Online Load Research Articles

An online long-term load forecasting method: Hierarchical highway network based on crisscross feature collaboration

In order to ensure the efficient operation of the power system, accurate online load forecasting is necessary. Existing studies face challenges in addressing unconventional load changes in long-term load forecasting (LTLF) and are unsuitable for online forecasting due to their reliance on measured Forecast Day Data (FDD). The development of a model independent of prior knowledge is essential to improve the interpretability and applicability of load forecasting methods. In this study, a hybrid model based on Crisscross Feature Collaboration (CFC) and Hierarchical Highway Network (HHN) is introduced. During the horizontal feature learning, the HHN is employed to learn temporal information from annual load curves across multiple time scales, reducing time complexity in LTLF. Subsequently, the prediction results of various time scales are combined to approximate FDD, addressing data leakage problems. In the vertical feature matching phase, the similar day data are screened by the multi-modal information of FDD and the Toeplitz Inverse Covariance Based Clustering (TICC) method, enabling the model to be independent of prior knowledge. Finally, through the HHN, all similar day data are utilized to reconstruct the forecasting load curve. In LTLF, the Root Mean Square Error (RMSE) of the proposed model reduces by 94.35 %, 85.74 %, 84.02 %, 90.72 %, 85.17 % and 87.76 % respectively in different scenarios compared to the direct forecasting. This signifies the effectiveness of our model in predicting the unconventional load change and achieving accurate online LTLF.

A novel collision detection method based on current residuals for robots without joint torque sensors: A case study on UR10 robot

Existing model-based collision detection methods rely on accurate torque dynamic parameters identified using measured joint torques. However, for robots lacking joint torque sensors, only joint currents can be measured, and joint torques must be estimated through the linear relationship between joint currents and joint torque constants. This way can lead to cumulative identification errors in torque dynamic parameters, thereby diminishing the performance of model-based collision detection algorithms. To tackle this challenge, this article proposes an innovative collision detection method based on current residuals, which represent the disparities between measured joint currents and predicted joint currents computed by current dynamic parameters. Then, a dynamic threshold method for current residuals is designed to mitigate the impact of modeling errors at zero-speed direction changes on collision detection performance. Additionally, a suppression strategy based on online load identification and compensation is introduced to reduce the interference of non-collision load factors on collision detection signals. The proposed method mitigates the accumulation of errors on torque dynamic identification resulting from inaccuracies in joint torque constants, ultimately enhancing collision detection performance for robots without joint torque sensors. Extensive experiment results validate the correctness and effectiveness of this approach.

LTI/LQE: A Physics-Based Approach for Online Estimation of Dynamic Loads in the Rotating Frame

Accurate and online load monitoring of vital components located in the rotor system is important for not only inferring usage and estimating fatigue in those components but also for developing load alleviation/limiting control schemes. In the literature, on-board models for online (i.e., real-time) estimation of rotor component loads using available in-flight aircraft fixed-frame measurements are developed using neural networks, a statistical-based approach, or a combination of both. These models are aircraft-specific, do not capture the higher harmonic dynamics of the rotor system vibratory loads, and their synthesis requires obtaining a lot of data. To remedy these issues, this paper introduces a purely physics-based approach to online estimation of rotor system loads. The proposed approach can be applied to any rotary-wing vehicle, does not rely on a large amount of data, and captures the higher harmonic dynamics of the rotor system vibratory loads. The developed model entitled LTI/LQE is synthesized using a linear time invariant (LTI) model of helicopter-coupled body-rotor-inflow dynamics and a linear quadratic estimator (LQE), that is designed to correct the LTI model state response using fixed system measurements. The estimation fidelity of the LTI/LQE model is evaluated in simulation using a high-fidelity nonlinear model of a generic helicopter for online prediction of rotor blade pitch link loads arising from vehicle maneuvers. Results obtained using the LTI/LQE model revealed an interesting finding. It was found that the N/rev (where N is the number of blades) fixed system load measurements have information that can be leveraged to make an inference about the N/rev dynamic loads in the rotating system.

