoS attacks in IoT environments. Furthermore, the paper recommends a number of directions for future research. This paper is intended to assist the research community with the design and development of effective defense systems capable of overcoming different types of DDoS attacks.

This paper provides the details of a study on the effects of electron irradiation on two Low Noise Amplifiers (LNA), the Gallium-Arsenide (GaAs) pseudomorphic high electron mobility transistor (pHEMT) based and the Silicon-Germanium (SiGe) Heterojunction Bipolar Transistor (HBT)-based. Previous studies have shown that the properties of GaAs and SiGe HBT’s are very tolerant of gamma, neutron, and proton irradiation without additional radiation hardening. Nowadays, commercials on the shelves (COTS) LNAs have been used in CubeSat space communication systems which may be connected to other communication networks for the implementation of the space-aerial- terrestrial integrated 5G network (SATIN) systems projects, for satellites, launched into Low and Medium Earth Orbits. Previous studies suggest that the electron radiation in space may degrade the LNAs’ performance and might even lead to its failure. Located at the front end of the communication receiver system, this paper conducted such investigation to evaluate the performance under the radiation of the GaAs and SiGe LNAs considering the physics of the technology of each LNA, respectively. The results indicate that both SiGe and GaAs technologies are affected after electron irradiation. As a result, this degradation of the LNAs’ performance affected the communications system performance of the inter-satellite radio link. After the assessment of the quality performance of the communication link at the system level, it has been found that the inter-satellite space link will be at risk under high space radiation dose and the link BER degrades proportionally to the radiation dose level.

A detection model of Internet of Things encrypted traffic based on edge intelligence is proposed in the paper, which can reduce the communication times of distributed Internet of Things gateways in the process of edge intelligence as well as the encrypted traffic detection model establishment time, in order to solve the problems that it is difficult to carry out efficient classification and accurate identification of the encrypted traffic of Internet of Things. In this paper, four new classification and identification methods for encrypted traffic are put forward, namely time-sequence behavior analysis, dynamic behavior analysis, key behavior analysis and two-round filtering analysis. The experimental results show that when the sample size is 1600, the encrypted traffic detection model establishment time is less than 100 seconds, and the accuracy of all the four new traffic classification methods is more than 92% and the recall rates of them are more than 83%.

In order to improve the economy and safety of multi-microgrids (MMGs) scheduling, this paper proposes a research on MMGs scheduling strategy that takes into account dynamic electricity prices based on the blockchain. We first introduce the principle of blockchain, analyze the security and economy of the combination of blockchain and MMGs scheduling, and design the scheduling architecture and process based on the blockchain platform. Second, we set a dynamic electricity price model according to the total power supply and demand of MMGs, and set a load optimization model. Finally, we take optimal system economy and minimum environmental pollution as the objective function, then use the linear programming method and the improved krill herd algorithm (KHA) with nonlinear changes in weights to solve the problem. The simulation results show that: (1) The dynamic electricity prices can reflect the power supply and demand of microgrids and optimize the load; (2) Comparing the three scheduling schemes, the strategy in this paper can improve the economic and environmental protection of MMGs by 37.33% and 39.34%, while reduce the interactive power between the microgrid and the distribution network by 56.28%, and the curtailment rate by 63.22%; (3) The improved krill herd algorithm has higher convergence speed and convergence accuracy;(4) The blockchain technology can ensure the security of scheduling data.

In most of existing Internet of Things (IoT) applications, data compression, data encryption and error/erasure correction are implemented separately. To achieve reliable communication, in particular, in harsh wireless environment with strong interference, error/erasure correction codes with higher correction capability or Automatic repeat request (ARQ) scheme are desirable but at the cost of increasing complexity and energy consumption. Due to resource-constrained IoT device, it is often challenging to implement all of them. In this paper, we propose a novel lightweight efficient secure error-robust scheme, ENCRUST, which is able to achieve these three functions using simple matrix multiplication. ENCRUST is built on the new theoretical foundation of projection-based encoding presented in this paper, by leveraging the sparsity inherent in the signal. We perform theoretical analysis and experimental study of the proposed scheme in comparison with the conventional schemes. It shows that the proposed scheme can work in low SINR range and the reconstructed signal quality shows graceful degradation. Furthermore, we apply the proposed scheme on real-life electrocardiogram (ECG) dataset and images. The results demonstrate that ENCRUST achieves decent compression, information secrecy as well as strong error recovery in one go.

