Data-Driven Operation Cost Modeling of Power System Considering Wind Power Fluctuation

With the increasing integration of distributed energy resources (DER) and corresponding fluctuations, phase shifting transformers (PSTs) have been considered as important electrical components to increase transmission capacity by integrating PST with optimal power flow (OPF). However, due to vulnerabilities of supervisory control and data acquisition (SCADA) systems, applications in power systems may be subject to various cyber-attacks. In this paper, we discuss the feasibility of load redistri-bution (LR) attacks against OPF with PST, and quantitatively evaluate the effect of LR attacks on power systems. Specifically, the interaction between LR attacker and OPF with PST is formulated as a Stackelberg game, where the attacker and power system, acting as the leader and follower, attempt to maximize and minimize the operation cost of power system, respectively. By replacing the lower OPF optimization with its Karush-Kuhn-Tucker (KKT) condition, the Stackelberg game is simplified into a single-level mixed integer programming, which can be easily solved. Simulations in the PJM 5-bus, IEEE 14-bus and IEEE 30-bus test systems verify the feasibility of LR attacks and quantitatively evaluate the effect of LR attacks on the operation cost of power systems.

Accurate wind power prediction can reduce the negative impact of wind power on power grid and reduce the operation cost of power system. Wind power forecasting is of great significance in spot trading, but the instability of wind power series brings difficulties to forecasting. With the rapid development of deep learning technology, prediction algorithm based on deep learning has been widely used in wind power prediction. The long short-term memory (LSTM) network performs well in a variety of prediction scenarios, but there is still room for improvement. In order to improve the prediction accuracy of LSTM in wind power prediction, this paper adopts a new gating mechanism in LSTM network, which can improve the context information modeling ability of LSTM network. The experimental results based on real data show that the proposed improved LSTM prediction model can improve the prediction accuracy of LSTM prediction model in ultra-short term wind power prediction scenarios.

The escalating grid-connected capacity of renewable energy sources, predominantly wind and photovoltaic (PV) power, along with its inherent volatility and anti-peaking attributes, exacerbates the peaking demands on the power system. Consequently, this trend necessitates enhanced flexibility in power system peaking capabilities. In response to this challenge, this paper introduces an optimal scheduling methodology grounded in a two-stage stochastic model tailored for power systems, which incorporates thermal-storage peaking pricing. Initially, a hierarchical decision-making framework, employing the group decision hierarchy analysis method, is devised to formulate a peaking pricing strategy for thermal power and energy storage. Subsequently, acknowledging the variability of wind power, multiple scenarios are generated and consolidated through the application of k-means clustering. A two-stage stochastic optimization approach is then utilized for day-ahead pre-dispatch of thermal power and storage units, and intraday dispatch adjustments are made to accommodate the uncertainties associated with renewable energy sources by leveraging flexible resources, thereby ensuring power balance, and maximizing wind power integration. The efficacy of the proposed approach is validated through simulations conducted on the enhanced IEEE39-bus system. The numerical results show that the flexibility of the system can be effectively improved, and the wind power and photovoltaic accommodation rate is increased by 98.11 and 99.42%, respectively.

In order to reasonably quantify the influence of wind and photovoltaic power output uncertainty on optimal scheduling, a day-ahead optimal scheduling model of wind-photovoltaic-thermal-energy storage combined power generation system considering opportunity-constrained programming is established. The model takes the system operation cost, which contains the operation cost of thermal power units, the penalty cost of pollutants, the operation cost of energy storage system and the penalty cost of new energy curtailment, as the optimization goal. In order to characterize the influence of wind and photovoltaic power prediction errors on the optimal scheduling, an opportunity-constrained programming method is used to calculate the rotating reserve capacity of the system in each period. The simulation calculation is performed on a typical day. The results show that the established model can realize the multi-energy complementary operation of the wind-photovoltaic-thermal-energy storage system. By reasonably quantifying the influence of the wind and photovoltaic power prediction error on the optimal scheduling, the system operation cost can be reduced, which verifies the validity of the model.

The growing share of renewable energies needs more flexible services to balance their intermittency and variance. The existing coal fired units and electrical energy storage (EES) systems may play an important role in delivering flexible services. The value of their flexibility services, along with the value of renewable energies, has to be analyzed from the perspective of the power system, in which the capacity costs and operation costs of renewable energy power units, EES systems, and thermal power generation units have to be taken into consideration. An optimal model is built to analyze the renewable energy integration and the flexibility services delivered by the EES systems and thermal power units in a power system. Taking the existing thermal power units and EES systems in North China Power Grid as an instance, the overall cost of the grid is examined for the penetration of renewable energies and flexible service provision. The results show that the growing shares of renewable energies are affected by their capacity credits and flexibility sources in the grid, and that the potential of thermal power units to provide flexible services will be reduced due to the replacement of renewable energies for thermal power generation. The results also indicate that the thermal units may be dispatched to have priority to delivering flexible services for the renewable energy integration, and that the curtailment of renewable energies may be regarded as one type of flexible service. According to these results, policy and strategy recommendations are put forward to weigh the role of existing coal-fired units and EES systems in providing flexible services, and to improve their compensation mechanism and their coordination.

