Operation Planning Of Power Systems Research Articles

A hybrid load forecasting system based on data augmentation and ensemble learning under limited feature availability

Accurate power load forecasting is an important part of power system operation planning, it can ensure the stable operation of power systems and improve the efficiency of energy utilization. The power load is affected by many factors including temperature, season, population density, and so on, however due to privacy protection and other reasons, it is difficult to obtain some characteristic information that affects the load. The lack of characteristic data will reduce the accuracy of load forecasting and the generalization ability. To solve it, a new hybrid load forecasting framework is proposed, which is composed of two subsystems: a data preprocessing system and a high-precision forecasting system. Based on the load sequence itself, subsystem 1 obtains the trend data and denoising data by variational mode decomposition method, obtains the indicator variable for the weekend according to the one-hot encoding, and also introduces the electricity price data, thus obtaining the 4-dimensional extended data. Subsystem 2 constructs a hybrid prediction model by synthesizing various models, including deep learning and machine learning models, to forecast the expanded data. Finally, the multi-objective JAYA algorithm based on tent chaotic mapping and cross-perturbation strategy is used to ensemble the prediction results of the sub-models. To verify the superiority of the proposed forecasting framework, we conducted experiments using four sets of load data from New South Wales, Australia. The experimental results show that the average absolute percentage error of the hybrid framework on the four data sets are MAPEMarch=0.8070, MAPEJune=0.8296, MAPESeptember=0.7238 and MAPEDecember=0.7709, which are significantly lower than other models and provide a basis for power system scheduling management.

Green hydrogen exports in New Zealand and Chile can improve electricity supply security if configured as local energy insurance

Extreme weather events, for example, prolonged droughts, are increasingly stressing electricity systems and threatening the achievement of energy transition goals. At the same time, countries such as Chile and New Zealand are developing strategies to export renewable energy through green hydrogen. This paper economically evaluates the use of future green hydrogen exports as strategic storage under the logic of energy insurance, based on the assumption that Chile and New Zealand will adopt a hydrogen economy. Our first contribution is to compare the marginal costs of re-electrified green hydrogen, for different scenarios, to the value of lost load in New Zealand and Chile. We then determine the marginal cost of energy produced from green hydrogen based on projected production, transportation, and re-electrification costs for the year 2030. Finally, we estimate the eventual penalties that exporters would incur in the case of breaking their contracts and the overall resulting system costs of using hydrogen for system security. We found that re-electrifying hydrogen can be competitive with conventional technologies at 3 USD/kg. At 1.5 USD/kg, hydrogen is more competitive than fossil fuel-based technologies, for both New Zealand and Chile. Thus, in 2030, it could make economic sense for New Zealand and Chile to use hydrogen as a strategic reserve. The system cost of the proposed insurance scheme in 2030 ranges from 0.53 to 1.76 USD/MWh, which is small compared to total electricity costs. These values can be used for power system expansion and operation planning.

Addressing the Security risk of a Massive Deployment of Photovoltaic Power for Electric Power Systems

Future electric power systems will incorporate a high percentage of Photovoltaic (PV) power generation as a means of mitigating environmental issues including global warming and energy depletion. The author observes that if the increase in uncertainties brought about by the PV penetration cannot be prevented, then it will be difficult to carry out the N-1 security level. To ensure the N-1 security requirement is met despite the uncertainties, paper has suggested a new notion of "robust power system security" along with many conditions to be met. In this research, and offer a model of uncertainties caused by PV generation, and use robust power system security to guide our choice of parameters. Next, simulate the system to learn more about, how PV and load disconnections affect stability, and how Fault Ride Through (FRT) and Dynamic Voltage Support (DVS) influence stability. Finally, it is demonstrated that the massive penetration of PV would significantly raise the complexity of burden jobs in future power system operation planning and real-time operation.