In order to enhance the tolerance of frequency misalignment and solve the amplitude-bandwidth limitation issue of the traditional linear resonance wireless power transfer (WPT) system, this paper proposes a non-linear resonance WPT (NLR-WPT) system which is mainly composed of high frequency power supply, primary linear resonator, secondary non-linear resonator, and load. Such non-linear resonator can be implemented with passive linear capacitor and magnetic saturation inductor, and it can be described by Duffing equation. It has been shown that the bandwidth of NLR-WPT system in the nonlinear state is 1.7 times larger than that in the linear state. In addition, the output voltage basically stabilizes at the same level, the voltage fluctuation is less than 5% over the power frequency ranging from 25.6 kHz to 31.8 kHz, and high overall efficiency of 86.5% can be achieved under a wide range of frequency (27.5 kHz-30.1 kHz). Furthermore, the range of frequency can be extended significantly over which the output voltage is maintained even while the compensate capacitance, load, or coupling coefficient values vary in a certain range. The experimental results agree well with the theoretical analysis and verify the validity of the NLR-WPT system with non-linear resonator. This paper provides a simple and effective scheme to suppress the fluctuation of output voltage, which is suitable for the applications of constant voltage charging such as moving devices or electric vehicles.

The outboard corner reflectors are widely used as a countermeasure to radar tracking and detection. Therefore, it is essential to determine the composite time-domain scattering characteristics of the corner reflector and ship against the sea background and the jamming effectiveness of the corner reflector array. In this paper, the composite electromagnetic scattering characteristics of ship target, corner reflector, and the sea surface are studied based on the time-domain bouncing ray method (TDSBR). The radar echo and broadband radar cross section (RCS) results for the time domain pulse incident are also obtained and the high-resolution range profile (HRRP) is calculated. We then devise an evaluation system based on centroid jamming intensity and range profile similarity to determine the jamming characteristics of passive jamming. We further propose an array composing of three sizes of corner reflectors. Finally, the centroid jamming effectiveness is analyzed by comparing the RCS of the corner reflector array and the ship, and the deception jamming effectiveness is measured by calculating the range profile similarity based on the cosine similarity method and the range profile peak envelope overlap. Simulation results show that the reflector array can achieve a good performance both for centroid and range profile deception. The simulation results given and the evaluation method proposed in this paper provides significant practical insights into the design and analysis of the passive jamming systems.

The appropriate control and management of reactive power is of great relevance in the electrical reliability, stability, and security of power grids. This issue is considered in order to increase system efficiency and to maintain voltage under the acceptable value range. In this regard, novel technologies as FACTS, renewable energies, among others, are varying conventional grid behavior leading to unexpected limit capacity reaching due to large reactive power flow. Thus, optimal planning of this must be considered. This paper proposes a new application for a simple and easy implementation optimization algorithm, called Rao-3, to solve the constrained non-linear optimal reactive power dispatch problem. Moreover, the integration of solar and wind energy as the most applied technologies in electric power systems are exploited. Due to the continuous variation and the natural intermittence of wind speed and solar irradiance as well as load demand fluctuation, the uncertainties which have a global concern are investigated and considered in this paper. The proposed single-objective and multi-objective deterministic/stochastic optimal reactive power dispatch algorithms are validated using three standard test power systems, namely IEEE 30-bus, IEEE 57-bus, and IEEE 118-bus. The simulation results show that the proposed optimal reactive power dispatch algorithms are superior compared with two recent algorithms (Artificial electric field algorithm (AEFA) and artificial Jellyfish Search (JS) algorithm) and other optimization algorithms used for solving the same problem.

To gather hydrological measurements is a difficult task for Autonomous Surface Vessels. It is necessary for precise navigation considering underwater obstacles, shallow and fast water flows, and also mitigate misreadings due to disturbs caused by their propulsion system. To deal with those problems, this paper presents a new topology of an Autonomous Surface Vessel (ASV) based on a catamaran boat with an aerial propulsion system with azimuth control. This set generates an over-actuated 3 Degree of Freedom (DoF) ASV, highly maneuverable and able of operating over the above-mentioned situations. To deal with the high computational cost of the over-actuated control allocation (CA) problem, this paper also proposes a Fast CA (FCA) approach. The FCA breaks the initial nonlinear system into partially-dependent linear subsystems. This approach generates smaller connected systems with overlapping solution spaces, generating fast and robust convergence, especially attractive for embedded control devices. Both proposals, i.e., ASV and FCA, are assessed through mathematical simulations and real scenarios.

This paper considers the novel problem of deriving a Time Division Multiple Access (TDMA) link schedule for rechargeable wireless sensor networks (rWSNs). Unlike past works, it considers: (i) the energy harvesting time of nodes, (ii) a battery cycle constraint that is used to overcome so called memory effects , and (iii) battery imperfections, i.e., leakage. This paper shows analytically that the battery cycle constraint and leaking batteries lead to unscheduled links. Further, it presents a greedy heuristic that schedules links according to when their corresponding nodes have sufficient energy. Our simulations show that enforcing the battery cycle constraint increases the link schedule by up to 1.71 (0.31) times for nodes equipped with a leaking (leak-free) battery. When nodes have a leaking battery, the derived schedules are on average 1.05 times longer than the case where nodes have a leak-free battery. Finally, the battery cycle constraint reduces the number of charge/discharge cycles by up to 47.41% (45.67)% when nodes have a leak (leak-free) battery. Between leak-free and leak battery scenarios, using the former produces up to 51.46% fewer cycles than the latter.