Identification and Stabilization of Constant Power Loads in AC Microgrids

Microgrids represent an attractive solution for a more flexible way to manage electrical power systems, potentially allowing for increased distributed generation and smarter usage of the available electrical power. However, the presence of power electronics and their control can challenge the system stability, especially power electronic converters with constant power load behavior. This paper considers a novel on-line load identification (OLLI) technique applied to the control of power electronic converters, as well as two control strategies to achieve stabilization of the whole system. The OLLI algorithm and the stabilizing effects of the proposed control strategies are validated through simulation and laboratory experiments, showing the feasibility of the presented work applied in future microgrids.

Minimum variance unbiased Bayesian smoothing for input and state estimation of systems without direct Feedthrough: Mitigating Ill-Posedness of online load identification

In this paper, the problem of moving load identification using pure displacement measurements is addressed. It is known that when no assumptions are made on the statistics of the unknown loads and a minimum variance unbiased (MVU) estimation approach is adopted, the existing methods in the literature suffer from a very elevated load estimation uncertainty. This elevated uncertainty is due to ill-posedness of the problem. In this paper a new method is proposed that addresses this issue via an MVU smoothing approach. To alleviate this problem, a MVU smoothing algorithm is proposed in this study, via modification of a MVU smoothing Bayesian estimator proposed by some of the co-authors of this paper hence referred to as MSBE, which leads to substantial decrease in the moving load estimation uncertainties using pure displacement measurements. The efficacy of the MSBE is studied through a simulated experiment corresponding to a numerical model of an operating steel railway bridge with riveted connections and a multi-axle load. The selection of the hyperparameters of the smoothing and filtering techniques are discussed, and the optimal values are presented. The parametric studies show that the proposed method can yield highly accurate results, and substantially outperforms a celebrated MVU filter proposed by Gillijns and De Moor.

Synchrophasor-Based Online Load Margin Estimation Using Incremental Learning Assisted LightGBM in Smart Grid

Traditional load margin estimation approaches, which require a retraining process under dynamic operation variation with increasing uncertainties and consuming excessive time, are insufficient to guarantee the smart grid operates in a stable region. To address this challenge, this article proposes an incremental learning-enhanced LightGBM (IL-LightGBM) method for online load margin estimation utilizing synchronized measurement data. As the key to achieving accurate load margin estimation using the LightGBM algorithm under operation variations, the weight parameters of the pre-trained model are updated online by IL technique through efficiently digesting synchrophasor measurements. The proposed method makes full use of the capability of LightGBM to handle massive measurements, while effectively improving the adaptability to large-scale operational variability. Case studies to evaluate the proposed approach are presented through the numerical simulations of the IEEE 39-bus test system and a larger IEEE 145-bus power system, demonstrating the effectiveness and robustness of the proposed method.

Online cardinality constrained scheduling

In online load balancing problems, jobs arrive over a list. Upon arrival of a job, the algorithm is required to assign it immediately and irrevocably to a machine. We consider such a makespan minimization problem with an additional cardinality constraint, i.e., at most k jobs may be assigned to each machine, where k is a parameter of the problem. We present both upper and lower bounds on the competitive ratio of online algorithms for this problem with identical machines.

Congestion games with priority-based scheduling

We reconsider atomic and non-atomic affine congestion games under the assumption that players are partitioned into p priority classes and resources schedule their users according to a priority-based policy, breaking ties uniformly at random. We derive tight bounds on both the price of anarchy and the price of stability as a function of p, revealing an interesting separation between the general case of p≥2 and the priority-free scenario of p=1. In fact, while in absence of priorities the worst-case prices of anarchy and stability of non-atomic games are lower than their counterparts in atomic ones, the two classes share the same bounds when p≥2. Moreover, while the worst-case price of stability is lower than the worst-case price of anarchy in atomic games with no priorities, their values become equal when p≥2. Said differently, the presence of priorities simultaneously irons out any combinatorial difference between atomic and non-atomic requests and among different pure Nash equilibria to produce a unique representative worst-case situation. Notably, our results keep holding even under singleton strategies. Besides being of independent interest, priority-based scheduling shares tight connections with online load balancing and finds a natural application within the theory of coordination mechanisms and cost-sharing policies for congestion games. Under this perspective, a number of possible research directions also arise.