Unit commitment with AC optimal power flow constraint

Because of the enormous economic gain it may offer, unit commitment (UC) has been a primary optimization job in the daily operation planning of contemporary power systems. The AC optimal power flow (ACOPF) constraint is a method for determining the ideal load arrangement in alternating current (AC) power systems. Using unit commitment and ACOPF limitations to determine optimal load distribution is an important approach to ensure the steady functioning of power systems. The unit commitment approach with ACOPF constraint is examined in this study. The aims and contents of the study are set by analyzing the current approaches, and a piecewise linearized unit commitment method is developed that takes into consideration the ACOPF limitations and has good accuracy and practicality. Finally, to validate the method’s effectiveness, it was tested using a six-bus test system.

Quantifying the value of probabilistic forecasting for power system operation planning

A recent key research area in renewable energy integration is the development of tools and methods to capture and accommodate the uncertainty associated with the forecast errors. While the research community has proposed numerous methods to improve the accuracy of probabilistic forecasts, their application to operational planning is still an open question. This work applies dynamic reserve determination methods to solar probabilistic forecasts and then feed them to a commercial production cost model simulator to assess the value of capturing the uncertainty endogenously in the reserve determination process. Testing is carried out on a calibrated real-size system representing the Southern Company for medium, and high solar penetration levels. Numerical results demonstrate the benefits that can be attained by explicitly modeling probabilistic uncertainty in terms of operating cost, and enhanced system reliability which is measured as the quantity of balancing and reserve violations. Additionally, these methods and results can pave the way for system operators to adopt probabilistic forecasting to draw the operating plans of the system, and this allowing the successful integration of variable renewable energy sources.

Short-Term Power Load Forecasting Based on an EPT-VMD-TCN-TPA Model

Accurate short-term load forecasting is the key to ensuring smooth and efficient power system operation and power market dispatch planning. However, the nonlinear, non-stationary, and time series nature of load sequences makes load forecasting difficult. To address these problems, this paper proposes a short-term load forecasting method (EPT-VMD-TCN-TPA) based on the hybrid decomposition of load sequences, which combines ensemble patch transform (EPT), variational modal decomposition (VMD), a temporal convolutional network (TCN), and a temporal pattern attention mechanism (TPA). In which, the trend component (Tr(t)) and the residual fluctuation component (Re(t)) of the load series are extracted using EPT, and then the Re(t) component is decomposed into intrinsic modal function components (IMFs) of different frequencies using VMD. The Tr(t) and IMFs components of the fused meteorological data are predicted separately by the TCN-TPA prediction model, and finally, the prediction results of each component are reconstructed and superimposed to obtain the final predicted value of the load. In addition, experiments after reconstructing each IMF component according to the fuzzy entropy (FE) values are discussed in this paper. To evaluate the performance of the proposed method in this paper, we used datasets from two Areas of the 9th Mathematical Modeling Contest in China. The experimental results show that the predictive precision of the EPT-VMD-TCN-TPA model outperforms other comparative models. More specifically, the experimental results of the EPT-VMD-TCN-TPA method had a MAPE of 1.25% and 1.58% on Area 1 and Area 2 test sets, respectively.

Non-Dominated Sorting-Based Hybrid Optimization Technique for Multi-Objective Hydrothermal Scheduling