Network pruning can effectively reduce a model's capacity and computational load, thereby making model deployment in mobile devices less difficult than that without pruning. To improve the pruning rate of the model while maintaining the kernel's feature extraction ability, this paper designs a progressive kernel pruning method for CNN model compression based on the proposed information mapping sparsity index. This method first prunes the kernels in the filter and then prunes the kernels in the convolution layer when the model reaches a certain compression ratio. The whole process is called progressive kernel pruning (PKP). For the kernel pruning process, this paper defines the information mapping sparse index (IMSI), which is used to measure the mapping ability of the convolution kernel related to the amount of information transferred by the convolution operation. When pruning the kernels of filters and layers, according to the IMSI, the kernels with the strongest mapping abilities are retained to transfer as much information as possible with the least number of kernels. Progressive kernel pruning can make use of the characteristic that the model is easy to optimize when kernel pruning in the filter, and it avoids having the model easily fall into local optima when kernel pruning in the layer directly. The experimental results on the CIFAR-10/100 and ImageNet datasets show that compared to the existing CNN model compression methods, the IMSI-based progressive kernel pruning method exhibits better compression performance in processing the model compression tasks that are currently popular. In particular, pruning VGG-16 on CIFAR-10 with our model achieves a compression ratio of 80.8x and an acceleration ratio of 14.8×, which are 5.8× and 4.2× higher than the best results at present, respectively, and the classification accuracy decreases by only 0.59% relative to that of the baseline.

Enhancement of low-light images is a challenging task due to the impact of low brightness, low contrast, and high noise. The inability to collect natural labeled data intensifies this problem further. Many researchers have attempted to solve this problem using learning-based approaches; however, most models ignore the impact of noise in low-lit images. In this paper, an encoder-decoder architecture, made up of separable convolution layers that solve the issues encountered in low-light image enhancement, is proposed. The architecture is trained end-to-end on a custom low-light image dataset (LID), comprising both clean and noisy images. We introduce a unique multi-context feature extraction module (MC-FEM) where the input first passes through a feature pyramid of dilated separable convolutions for hierarchical-context feature extraction followed by separable convolutions for feature compression. The model is optimized using a novel three-part loss function that focuses on high-level contextual features, structural similarity, and patch-wise local information. We conducted several ablation studies to determine the optimal model for low-light image enhancement under noisy and noiseless conditions. We have used performance metrics like peak-signal-to-noise ratio, structural similarity index matrix, visual information fidelity, and average brightness to demonstrate the superiority of the proposed work against the state-of-the-art algorithms. Qualitative results presented in this paper prove the strength and suitability of our model for real-time applications.

There is a wide application prospect of Nearest level pulse width modulation (NL-PWM) in cascaded H-bridge (CHB) converter, however, it will produce narrow- and error-pulse in the process of implementation, which has a negative impact on CHB efficiency, output voltage quality and safe operation of power devices. In this paper, the NL-PWM strategy is improved based on round function to solve the narrow pulse of traditional NL-PWM, and a step wave delay update strategy is proposed to solve the error-pulse of NL-PWM. This paper analyzes the basic principle of improved NL-PWM and its ability to avoid narrow pulse of power device, shows the harmonic characteristics of improved NL-PWM, analyzes the generation mechanism of error pulse, and calculates the error of output voltage caused by the proposed step wave delay update strategy. The correctness of theoretical analysis and the effectiveness of the proposed method are verified by simulation and experiment. The research shows that the improved NL-PWM can effectively prevent narrow pulse while maintaining good harmonic characteristics, and the proposed step wave delay update strategy can avoid error pulse and has little impact on the output of NL-PWM.

Integrating renewable energy sources (RESs) into electrical power systems has gotten highly noticeable among researchers and those interested in electrical energy production due to the increase in energy demands, fossil fuel exhaustion, and ecological effects. PV-based renewable energy generation is one of the essential RESs that has appeared and had played a vital role in electrical power systems recently due to their advantages. In this regard, this paper presents a multi-objective computation problem for optimal siting and the design of grid-tied PV systems to achieve optimum generating reliability, considering some states of different generation probabilities. The proposed paper studies the evaluation of the grid-tied PV systems reliability, the states of generation probabilities, the generation buses availabilities, the capacities of the generation’s system in or out of service for each failure state, and the frequency and mean duration of generation failure states. The presented multi-objective computation problem is optimized using a modified adaptive accelerated particle swarm optimization (MAACPSO) algorithm. The effectiveness of the proposed method is demonstrated through IEEE_EPS_24_bus integrated with PV systems. Results revealed the ability of MAACPSO to solve the multi-objective optimization problem presented, consequently supporting the system reliability.