Guiding the coordinated charging of electric vehicles can alleviate the load fluctuation of power systems within a local area caused by uncoordinated charging of electric vehicles and greatly reduce the cost of power system operation. This will become an inevitable development trend of future energy system transformation. In this paper, a new mobile application is built to realize the dynamic adjustment of electric vehicle charging prices according to the change in weather conditions to guide the coordinated charging of electric vehicles. After systematically introducing the structure and data flow process of the application, we simulate the fluctuation of charging prices under various weather conditions using the electricity load data of North China and verify the good performance of the application. We believe that this application can help power systems to achieve low-carbon transformation by adopting a new dynamic time-of-use pricing charging model.

Marine Predator Algorithm (MPA) is an optimization algorithm inspired by the behavior of predator and prey to catch their own food. MPA is simple and easy to implement. To further improve the performance of MPA, this paper proposes a Multigroup Marine Predator Algorithm (MGMPA). The multigroup mechanism is to divide the initial population into several independent groups. These groups generate the top predator and the Elite matrix based on different strategies and share information after a fixed iteration. Above strategies include the maximum of the same group, the average of the same group, the maximum of different groups and the average of different groups. To verify its performance, the paper compares MGMPA with some classic algorithms such as Particle Swarm Optimization (PSO), Parallel Particle Swarm Optimization (PPSO), Slap Swarm Algorithm (SSA), and Marine Predator Algorithm (MPA). In addition, the proposed MGMPA is also applied to solve Economic Load Dispatch problem (ELD). The experimental results show that the proposed MGMPA has significant advantages under the CEC2013 suite and obtains the minimum cost of power system operation and the maximum economic benefits in the application.

In this article, the optimal transmission switching (OTS) problem is solved in order to optimize the operation cost of an electrical power system. This is accomplished by disconnecting some transmission lines, which enables a change to the profile of the power flow distribution in the system, allowing for increased generation at the buses with lower costs; thus, the hourly operation cost of the generation is minimized to meet the demand of the system. This article presents contributions to topics related to the issue of the high number of transmission lines that are disconnected when the OTS problem is solved, the problem of system islanding, and the causes of Braess's paradox, in the context of the OTS problem. Finally, tests are conducted using the 41-bus southern Brazilian system and the 92-bus Colombian system. The results demonstrate the effectiveness of the proposals for reducing the number of lines disconnected from the system and for avoiding islanding.

In this paper, The electric power dispatching equivalent model (DEM) of the integrated energy systems (IES) and the global dispatching optimization issue with IES DEM integrated of the electric power system are studied. The electric power DEM of the IES is constructed as two models: one is the dispatching cost equivalent model (DCEM), and the other one is the dispatching feasibility equivalent (DFEM) model, considering that an electric power dispatching curve for IES is not always feasible. The DCEM is trained by deep neural network (DNN) model based on the data sets of the electric power dispatching curve and dispatching cost. The DFEM is trained by ensemble classifier model based on the data sets of the electric power dispatching curve of the IES and feasibility flag. Global dispatching optimization model is built with IES DEM integrated and solved by genetic algorithm. Case studies are carried out to verify the proposed electric power DEM and the global dispatching optimization model, and results show that the proposed IES DEM makes it possible for the power dispatching center to take advantage of the IES’s dispatchability so as to minimize the total operation cost of the entire electric power system.

The short-term wind power scenarios have a significant impact on the operation cost and power system reliability due to the stochastic generation scheduling of wind-integrated power systems. In order to obtain the scenarios containing the information of forecast error distribution and fluctuation distribution for short-term wind power, a scenario generation method is proposed. This paper characterizes forecast error via empirical distributions of a set of forecast bins and assumes that wind power fluctuations over unit interval follow t location-scale distribution. An inverse transform sampling from a multivariate normal distribution is adopted to generate a large number of wind power scenarios. The covariance matrix of the multivariate normal distribution is estimated to fit the distribution of historical wind power fluctuations. The proposed scenario generation method is applied to the actual aggregate wind power data in the whole regions of Ireland's Power System. The results indicate that the variability of wind power scenarios can be adjusted by estimating the key range parameter in the exponential covariance structure of a multivariate normal distribution.