Short-term hydrothermal scheduling problem plays an important role in maintaining a high degree of economy and reliability in power system operational planning. Since electric power generation from fossil fired plants forms a major part of hydrothermal generation mix, therefore their emission contributions cannot be neglected. Hence, multi-objective short term hydrothermal scheduling is formulated as a bi-objective optimization problem by considering (a) minimizing economical power generation cost, (b) minimizing environmental emission pollution, and (c) simultaneously minimizing both the conflicting objective functions. This paper presents a non-dominated sorting disruption-based oppositional gravitational search algorithm (NSDOGSA) to solve multi-objective short-term hydrothermal scheduling (MSHTS) problems and reveals that (i) the short-term hydrothermal scheduling problem is extended to a multi-objective short-term hydrothermal scheduling problem by considering economical production cost (EPC) and environmental pollution (EEP) simultaneously while satisfying various diverse constraints; (ii) by introducing the concept of non-dominated sorting (NS) in gravitational search algorithm (GSA), it can optimize two considered objectives such as EPC and EEP simultaneously and can also obtain a group of conflicting solutions in one trial simulation; (iii) in NSDOGSA, the objective function in terms fitness for mass calculation has been represented by its rank instead of its EPC & EEP values by using the NS approach; (iv) an elite external archive set is defined to keep the NS solutions with the idea of spread indicator; (v) the optimal schedule value is extracted by using fuzzy decision approach; (vi) a consistent handling strategy has been adopted to handle effectively the system constraints; (vii) finally, the NSDOGSA approach is verified on two test systems with valve point loading effects and transmission loss, and (viii) computational discussion show that the NSDOGSA gives improved optimal results in comparison to other existing methods, which qualifies that the NSDOGSA is an effective and competitive optimization approach for solving complex MSHTS problems.

An Analytical Formulation for Mapping the Spatial Distribution of Nodal Inertia in Power Systems

Mapping the spatial distribution of nodal inertia is crucial for helping system operators identify locational frequency stability issues in the power system operation planning process. Currently, index-based methods cannot retrieve the nodal inertia values for all load buses. To do so, this paper proposes an accurate analytical formulation relying only on steady-state system parameters to determine the nodal inertia values and, thus, map the spatial distribution of system inertia. The performance and reproducibility of the proposed formulation are evaluated using three test systems: a 5-bus radial system with two synchronous generators, a multimachine IEEE 68-bus benchmark system, and a larger NPCC 140-bus test system. We validate the proposed spatial distribution of nodal inertia using time-domain simulations and numerical estimations. In addition, we compare our method against the inertia distribution indexes presented in the literature. The results indicate that our formulation adequately quantifies the nodal inertia value in system buses, with low computational cost, and providing a suitable tool for operation planning analysis.

Optimal adjustment of control parameters in short‐term operational planning of power systems based on cost‐effective security solutions

Optimal adjustment of control parameters in short‐term operational planning of power systems based on cost‐effective security solutions

Forecasting Short-Term Electricity Load Using Validated Ensemble Learning

As short-term load forecasting is essential for the day-to-day operation planning of power systems, we built an ensemble learning model to perform such forecasting for Thai data. The proposed model uses voting regression (VR), producing forecasts with weighted averages of forecasts from five individual models: three parametric multiple linear regressors and two non-parametric machine-learning models. The regressors are linear regression models with gradient-descent (LR), ordinary least-squares (OLS) estimators, and generalized least-squares auto-regression (GLSAR) models. In contrast, the machine-learning models are decision trees (DT) and random forests (RF). To select the best model variables and hyper-parameters, we used cross-validation (CV) performance instead of the test data performance, which yielded overly good test performance. We compared various validation schemes and found that the Blocked-CV scheme gives the validation error closest to the test error. Using Blocked-CV, the test results show that the VR model outperforms all its individual predictors.

Distribution-free chance-constrained load balance model for the operation planning of hydrothermal power systems coupled with multiple renewable energy sources

We are interested in the potential risk of short-term load imbalance in hydro-thermal power systems under high integration of renewable energy sources, including wind, solar, and biomass. Toward this end, the short-term load balance across the system is formulated as a joint chance constraint. Due to non-convexity issues and the lack of information about the joint probability distribution of the renewable energy resources, we analyze a distribution-free robust relaxation and a p-efficiency approximation of the joint chance constraint. Our proposal is embedded within the widespread SDDP algorithm, and compared with the popular CVaR scheme on a simplified version of the Colombian power system.