The investment demand forecast of the power grid is the further deepening of grid supply and demand management. Under the current reform of the power system in China, we are facing new challenges in analyzing and predicting the investment demand of the power grid. Firstly, the influencing factors of power grid investment demand from both internal and external aspects are analyzed in this paper. Then the selection principles of power grid investment demand factors by domestic and foreign scholars are summarized in this paper. Based on the above content, the forecast indicators and selection methods of domestic and foreign scholars on power grid investment demand are summarized in this paper. The common influencing factor mining algorithms are the grey correlation analysis, the core principal component analysis, and their combination methods. Moreover, the research status of domestic and foreign scholars in power grid investment demand, data mining technology, and forecasting technology is investigated. The current situation of power grid investment demand in China and the common forecasting methods are also elaborated. Finally, the key issues that need to be solved in future research on the investment demand of power grid infrastructure are summarized in this paper. This review will help power utilities find problems in power grid infrastructure investment demand and guide the future investment demand calculations of power utilities.

As the manufacturing industry gradually expands and tends to be intelligent, the industrial networks have been greatly applied and are of paramount importance. Furthermore, taking the advantages of wireless networks into account, such as flexible deployment, low cost, and easy maintenance, etc., this paper is dedicated to the research of industrial wireless networks. Facing the problem of limited radio resource, in order to make full use of frequency band, this paper proposes a resource scheduling scheme with puncturing, which guarantees the reliability of subsequent ultra-reliable and low-latency communications (URLLC) services under the conditions of existing enhanced mobile broadband (eMBB) services. By optimizing the bandwidth pre-allocation for eMBB services, puncture weight and transmit power, this scheme can effectively meet the needs of the system. Specifically, the purpose of this paper is to minimize the decoding error rate of the devices carrying URLLC service while ensuring the demand for throughput of eMBB services. The problem proposed is a nonlinear stochastic optimization problem involving complex Q-function. The closed-form expression of the objective function is obtained through linear approximation and probability theory, which is further decoupled into three sub-problems. To this end, a block coordinate descent optimization (BCDO) algorithm is proposed to obtain the optimal bandwidth allocation, puncture weight and transmit power. The simulation results verify the rationality of the theoretical analysis and the effectiveness of the proposed algorithm, and also elaborate the influence of different parameters on the decoding error rate of the URLLC devices.

As the state-of-the-art technology of Bayesian inference, based on low-dimensional principal components analysis (PCA) subspace inference methods can provide approximately accurate predictive distribution and well calibrated uncertainty. However, the main problem of PCA method is that it is a linear subspace feature extractor, and it cannot effectively represent the nonlinearly high-dimensional parameter space of deep neural networks (DNNs). Firstly, in this paper, in order to solve the main problem of the linear characteristics of PCA in high-dimensional space, we apply kernel PCA to extract higher-order statistical information in parameter space of DNNs. Secondly, to improve the efficiency of subsequent computation, we propose a strictly ordered incremental kernel PCA (InKPCA) subspace of parameter space within stochastic gradient descent (SGD) trajectories. In the proposed InKPCA subspace, we employ two approximation inference methods: elliptical slice sampling (ESS) and variational inference (VI). Finally, to further improve the memory efficiency of computing the kernel matrix, we apply Nyström approximation to determine the suitable size of subsets in the original datasets. The novelty of this paper is that it is the first time to apply the proposed InKPCA subspace with Nyström approximation for Bayesian inference in DNNs, and the results show that it can produce more accurate predictions and well-calibrated predictive uncertainty in regression and classification tasks of deep learning.

This paper presents a simple, didactical and compact solution for the components of the magnetic vector potential and magnetic field generated by a circular arc filament carrying a direct current using the well-known incomplete elliptic integrals, their properties and formulas. Using a well-defined method, the expressions are obtained in Cartesian coordinates and then easily transformed to a compact expression in cylindrical coordinates. The paper provides numerical data for different study cases where arbitrary arc filaments are considered. Through Gauss’ magnetic law, special cases, and comparisons with numerical methods, a degree of validation is provided for the derived expressions.These solutions represent a computational advantage for magnetic field calculations for applications like wireless power transfer. Finally, new integral formulas are derived using elliptic integrals.

When taking pictures of electronic screens or objects with high-frequency textures, people often run across colorful rainbow patterns that are known as “moire”, seriously affecting the image quality and subsequent processing. Current methods for removing moire patterns mostly extract multiscale information by downsampling pooling layers, which may inevitably cause information loss. To address this issue, this paper proposes a demoireing method in the wavelet domain. By employing both discrete wavelet transform (DWT) and inverse discrete wavelet transform (IDWT) instead of traditional downsampling and upsampling, this method can effectively increase the network receptive field without information loss. In addition, to further reconstruct more details of moire patterns, this paper proposes an efficient attention fusion module (EAFM). With a combination of efficient channel attention, spatial attention and local residual learning, this module can self-adaptively learn various weights of feature information at different levels and inspire the network to focus more on effective information such as moire details to improve learning and demoireing performance. Extensive experiments based on public datasets have shown that this suggested method can efficiently remove moire patterns and has a good quantitative and qualitative performance.

