Search

Compared with time-domain simulation (TDS), data-driven models show great advantage in time consumption of the power system security analysis. This paper proposes a novel graph neural network (GNN) based model for the short-term voltage stability (STVS) assessment. Based on mainly the steady-state information as the inputs, the model can provide multivariate stability indices for each bus and adapt to minor topological changes. Both the fast voltage collapses (FVC) and fault-induced delayed voltage recovery (FIDVR) events can be identified. A modified multi-task learning (MTL) process improves the multi-index evaluation performance of the model. As the kernel of the model, a heterogeneous graph attention deep network (HGAT) with various types of nodes works to aggregate the local information and generate new features. Then several micro-detectors scan over the network to provide the STVS indices of each bus with only the feature vector of one node as input for the prediction. Comparisons on the test system show that the proposed method can provide fine-grained STVS evaluation and show good robustness in the scenarios of different dynamic load distribution and network topology changes.

Phasor measurement units (PMUs) have been widely deployed in modern power grids, which provides a great opportunity for high-precision and real-time power system state estimation. If power networks are completely observable by PMUs, linear state estimation can be achieved with massive synchronous data. In this paper, the analysis of phasor error is presented and a complex number weighted least squares (CWLS) algorithm for state estimation is proposed, which shows high computational efficiency and stronger robustness to phasor errors. Meanwhile, self-adaptive weight and bad data identification techniques are developed to enhance the CWLS’s performance. Test results on various IEEE test systems and the realistic power grid validate the effectiveness and superiority of the proposed CWLS algorithm, outperforming the conventional weighted least squares (WLS) state estimation method as well as other existing industrial-grade approaches.

Reliable and fast transient stability assessment (TSA) is significantly required for the power system emergency control. We propose a topology adaptive high-speed transient stability assessment (HSTSA) scheme, where the inputs of the model adopt only the pre-fault state and the dynamics at the fault occurrence snapshot. A novel multi-graph attention network with residual structure (ResGAT) is designed to capture the stability characteristics. ResGAT applies improved graph attention mechanism to enhance its adaptability to the power system topology changes and the residual structure helps to avoid network degeneration. Meanwhile, a new piece-wise transient stability index (PSI) is proposed for the stability level prediction. Integration of both the stability category and the stability level results increase the precision of the HSTSA scheme. Test results on IEEE 39 Bus system and IEEE 300 Bus system indicate the superiority of the proposed scheme over existing models and its robustness under various scenarios. .

With respect to the features in optimal scheduling of long-term power generation such as long cycle, large scale and strong randomness, the paper proposed a multi-stage optimized decision model based on approximate dynamic programming. In this model, the influence of random variables including load, fuel price, and water was tested by means of stochastic simulations. Staged solution was used to reduce the size and difficulty of solving the problem. The proposed value function approximation strategy of pondage and coal inventory solved the "curse of dimensionality" in staged recursion and maintained the overall optimization characteristics after staged decomposition. Through the iterative update between the decision and the approximation function, the approximate optimal decision sequence was solved to obtain the optimization scheme of power generation scheduling. The calculation and analysis results in a real system in some province showed that the modeling by approximate dynamic programming is an effective way to solve such large-scale optimization and has a promising future by virtue of its conciseness, effectiveness, convenience in dealing with random factors, high calculation accuracy and rapidity of solution.

With the large-scale development of the power industry, new requirements are put forward for the stable operation of the power system. Power system load forecasting refers to the use of historical load data to predict the future load value, which is an important part of energy management system. The short-term load forecasting process is often combined with the basic mechanism of power grid dispatching to achieve the balance of power grid supply and demand, reflecting the highly nonlinear computing ability. Power load forecasting is the premise of power grid real-time control, operation planning and development planning. At present, the grey model model used for power system load forecasting generally has the problems of large calculation amount, no mature theoretical basis for selecting structures and parameters, etc. This paper discusses the application of grey model in short-term power load forecasting, and puts forward a principal component analysis method suitable for ordinary daily power load forecasting data, which improves the accuracy of short-term power load forecasting.

This paper proposes a robust optimal reactive power flow (RORPF) model to cope with uncertainties of load perturbations. It incorporates information of load-increase directions to reduce the uncertain effect of system loads, searching for solutions that are immune to parameter drifts and load changes. According to the numerical study on IEEE-118 systems, optimal solutions of RORPD are superior in the presence of uncertain load perturbations. The robust solutions are proved to be comparatively less sensitive to load perturbations and can maintain their objective values against uncertain load changes. RORPD can provide optimal solutions with a higher degree of stability in the face of perturbations and can be more practical in the reactive power optimization of real-time operation systems.