Modelling of renewable energy power plant controllers for steady-state studies using an extended power flow formulation

The increasing penetration of large-scale wind and photovoltaic plants has created new challenges for the secure and reliable operation of power systems. To face these challenges, these renewable plants are composed of numerous inverter-based resources that use centralized controllers to satisfy grid code requirements related to frequency and voltage control, both under dynamic and steady-state conditions. In this paper, an extended power flow formulation is proposed for the steady-state modelling of power plant controllers along with their power flow controls including those associated with droop control, suitable for any plant topology and number of regulation devices. The effectiveness of this modelling approach is validated against WECC generic models using dynamic simulations in PSS®E. For this purpose, a 13-bus 5-generator test system representing a 100-MW renewable power plant is studied. The accuracy of the introduced formulation is demonstrated since it presents absolute errors smaller than 0.2% between both fundamentally different methods. Additionally, the well-known IEEE RTS-96 test system is used to model 29 power plant controllers with 93 generators, thus showing the practicality of the proposed formulation. It can be envisaged that this timely tool is valuable for power system operation and expansion planning considering large-scale variable renewable energy power plants.

Development a New Flexibility Index Suitable for Power System Operational Planning

Power system flexibility is an important characteristic in both power system planning and operation, which should be evaluated and maintained in the desired value. On the other hand, more renewable energy integration leads to increasing uncertainty and variability in the power system. Therefore, the power system should have the sufficient ability to overcome the adverse effects of uncertainty and variability named as flexibility, which should be improved by suitable tools such as adequate reserve, fast ramp up/down generation sources and suitable energy storage capacity. Power system flexibility evaluation is the main task that needs suitable indices to indicate the level of system flexibility correctly. In the current paper, a well-known system flexibility index named normalized flexibility index, which is used for power system planning horizon is modified to use for the operational planning time zone. In this concept, the flexibility index is separated into two components, each of them indicating the ability of the power system to withstand upward/downward net-load uncertainty and variability. In the further, this is shown these two components are the same as the upward/downward system reserve and can be converted to economic value simply. So, this concept facilitates the economic trade-off between operation cost and system flexibility, improving cost to achieve the best level of system flexibility.

Sequencing paths of optimal control adjustments determined by the optimal reactive dispatch via Lagrange multiplier sensitivity analysis

Optimal power flows play a key role in power system operation planning. While most papers in the literature focus on attaining optima, sequencing paths of optimal control adjustments that lead the system from an initial operating point towards the optimum remain scarcely accounted for. Thus, this work proposes a practical framework based upon power system steady-state analysis for sequencing strictly feasible paths of optimal control adjustments determined by the Optimal Reactive Dispatch (ORD) via Lagrange multiplier sensitivity analysis. The proposed framework is methodologically founded on the reformulation of the ORD in terms of optimal control adjustments rather than optimal control values, successive Newton’s power flow calculations to assure a strictly feasible path from the initial operating point towards the optimum, and successive resolutions of the reformulated ORD’s associated dual problem to determine Lagrange multipliers along such sequence path. Thus, pondering optimal control adjustments by their respective Lagrange multipliers indicates which control action must be realised. Numerical results for IEEE test-systems with up to 300 buses with an increased number of controllable variables are obtained to validate and illustrate the efficiency and robustness of the proposed framework.

Automated approach to power system operational planning and congestion management for a system with high levels of renewable generation

Automated approach to power system operational planning and congestion management for a system with high levels of renewable generation

A Full-Newton AC-DC Power Flow Methodology for HVDC Multi-Terminal Systems and Generic DC Network Representation

The use of Direct Current (DC) transmission links in power systems is increasing continuously. Thus, it is important to develop new techniques to model the inclusion of these devices in network analysis, in order to allow studies of the operation and expansion planning of large-scale electric power systems. In this context, the main objective of this paper is to present a new methodology for a simultaneous AC-DC power flow for a multi-terminal High Voltage Direct Current (HVDC) system with a generic representation of the DC network. The proposed methodology is based on a full Newton formulation for solving the AC-DC power flow problem. Equations representing the converters and steady-state control strategies are included in a power flow problem formulation, resulting in an expanded Jacobian matrix of the Newton method. Some results are presented based on HVDC test systems to confirm the effectiveness of the proposed approach.