Online load balancing on uniform machines with limited migration

Consider that one job after another arrives online and must be placed on machines immediately, without knowing which jobs might arrive later. The goal is to balance the load of the machines – the sizes of placed jobs divided by the machine speed. Each job contributes a migration potential proportional to its size, which can be used to relocate some previously placed jobs. This paper presents results with improved competitiveness compared to the pure online case using only very limited migration.

Model predictive control of power plant cycling using Industry 4.0 infrastructure

This work involves the Industry 4.0 infrastructure developed at West Virginia University (WVU) for process systems applications. This infrastructure emulates an interconnected environment, enabling communication and data sharing among different components for use in academic and industrial settings. The current infrastructure encompasses a power plant model interacting with online load demand, distributed control systems, and data analytics components. The developed model of a sub-critical coal-fired power plant is employed to evaluate classical and advanced control strategies using this infrastructure under different operating conditions. Specifically, the control strategies evaluated include classical proportional–integral–derivative (PID) and advanced model predictive control (MPC) structures, focusing on the dynamic matrix control (DMC) approach with an in-house modified sequential quadratic programming (SQP) solver. The MPC approach is developed and simulated in closed loop to address setpoint tracking and load-following scenarios under power plant cycling conditions. In this infrastructure, the PI System centralizes all the information received from the power plant model and the online power demand and sends the control actions calculated by the MPC back to the power plant model for implementation. Results of the implementation of these control strategies are discussed focusing on power plant operating regions associated with cycling.

Well-behaved online load balancing against strategic jobs

In the online load balancing problem on related machines, we have a set of jobs (with different sizes) arriving online, and the task is to assign each job to a machine immediately upon its arrival, so as to minimize the makespan, i.e., the maximum completion time. In classic mechanism design problems, we assume that the jobs are controlled by selfish agents, with the sizes being their private information. Each job (agent) aims at minimizing its own cost, which is its completion time plus the payment charged by the mechanism. Truthful mechanisms guaranteeing that every job minimizes its cost by reporting its true size have been well studied (Aspnes et al. JACM 44(3):486–504, 1997, Feldman et al. in: EC, 2017). In this paper, we study truthful online load balancing mechanisms that are well-behaved (machine with higher speed has larger workload). Good behavior is important as it guarantees fairness between machines and implies truthfulness in some cases when machines are controlled by selfish agents. Unfortunately, existing truthful online load balancing mechanisms are not well behaved. We first show that to guarantee producing a well-behaved schedule, any online algorithm (even non-truthful) has a competitive ratio at least \(\varOmega (\sqrt{m})\), where m is the number of machines. Then, we propose a mechanism that guarantees truthfulness of the online jobs and produces a schedule that is almost well-behaved. We show that our algorithm has a competitive ratio of \(O(\log m)\). Moreover, for the case when the sizes of online jobs are bounded, the competitive ratio of our algorithm improves to O(1). Interestingly, we show several cases for which our mechanism is actually truthful against selfish machines.

A deep model for short-term load forecasting applying a stacked autoencoder based on LSTM supported by a multi-stage attention mechanism

This paper presents an innovative univariate Deep LSTM-based Stacked Autoencoder (DLSTM-SAE) model for short-term load forecasting, equipped with a Multi-Stage Attention Mechanism (MSAM), including an input AM and several temporal AM in the pre-training phase. The input AM is used to capture the high-impact load sequence time steps of univariate input data. It should be noted that the model's performance is improved by increasing the network depth; however, finding the optimal network parameters is a challenging task due to the random assignment of the initial weights of the network. An unsupervised greedy layer-wise pre-training structure equipped with the MSAM is expanded to solve setting the random initial weight problem of the DLSTM-SAE model. The multi-stage temporal AM in the pre-training structure leads the DLSTM-SAE to properly learn the time dependencies related to remarkably long sequence input data and capture the temporal merit features lied in the LSTM memory. The performance of the proposed model is evaluated through various comparative tests with current prevalent models using actual energy market data New England ISO using three criteria indexes. The results show the superiority of the proposed model and its robustness in offline and online load forecasting.