The low efficiency of the electromagnetic forming is one of the obstacles for the advancement of this technology. Aiming at solving this issue, this paper presents a double tube-task electromagnetic forming technology method by employing a single driving coil to achieve tube compression and expansion at the same time. On the basis of electromagnetic forming principles and the law of energy conversion, this paper presents a new electromagnetic forming model for double tubes. The magnetic flux density distribution, electromagnetic force distribution and forming efficiency of the tube are analyzed and the effect of the equivalent pulse width of the driving coil current on the forming efficiency is investigated. Results show that in order to maintain the same plastic strain energy of the tube, the initial discharge energy required by the proposed double tube model is 25 kJ which is smaller than that required by a traditional single tube model (33.58kJ) and the efficiency can be improved by about 34.3%. The double tube model attains the maximum forming efficiency near the optimal current pulse width and can improve the low efficiency of traditional electromagnetic forming to a certain extent and provide more possibilities for this technological process.

Recent advances in the cyber-physical smart grid (CPSG) have enabled a broad range of new devices based on the information and communication technology (ICT). However, these ICT-enabled devices are susceptible to a growing threat of cyber-physical attacks. This paper performs a thorough review of the state-of-the-art cyber-physical security of the smart grid. By focusing on the physical layer of the CPSG, this paper provides an abstracted and unified state-space model, in which cyber-physical attack and defense models can be effectively generalized. The existing cyber-physical attacks are categorized in terms of their target components. We then discuss several operational and informational defense approaches that present the current state-of-the-art in the field, including moving target defense, watermarking, and data-driven approaches. Finally, we discuss challenges and future opportunities associated with the smart grid cyber-physical security.

This paper proposes a non-contact method for fault location in transmission lines, which is based on magnetic fields produced by current signals measured using magnetoresistive sensors installed only at transmission line terminals, under the phase conductors of the first transmission tower at both terminals or at the substations portals. The proposed method uses the Extended Kalman filter to process these measurements and is based on a travelling wave approach in order to perform the fault localization. This paper also describes the implementation and testing of the method, firstly introducing its overview, followed by an analysis of the magnetic fields produced by the current signals, as well as considerations on their measurement; secondly, detailing the Extended Kalman filter and the travelling wave approach; and, finally, presenting the results of the method with regards to simulations built using EMTP/ATP to evaluate its robustness under different conditions varying the fault resistance, fault inception angle, phases involved and fault location. The results indicate that the proposed method is robust and accurate.

In this paper, an integration approach (IA) and parametric approach (PA) fusion-based pedestrian dead reckoning (PDR) using low-cost inertial sensors is proposed to improve the indoor position estimation. When moving indoors using a smartphone, various poses such as texting and putting a pocket are possible. In the existing PA-based positioning algorithms, it can be used only when the pedestrian’s walking direction and the device’s direction coincide. In order to solve the problem, this paper proposes an algorithm that constructs state variables based on the IA and uses the position vector from the PA as a measurement. If the walking direction and the device heading do not match based on the poses, the position is updated according to the direction calculated using principal component analysis (PCA) and the step length obtained from the PA. Through experiments considering various operating conditions and paths, it shows that the proposed method stably estimates the position and improves performance even in various indoor environments with the accuracy of 2.13m for Xsens MTw 2.49m for Smartphone.

This paper applies the elitist genetic algorithm to the electric vehicle routing problem with time window. In initialization, the paper proposes an improved neighbor routing initialization method for adaptive elitist genetic algorithm. The improved neighbor routing method is used to select the nearest EV customer as the next route to be scheduled and make the route start from the suitable first customer in the initialization of the elitist GA. It makes the scheduled route begins with a neighboring directionality, which can be inherited in selection, crossover, and mutation operations. For effective convergence, new adaptive crossover probability and mutation probability are provided to make the algorithm converge faster. Experimental studies on randomly distributed customers and Solomon benchmark cases show the effective performance of the algorithm. The algorithm is demonstrated in the simulation of a U.S. Postal Service system.

How to extract discriminative features from redundant video information is a key issue for video pedestrian re-identification. Factors such as occlusion, perspective, and posture changes in complex environments pose severe challenges to pedestrian re-identification based on local methods. In this paper, a posture-guided multi-scale structural relationship learning pedestrian re-identification method is proposed. The purpose is to analyze the video sequence of pedestrians based on the reference pose and the pose alignment model, and extract the sample frame with the highest image quality and the most complete spatial information in the reference pose. The method based on posture guidance can more accurately eliminate the interference of background, occlusion and perspective factors. To further explore the potential relationship between local regions, this paper calculates the relationship matrix between the local regions based on the relationship model to further calculate the relationship weight, and the graph convolutional network based on the relationship weight learns the structural relationship feature of multi-scale regions. The input of the graph convolutional network is a local region divided by a multi-scale method, and the output is a pose-guided multi-scale structural relationship feature. The experimental results on three public datasets show that the proposed method performs favorably against state-of-the-art methods.

As for the direct-current (DC) power systems, the DC leakage current sensor is indispensable for positioning insulation faults. And in this paper, a quasi-digital flux gate sensor was designed based on duty ratio model, and the mathematical model between the excitation voltage period and the measured current which exceeded the measurement range was built by analyzing the working principles of the duty-ratio model digital sensor. In this way, the measurement range of the original sensor could be expanded to 1.5 times of its original. Meanwhile, without changing the hardware, the solution provided in this paper also solved the measurement failure problem of part of sensors when the any overcurrent situations happen in the system, such as overload or short-circuit. Besides, the sensor designed in this paper could achieve Level-0.1 measurement accuracy within range, and reach Level-2 over-range measurement accuracy, satisfying the demands of DC power supply system on leakage current measurement.