With the deregulation of power market and integration of large-scale intermittent renewable generation, the operation of modern power system is faced with more and more uncertain factors. Traditional deterministic methods are hard to meet the requirements for the analysis and control of power system under uncertainty. Therefore, it's necessary to introduce risk assessment methods into the daily operation analysis of power systems to help power system operators make more scientific decisions. In this paper, a new software of grid operation risk assessment named GORA is presented, which is able to take into various uncertain factors such as faults, equipment status and weather. Several measures are taken by GORA to meet the requirements of on-line risk assessment for power system. Besides daily rolling analysis of operation risk, GORA provides comprehensive risk analysis of power system at each specified time interval, such as sensitivity analysis, fault status analysis and weak point analysis, which is useful for power system operators to find the root of risk and take risk control measures. The correctness and effectiveness of the developed software are verified by an actual power grid in China.

Integration of wind power generation is increasing rapidly worldwide, hence bringing more challenges to the safe operation of the power system. In this paper, the basic construction method of dynamic security region (DSR) boundary is introduced. A model is established based on DSR, whose object is the minimization of controlling cost in dispatching operations. The coordinated control strategy based on DSR is proposed, with which transient risk assessment of the power system connecting wind farms is studied. The risk indexes calculation and comprehensive analysis method of wind power integration system are defined based on DSR, which can give quantitative indicators to the system day-ahead scheduling plan and present the risk sources as well as weak buses information. Test results on the New England 10-genrator 39-bus system are presented and prove the effectiveness of this approach.

Recently, many aspects of power system dynamic analysis, including oscillation identification, load model identification and system state estimation, are conducted based on the measurement data from phasor measurement unit (PMU). However, the impact of PMU measurement error, especially in power system dynamic process, has an assignable impact on these analysis results and thus cannot be neglected. In this paper, the requirements from IEEE standards and the analyze method for PMU measurement error are proposed first. Then, the ambient signal based identification method and the impact of measurement error for dynamic load model parameters is introduced. Afterwards, the stochastic distribution and frequency spectrum of PMU measurement error are measured through an experiment carried in a power system hardware simulation platform. Finally, the impact of PMU measurement error on the identification results of load model is simulated in WSCC 9 bus system.

Load modeling is an important but complicated task for power system analysis and simulation. In previous research, much work has been done on load model parameter identification. However, the effectiveness of the identified load model parameters still need further validation. In practical situation, validation of load model parameters is commonly based on Post-Disturbance Simulation method, which has some inadequacies actually. Learning from the idea of load model parameter identification, a decision tree based method is proposed to validate the load model parameters in this paper. The structure and parameters of load model are selected first. Then, the parameters are discretized and represented as different classes so that the iteration of shapelet searching is carried on until all the samples are classified correctly. Finally, the decision trees are built using C4.5. Extensive simulation results in the CEPRI 36 bus system have demonstrated its feasibility and validity.

For China Southern Power Grid (CSG), which has the feature of long distance power transmission with strong DC system and weak AC interconnection, low-frequency oscillation is more serious than before for specific operating situations. Therefore, using DC modulation technique to suppress inter-area low-frequency oscillation becomes a very important topic. This paper investigates low-frequency oscillation problem of CSG, and carries out system modal analyses to find out dominant inter-area oscillation modes, and then applies HVDC modulation to damping system oscillations. Gain optimization of modulation controllers of Yunnan-Guangdong DC systems is carried out to improve system damping performance. Simulations with and without modulation controllers are performed in frequency-domain and time-domain to verify the effectiveness of controllers.

China Southern Power Grid Company (CSG) and eMIT, LLC recently developed and implemented an online risk-based security assessment system-Probabilistic Security Risk Assessment (PSRA). PSRA calculates the probability of undesirable events in the transmission system and evaluates their corresponding severities. PSRA combines the probability measure of each undesirable event with the consequence measure into a single reliability index-Probabilistic Loss of Load Index (PLLI). When PSRA determines the system condition to be insecure, it optimally re-dispatches the generation to ensure the system goes back to the secure region. Under emergency conditions, when generation re-dispatch alone is not enough, load shedding will be enforced to bring the system back to the secure region. The salient aspect of PSRA is automatic application of program-determined remedial actions inclusive of load curtailment. Weak point analysis is developed and used to identify the buses and branches that are most vulnerable. A novel method is developed to visualize the Probabilistic Loss of Load Index, which makes it easy for the system operators to make decisions within a shorter time period. Online implementation of PSRA is also discussed and it is concluded that the proposed probabilistic reliability analysis is much more accurate than traditional methods and it provides system operators the capability to make faster trade-off decisions between economy and technical constraints.

Dynamic security region (DSR), containing lots of security information, can be used for emergence control and security assessment of power system, and it draws more and more attention in recent years. To obtain the boundary of DSR, the direct and approximation method are used, and the latter is the most credible way. This paper makes some improvements on the approximation of DSR mainly based on linear regression theory. Two goodness-of-fit indicators are introduced to test the fittingness of hyper-plane, as an alternative to maximum error used generally. Then utilize the total partial regression sum of squares theory to filter out the unnecessary nodes, in which way the DSR will be more reliable. Examples on the 10-generator, 39-bus New England Test System show the practicality of these methods.