Digital Twin Based Virtual Sensor for Online Fatigue Damage Monitoring in Offshore Wind Turbine Drivetrains

Abstract In this article a virtual sensor for online load monitoring and subsequent remaining useful life (RUL) assessment of wind turbine gearbox bearings is presented. Utilizing a Digital Twin framework the virtual sensor combines data from readily available sensors of the condition monitoring (CMS) and supervisory control and data acquisition (SCADA) system with a physics-based gearbox model. Different state estimation methods including Kalman filter, Least-square estimator, and a quasi-static approach are employed for load estimation. For RUL assessment the accumulated fatigue damage is calculated with the Palmgren–Miner model. A case study using simulation measurements from a high-fidelity gearbox model is conducted to evaluate the proposed method. Estimated loads at the considered intermediate and high-speed shaft bearings show moderate to high correlation (R = 0.50 − 0.96) to measurements, as lower frequency internal dynamics are not fully captured. The estimated fatigue damage differs by 5–15% from measurements.

Design and testing of a smart rubber stave for marine water-lubricated bearings

Abstract The water-lubricated bearing is mainly installed at the stern of the ship to support the rotation of the main shaft of the ship, which is an important component of the ship’s power plant. The rubber stave of the water-lubricated bearing acts as an elastic body, which can effectively suppress the rotational vibration of the rotating shaft and improve the shaft alignment effect. To monitor the operational status of the water-lubricated bearing at the stern of a ship, a design method of smart rubber stave for water-lubricated bearings is proposed in this article. According to the structure of the rubber stave, the packaging and protection form of the optical fiber, the forming process of the rubber stave, and the mechanical analysis were carried out to determine the distribution design principle of the fiber Bragg grating (FBG) in the rubber stave. Then, FBGs were packaged, protected, and implanted with an uneven distribution into the rubber stave to achieve force sensing in the bearing. With reference to the force state of the water-lubricated bearing during shipbuilding and ship operation, various types of condition monitoring tests were conducted on the smart rubber stave using an elastomer testing machine, which were to analyze the effects of load sizes, frequencies, and ambient temperatures. The tests showed that the FBGs in the rubber stave could monitor the static and dynamic loads of the bearing under the seawater environment or inland water environment below 30°C. This method can be used for the smart designing and online load condition monitoring of water-lubricated bearings.

Nash Social Welfare in Selfish and Online Load Balancing

In load-balancing problems there is a set of clients, each wishing to select a resource from a set of permissible ones to execute a certain task. Each resource has a latency function, which depends on its workload, and a client’s cost is the completion time of her chosen resource. Two fundamental variants of load-balancing problems are selfish load balancing (a.k.a. load-balancing games ), where clients are non-cooperative selfish players aimed at minimizing their own cost solely, and online load balancing , where clients appear online and have to be irrevocably assigned to a resource without any knowledge about future requests. We revisit both problems under the objective of minimizing the Nash Social Welfare , i.e., the geometric mean of the clients’ costs. To the best of our knowledge, despite being a celebrated welfare estimator in many social contexts, the Nash Social Welfare has not been considered so far as a benchmarking quality measure in load-balancing problems. We provide tight bounds on the price of anarchy of pure Nash equilibria and on the competitive ratio of the greedy algorithm under very general latency functions, including polynomial ones. For this particular class, we also prove that the greedy strategy is optimal, as it matches the performance of any possible online algorithm.