Smart Grid has two main objectives on both supply and demand aspects which are to distribute the renewable energy sources on supply side and to develop realtime price responses on demand side.BRRenewable energy does not consume fossil fuels, therefore it improves the eco-friendliness and saves the cost of power system operation at the same time. Demand response increases the flexibility of the power system by mitigating the fluctuation from renewable energies, and reduces the capacity investment cost by shedding the peak load to off-peak periods. Currently Smart Grid technologies mainly focus on energy monitoring and display services but it has been proved that enabling technologies can induce the higher demand responses through many pilot projects in USA. On this context, this paper provides a price responsive algorithm for HEMS (home energy management system) on the real time pricing environment. This paper identifies the demand response as a core function of HEMS and classifies the demand into 3 categories of fixed, transferable, and realtime responsive loads which are coordinated and operated for the utility maximization or cost minimization with the optimal usage combination of three kinds of demand.

Aiming at the large-scale wind power transmission end grid wind and thermal power shared transmission channel decoupling out, the wind power output power is blocked, the transmission channel utilization rate is not high and so on. Based on a comparative analysis of wind and fire electrolytic coupling / coupling transmission operation characteristics, this paper The goal is to maximize the output power of wind power, maximize the utilization of transmission channels, and minimize the operating costs of thermal power units. Aiming at the dispatching output of wind power and thermal power units as optimization variables, a multi-objective optimization model of coupled wind power and thermal power transmission is proposed. The target differential evolution algorithm solves the model, and finally obtains a set of optimized scheduling output plans for wind and thermal power, which provides a scientific basis for operators to make scheduling decisions. Finally, a regional power grid is taken as an example to prove the rationality of the model and the effectiveness of the proposed algorithm.

As more wind power and other renewable resources are being integrated into the electric power grid, the forecast uncertainty brings operational challenges for the power system operators. In this report, different operational strategies for uncertainty management are presented and evaluated. A comprehensive and consistent simulation framework is developed to analyze the performance of different reserve policies and scheduling techniques under uncertainty in wind power. Numerical simulations are conducted on a modified version of the IEEE 118-bus system with a 20% wind penetration level, comparing deterministic, interval, and stochastic unit commitment strategies. The results show that stochastic unit commitment provides a reliable schedule without large increases in operational costs. Moreover, decomposition techniques, such as load shift factor and Benders decomposition, can help in overcoming the computational obstacles to stochastic unit commitment and enable the use of a larger scenario set to represent forecast uncertainty. In contrast, deterministic and interval unit commitment tend to give higher system costs as more reserves are being scheduled to address forecast uncertainty. However, these approaches require a much lower computational effort Choosing a proper lower bound for the forecast uncertainty is important for balancing reliability and system operational cost in deterministic and interval unit commitment. Finally, we find that the introduction of zonal reserve requirements improves reliability, but at the expense of higher operational costs.

The power system has experienced significant changes due to digitalization, decarbonization, and decentralization, leading to the emergence of the smart grid concept. Metaverse is a virtual realm that improves the efficiency and cost of smart power system operation and development in various ways that are rarely considered in the literature. This paper aims to integrate the metaverse into the smart grid architecture model and explain its various applications through different use cases. It also examines the challenges and issues related to the metaverse deployment within the smart grid.

In order to effectively solve the problem of wind and solar energy curtailment or load shedding caused by the insufficient regulation capacity of traditional power sources in renewable energy high-penetration power systems, a generalized source-storage system is proposed. It centralizes the scattered renewable energy and then shifts it to peak load period, so that the power output can follow the load changes of the power grid. Thus, the purpose of using renewable energy installed capacity to properly supplement the power shortage needs is achieved. Firstly, according to the operating status of traditional power systems, a schedulable range constraint that takes into account the marginal amount of net load indicators is proposed. Then, in order to reduce the prediction errors caused by wind power fluctuations, a mathematical model of the energy storage system and a centralized control strategy are established. Finally, an optimal scheduling model for wind and fire storage systems with energy storage systems is established, and a hybrid particle swarm algorithm combining standard particle swarm and simulated annealing algorithms are used to solve the problem. The results of calculation examples show that the power system model considering the constraints of the schedulable range effectively reduces the feasible range of the variables and improves the speed of solving the model. In addition, the centralized control strategy that considers distributed energy storage to participate in the optimal dispatch. Compared with the traditional strategy, while improving the accuracy of the control, the solution results are further in line with the actual situation, which provides a theoretical basis for the optimal dispatching of the actual complex power system.