This paper reports a soft fingertip-mountable tactile actuator based on a Dielectric Elastomer Actuator (DEA), which exhibits high output force over a wide frequency range with a lightweight and soft structure. DEA is a soft actuator characterized by its large area strain, fast response speed, and high specific energy density. The proposed soft tactile actuator is constructed of a multi-layered conical DEA structure. This design has safety benefits because it isolates the high voltage components from the contact point. In this paper, the resonance frequency of the tactile actuator was designed to be at 250 Hz to maximize vibrotactile stimulation. In addition, the geometric design parameters of the soft tactile actuator were optimized by conducting the simulations and the experiments. Based on these efforts, the proposed actuator produces a high output force of 8.48 N at the resonance frequency, with a maximum displacement of 0.46 mm. Our wearable prototype was an entirely soft haptic system, which exhibits high output force, as well as flexibility and conformity with a total weight of 2.6 g.

Wireless energy transmission and harvesting techniques have recently emerged as attractive solutions to realize wireless powered networks. By eliminating fundamental power constraints arising from the use of conventionally battery sources, wireless modes of energy transmission provide viable means to power wireless network devices away from the grid. Metasurfaces have emerged as key enablers for the use of microwave energy as a power source. Their unique abilities to tailor electromagnetic waves have motivated significant research interest into their use for power-focused microwave systems. This article provides an overview of progress in the development of metasurface implementations for microwave energy transmitters and energy harvesters. First, the paper provides a basic introduction to metasurfaces, after which it reviews research progress in metasurfaces for microwave energy transmission and harvesting. Also highlighted are key parameters by which the performance of such metasurface designs are characterized. In addition, an overview of studies on metasurfaces as reconfigurable intelligent surfaces in wireless networks supporting the simultaneous transmission of information and energy is presented. Finally, the paper highlights existing challenges, and explores future directions, including opportunities to control radio environments through ambiently energized reconfigurable intelligent surfaces in next-generation wireless networks.

Compared with AC power system, DC distribution network is a low-damping system without natural zero crossing. The DC fault current will reach its peak value within a few milliseconds, posing a great threat to power electronic devices. Therefore, the rapid and high-precision identification of DC line faults is one of the technical difficulties faced by the VSC based DC distribution network. In this paper, a new fault location method based on traveling wave differential current with Hausdorff distance and cubic spline interpolation is proposed. First, the forward and reverse traveling waves at both ends of the VSC based DC lines are extracted, the Karenbauer transform is used to decouple the positive and negative electrodes to obtain accurate fault information. Then, the internal relationship between the differential current and the fault current is obtained according to the principle of traveling wave transmission invariance, the cubic spline interpolation algorithm is used to solve the traditional sampling frequency limitation, and the positive differential current in the interception time window is smoothly displaced. At last, the Hausdorff distance algorithm is used to analyze the correlation between the translational differential current and the reverse differential current. The fault location corresponds to the highest correlation. The proposed method uses the time-domain information at both ends of the VSC based DC lines, and it is less affected by the transition resistance. Lots of simulation experiments prove that it has a strong anti-noise interference ability and high reliability, and is less affected by the sampling frequency after cubic spline interpolation. Compared with traveling wave algorithm based on Pearson correlation coefficient, the method in this paper shows a significantly shortened error, 27% on average.

In the advanced television systems committee (ATSC) 3.0 system, the concept of flexibility is significant for supporting backward compatibility within the same ATSC 3.0 system. However, since the conventional bootstrap signal detection scheme is difficult to support the flexibility, the conventional bootstrap signal detection scheme should be newly designed according to the change of version. In this paper, a convolution neural network (CNN) model for bootstrap signal detection in ATSC 3.0 is proposed to maintain the flexibility of bootstrap. Additionally, for minimizing the loss of error performance of CNN-based bootstrap detection scheme, this paper proposes two dimensional alternate array to utilize the correlation of the adjacent bootstrap symbol and proposes the offline learning method using the bootstrap signal corrupted by noise to improve the error performance.