Cooperated Online Learning and Optimization Operation of the Low Voltage Distribution System Considering User Electrical Characteristics

With the acceleration of energy reform, photovoltaic, energy storage, electric vehicles, and other new loads in low-voltage distribution networks have been rapidly developed. However, the distribution network with distributed power supply has some problems, such as imprecise power flow modeling and difficult coordination between various energy sources and loads, which bring challenges to the online optimization of the distribution network load curve. In this study, a multi-time scale online load optimal operation scheme of the distribution network is proposed by using the Bayesian online learning method. This scheme transforms the online power optimization of the distribution network into a Markov decision process. The output time of different energy sources is different, and the load with different user load characteristics is optimized. The scheme can track the state of the distribution network in real time and make the optimization scheme of multi-energy output online. Finally, an example is given to verify the effectiveness of the proposed method, which has theoretical significance for promoting the diversified development of low-voltage user-side load.

Adaptive bus‐voltage control of ship power system with online fluctuation prediction

To ensure the power quality of a ship power system under disturbances due to changing sea states, an adaptive bus-voltage controller with online load power fluctuation prediction is proposed to stabilize the ship bus voltage. The controller uses an online sequence extreme learning machine to update the output weight matrix and achieve an accurate online prediction of the ship load power fluctuations under the disturbance of random uncertainties. A stable additional control signal is provided to the bus-voltage controller for adaptive operation. The prediction results are used to optimize the coordinated control of multiple generation systems and improve the bus-voltage control performance by updating the droop control coefficients online. The adaptability provided by the online prediction algorithm and the performance of the bus-voltage control are verified using recorded voyage data of an offshore marine engineering transport vessel.

Model-predictive energy management system for thermal batch production processes using online load prediction

Predictive energy management systems (EMS) enable industrial plants to participate in the modern power market and reduce energy cost. In this paper, a novel modular model predictive EMS specifically designed for industrial thermal batch processes is presented. The EMS consists of a two-layer mixed-integer model predictive controller and an online load predictor, and thus solves the main challenges of EMS in industry - high implementation costs and the possible reduction of production reliability. The modular formulation of the optimization problem enables system integrators to implement the EMS without time-consuming modelling tasks and elaborate parameter tuning. The online load predictor estimates the typical pulse-like heat loads of batch processes ensuring both - reliable production and maximal flexibility of the power demand. The utilization of real-time data provides additional robustness against uncertainties caused by human operators. The performance of the EMS is evaluated in a case study of an existing food plant.

Deep Learning Based Model-Free Robust Load Restoration to Enhance Bulk System Resilience With Wind Power Penetration

This paper proposes a new deep learning (DL) based model-free robust method for bulk system on-line load restoration with high penetration of wind power. Inspired by the iterative calculation of the two-stage robust load restoration model, the deep neural network (DNN) and deep convolutional neural network (CNN) are respectively designed to find the worst-case system condition of a load pickup decision and evaluate the corresponding security. In order to find the optimal result within a limited number of checks, a load pickup checklist generation (LPCG) algorithm is developed to ensure the optimality. Then, the fast robust load restoration strategy acquisition is achieved based on the designed one-line strategy generation (OSG) algorithm. The proposed method finds the optimal result in a model-free way, holds the robustness to handle uncertainties, and provides real-time computation. It can completely replace conventional robust optimization and supports on-line robust load restoration which better satisfies the changeable restoration process. The effectiveness of the proposed method is validated using the IEEE 30-bus system and the IEEE 118-bus system, showing high computational efficiency and considerable accuracy.

Online load balancing with general reassignment cost

We investigate a semi-online variant of load balancing with restricted assignment. In this problem, we are given n jobs, which need to be processed by m machines with the goal to minimize the maximum machine load. Since strong lower bounds rule out any competitive ratio of o(log⁡n), we may reassign jobs at a certain job-individual cost. We generalize a result by Gupta, Kumar, and Stein (SODA 2014) by giving a O(log⁡log⁡mn)-competitive algorithm with constant amortized reassignment cost.