A Methodology for Provision of Frequency Stability in Operation Planning of Low Inertia Power Systems

Along with the increasing share of non-synchronous power sources, the inertia of power systems is being reduced, which can give rise to frequency containment problems should an outage of a generator or a power infeed happen. Low system inertia is eventually unavoidable, thus power system operators need to be prepared for this condition. This paper addresses the problem of low inertia in the power system from two different perspectives. At a system level, it proposes an operation planning methodology, which utilises a combination of power flow and dynamic simulation for calculation of existing inertia and, if need be, synthetic inertia (SI) to fulfil the security criterion of adequate rate of change of frequency (RoCoF). On a device level, it introduces a new concept for active power controller, which can be applied virtually to any power source with sufficient response time to create synthetic inertia. The methodology is demonstrated for a 24 h planning period, for which it proves to be effective. The performance of SI controller activated in a battery energy storage system (BESS) is positively validated using a real-time digital simulator (RTDS). Both proposals can effectively contribute to facilitating the operation of low inertia power systems.

Multi-Convolution Feature Extraction and Recurrent Neural Network Dependent Model for Short-Term Load Forecasting

Load forecasting is critical for power system operation and market planning. With the increased penetration of renewable energy and the massive consumption of electric energy, improving load forecasting accuracy has become a difficult task. Recently, it was demonstrated that deep learning models perform well for short-term load forecasting (STLF). However, prior research has demonstrated that the hybrid deep learning model outperforms the single model. We propose a hybrid neural network in this article that combines elements of a convolutional neural network (1D-CNN) and a long short memory network (LSTM) in novel ways. Multiple independent 1D-CNNs are used to extract load, calendar, and weather features from the proposed hybrid model, while LSTM is used to learn time patterns. This architecture is referred to as a CNN-LSTM network with multiple heads (MCNN-LSTM). To demonstrate the proposed hybrid deep learning model’s superior performance, the proposed method is applied to Ireland’s load data for single-step and multi-step load forecasting. In comparison to the widely used CNN-LSTM hybrid model, the proposed model improved single-step prediction by 16.73% and 24-step load prediction by 20.33%. Additionally, we use the Maine dataset to verify the proposed model’s generalizability.

Predispatch Linear System Solution With Preconditioned Iterative Methods

Primal-dual interior point method is used to minimize the predispatch generation costs and transmission losses on short-term operation planning of hydrothermal power systems with previously scheduled maneuvers and ramp rate constraints. Despite the efficiency shown by interior point methods for very large-scale problems, they generally perform only reasonably when applied to multiple-dimensional flow problems, the new specialized interior point algorithm performed here, overcomes this disadvantage. This specialization uses the preconditioned conjugated gradient method with an incomplete Cholesky factorization to solve a linear system in each iteration of the algorithm. The preconditioner developed exploring the structure of the problem is fundamental to ensure the efficiency of the method.

A Cross-Domain Approach to Analyzing the Short-Run Impact of COVID-19 on the US Electricity Sector.

The novel coronavirus disease (COVID-19) has rapidly spread around the globe in 2020, with the US becoming the epicenter of COVID-19 cases since late March. As the US begins to gradually resume economic activity, it is imperative for policymakers and power system operators to take a scientific approach to understanding and predicting the impact on the electricity sector. Here, we release a first-of-its-kind cross-domain open-access data hub, integrating data from across all existing US wholesale electricity markets with COVID-19 case, weather, mobile device location, and satellite imaging data. Leveraging cross-domain insights from public health and mobility data, we rigorously uncover a significant reduction in electricity consumption that is strongly correlated with the number of COVID-19 cases, degree of social distancing, and level of commercial activity.