Being complex and combinatorial optimization problems, Permutation Flow Shop Scheduling Problems (PFSSP) are difficult to be solved optimally. PFSSP occurs in many manufacturing systems i.e. automobile industry, glass industry, paper industry, appliances industry, and pharmaceutical industry, and the generation of the best schedule is very important for these manufacturing systems. Evolution Strategy (ES) is a subclass of Evolutionary algorithms and in this paper, we propose an Improved Evolution Strategy to reduce the makespan of PFSSP. Two variants of the Improved Evolution Strategy are proposed namely ES5 and ES10. The initial solution is generated using the shortest processing time rule. In ES5, four offsprings are generated from one parent while in ES10, nine offsprings are generated from one parent. The selection pool consists of both the parents and offsprings. Quad swap mutation operator has been proposed to minimize computational time and for the maximum search of solution space. Also, a variable mutation rate is used for the fine-tuning of results, with the increasing number of iterations the mutation rate is reduced. The performances of both ES variants were tested on two test domains. First, it is applied to benchmark the PFSSP of Carlier and Reeves. Computational results are matched with other well-known techniques available in the literature, and the results show the effectiveness and robustness of the proposed techniques. Secondly, ES is applied to the real-life problem for the manufacturing of batteries to demonstrate its effectiveness. Data was taken from Pakistan Accumulator for NS30-40 Plates battery, the company is daily producing 1400 units of NS30-40 Plates battery. ES is applied to different batch sizes i.e. 35, 140, 1120 & 1400. Our results show that a Min %GAP of 1.25 is found using ES10. Hence the company can increase monthly 450 units of NS30 batteries using the ES10 algorithm.

Alternative models of the electrical resistivity of the soil to the well-known constant piecewise multilayer model are presented in this paper. By their special features, the proposed models lead to reasonably simple semi-analytical expressions for the electric potential. The parameters associated with these models are obtained from apparent resistivity measurements using an optimization procedure. The proposed models allow obtaining a continuous piecewise resistivity profile as a function of depth from the soil surface. From the full definition of the models, the grounding resistance of several electrodes is calculated and compared with that obtained from the classic multilayer model. Finally some application examples including data from field measurements are considered in this paper.

Graph embedding models are widely used in knowledge graph completion (KGC) task. However, most models are based on the assumption that knowledge is completely certain, and this is inconsistent with real-world situations. Although there are multiple studies on uncertain knowledge embedding tasks, they often use knowledge confidence to learn embedding and cannot make full use the structural and uncertain information of knowledge. This paper presents a new embedding model named Structural and Uncertain Knowledge Embedding (SUKE), which comprises two components: an evaluator and a confidence generator. For unknown triples, the evaluator learns the structural and uncertain information to evaluate its rationality and obtain a candidate set. The confidence generator then determines the confidence of the candidate set to achieve KGC. To verify the effectiveness of the proposed model, confidence prediction, triple evaluation, and fact classification tasks are performed on three data sets. Experimental results show that SUKE performs better than mainstream embedding methods. The model proposed in this paper can help advance the research on the embedding of uncertain knowledge graphs.

Crude oil is a vital and valuable commodity in the energy industry. In order to maintain continuous, stable, and reasonably priced supplies, oil producers need cheaper exploration and extraction techniques. Permeability is one of the formation parameters that is a key interest to petroleum engineers in determining the economic worth and yield of crude deposits, yet permeability prediction remains a difficult problem. Many approaches have been applied to solving this important issue. Soft computing has been deployed to predict permeability. In this paper, we present an extensive review of the existing research that has been conducted on applications of soft computing for permeability prediction. This paper finds out that traditional approaches for permeability prediction are still relevant in the oil and gas industry. Soft computing methods in particular are worthy of addition to this interesting area. This extensive review is intended to be an entry point for further exploration of other approaches that have received little or no attention from researchers.

Remote laboratories have been in use for 25 years now. Whereas several learning oriented meta-analyses exist, validated and agreed-upon tools for assessing the user experience are not readily available. The present paper fills this gap by designing and evaluating a questionnaire focused on the needs of remote labs developers and educators using them. Building from pre-existing tools, a first version of the User eXperience Questionnaire (UXQ version 20190308) was designed to contain four scales, usability, utility, satisfaction and immersion. A total of 180 completed responses were collected from two different remote labs (VISIR and FPGA), in different campuses and in different courses to evaluate the questionnaire. The questionnaire was analyzed in terms of reliability, using Cronbach’s alpha and McDonalds’ omega, and validity of construct, through factor analysis. The reliability of the questionnaire and its four subscales is acceptable, but its validity should be improved. Accordingly, the questionnaire was redesigned to obtain the User eXperience Questionnaire (UXQ version 20191126), which includes three scales: usability, utility and immersion, and three questions per scale. This questionnaire was assessed using the same data. Reliability coefficients are above 0.7 and construct validity is satisfactory. A new questionnaire to evaluate the user experience in remote laboratories has been designed and validated. The questionnaire, now renamed as UXQ4RL v. 1.0, is presented and made available in this paper.

Repetitive controllers (RCs) are known for their ability of controlling periodic exogenous signals, even if these signals have high harmonic content. Due to the variety of repetitive controllers proposed in the literature, which are significantly different from each other, a comparative evaluation in a collective way is analytically complex. This fact implies a greater difficulty to select the appropriate RC strategy when designing a control system, what makes most control system designers to not use this class of controllers. In order to solve this problem, the present paper develops a unified approach for representation of (real and complex) repetitive controllers, which is based on the use of multiple primitive repetitive cells in parallel. Through this approach the main characteristics of a repetitive controller, such as stability properties, dynamic response, set of harmonic components that are effectively compensated and computational burden, are easily identified. Furthermore, in order to validate the potential of the proposed unified approach, comparative studies of $nk \pm m$ RCs and $nk+m$ RCs are performed. An experimental application based on a three-phase shunt active power filter is implemented to validate the theoretical evaluation presented in this paper.

In this paper, we investigate a multitone simultaneous wireless information and power transfer (SWIPT) based modulation scheme for battery-less Internet of Things (IoT) nodes that works in the ultra high frequency (UHF) region. The conventional SWIPT system is assumed to have power-consuming communication modules. Having such modules on to the IoT nodes whose power is harvested from radio frequency (RF) power sources is too unrealistic. In addition, waveform design from the aspect of power harvesting through SWIPT still has room for consideration. Recent studies have explored multitone based SWIPT to increase the power conversion efficiency (PCE). In these schemes, information is modulated by changing each tone’s nature or varying the number of tones. Among these methods, we focused on modulation schemes known as frequency-shift multitone based SWIPT, which shifts frequencies among the tones for information encoding. Unlike previously proposed methods where demodulation requires some power-consuming fast Fourier transfers (FFTs), especially under small communication bandwidths, we applied a signal detection method by measuring output peak to average power ratios (PAPRs) for frequency-shift multitone based SWIPT to reduce power consumption. Based on our analysis, different tone configurations in the frequency domain would yield varieties of nonlinear outputs during the rectification process. In addition, these specific nonlinear output patterns depend on the tone configurations. With this feature, it is possible to demodulate in the time domain at the receiver side using PAPR based measurements, which could eliminate FFT operation. This paper describes how measuring PAPRs enables the detection of signals in theory and validates this through simulations and experiments. We also estimate the communication rates. Based on our results, we achieved 0.46 bits/s/Hz when the number of tones was 6 and estimated that there were $(N-1)(N-2)/2 + 1$ different PAPRs from a given multitone waveform whose number of tones was $N$ .

Wireless communication systems play a very crucial role in modern society for entertainment, business, commercial, health and safety applications. These systems keep evolving from one generation to next generation and currently we are seeing deployment of fifth generation (5G) wireless systems around the world. Academics and industries are already discussing beyond 5G wireless systems which will be sixth generation (6G) of the evolution. One of the main and key components of 6G systems will be the use of Artificial Intelligence (AI) and Machine Learning (ML) for such wireless networks. Every component and building block of a wireless system that we currently are familiar with from our knowledge of wireless technologies up to 5G, such as physical, network and application layers, will involve one or another AI/ML techniques. This overview paper, presents an up-to-date review of future wireless system concepts such as 6G and role of ML techniques in these future wireless systems. In particular, we present a conceptual model for 6G and show the use and role of ML techniques in each layer of the model. We review some classical and contemporary ML techniques such as supervised and un-supervised learning, Reinforcement Learning (RL), Deep Learning (DL) and Federated Learning (FL) in the context of wireless communication systems. We conclude the paper with some future applications and research challenges in the area of ML and AI for 6G networks.

Sentiment Analysis is an important research direction of natural language processing, and it is widely used in politics, news and other fields. Word embeddings play a significant role in sentiment analysis. The existing sentiment embeddings methods directly embed the sentiment lexicons into traditional word representation. This sentiment representation methods can only differentiate the sentiment information of different words, not the same word in different contexts, so it cannot provide accurate sentiment information for word in different contexts. This paper proposes sentiment concept to solve the problem. First, we found the optimal sentiment concept of words in Microsoft Concept Graph according to the context of words. Then we obtained the sentiment information of words under optimal sentiment concept from the multi-semantics sentiment intensity lexicon which we constructed in this paper to achieve accurate embedding of sentiment information and provide more accurate semantics and sentiment representation for words. Finally, we combined two refined word embeddings methods to achieve a more comprehensive word representation. Compared with traditional and sentiment embeddings methods on six representative datasets, the validity of the word embeddings method based on sentiment concept proposed in this paper is verified.

For the utility to plan the resources accurately and balance the electricity supply and demand, accurate and timely forecasting is required. The proliferation of smart meters in the grids has resulted in an explosion of energy datasets. Processing such data is challenging and usually takes a longer time than the requirement of a short-term load forecast. The paper addresses this concern by utilizing parallel computing capabilities to minimize the execution time while maintaining highly accurate load forecasting models. In this paper, a thousand smart meter energy datasets are analyzed to perform day ahead, hourly short-term load forecast (STLF). The paper utilizes multi-processing to enhance the overall execution time of the forecasting models by submitting simultaneous jobs to all the processors available. The paper demonstrates the efficacy of the proposed approach through the choice of machine learning (ML) models, execution time, and scalability. The proposed approach is validated on real energy consumption data collected at distribution transformers' level in Spanish Electrical Grid. Decision trees have outperformed the other models accomplishing a tradeoff between model accuracy and execution time. The methodology takes only 4 minutes to train 1,000 transformers for an hourly day-ahead forecast of (~24 million records) utilizing 32 processors.