Migratable Power System Transient Stability AssessmentMethod Based on Improved XGBoost

The data-driven transient stability assessment(TSA)of power systems can predict online real-time prediction by learning the temporal features before and after faults.However,the accuracy of the assessment is limited by the quality of the data and has weak transferability.Based on this,this paper proposes a method for TSA of power systems based on an improved extreme gradient boosting(XGBoost)model.Firstly,the gradient detection method is employed to remove noise interference while maintaining the original time series trend.On this basis,a focal loss function is introduced to guide the training of theXGBoostmodel,enhancing the deep exploration of minority class samples to improve the accuracy of the model evaluation.Furthermore,to improve the generalization ability of the evaluation model,a transfer learning method based on model parameters and sample augmentation is proposed.The simulation analysis on the IEEE 39-bus system demonstrates that the proposed method,compared to the traditional machine learning-based transient stability assessment approach,achieves an average improvement of 2.16%in evaluation accuracy.Specifically,under scenarios involving changes in topology structure and operating conditions,the accuracy is enhanced by 3.65%and 3.11%,respectively.Moreover,the model updating efficiency is enhanced by 14–15 times,indicating the model’s transferable and adaptive capabilities across multiple scenarios.

Physics-informed transient stability assessment of microgrids

With the integration of a voltage source converter(VSC),having variable internal voltages and source impedance,in a microgrid with high resistance to reactance ratio of short lines,angle-based transient stability techniques may find limitations.Under such a situation,the Lyapunov function can be a viable option for transient stability assessment(TSA)of such a VSC-interfaced microgrid.However,the determination of the Lyapunov function with the classical method is very challenging for a microgrid with converter controller dynamics.To overcome such challenges,this paper develops a physics-informed,Lyapunov function-based TSA framework for VSC-interfaced microgrids.The method uses the physics involved and the initial and boundary conditions of the system in learning the Lyapunov functions.This method is tested and validated under faults,droop-coefficient changes,generator outages,and load shedding on a small grid-connected microgrid and the CIGRE microgrid.

Hybrid Analytical and Data-Driven Model Based Instance-Transfer Method for Power System Online Transient Stability Assessment

Data-driven methods are widely recognized and generate conducive results for online transient stability assessment.However,the tedious and time-consuming process of sample collection is often overlooked.The functioning of power systems involves repetitive sample collection due to the constant variations occurring in the operation mode,thereby highlighting the importance of collection efficiency.As a means to achieve high sample collection efficiency following the operation mode change,we propose a novel instance-transfer method based on compression and matching strategy,which facilitates the direct acquisition of useful previous samples,used for creating the new sample base.Additionally,we present a hybrid model to ensure rationality in the process of sample similarity comparison and selection,where features of analytical modeling with special significance are introduced into data-driven methods.At the same time,a data-driven method can also be integrated in the hybrid model to achieve rapid error correction of analytical models,enabling fast and accurate post-disturbance transient stability assessment.As a paradigm,we consider a scheme for online critical clearing time estimation,where integrated extended equal area criterion and extreme learning machine are employed as analytical model part and data-driven error correction model part,respectively.Derived results validate the credible efficacy of the proposed method.

Safety assessment of waste rock dump built on existing tailings ponds

The construction of waste rock dumps on existing tailing ponds has been put into practice in China to save precious land resources. This work focuses on the safety assessment of the Daheishan molybdenum mine waste rock dump under construction on two adjoining tailings ponds. The consolidation of the tailings foundation and the filling quality of the waste rock are investigated by the transient electromagnetic method through detecting water-rich areas and loose packing areas, from which, the depth of phreatic line is also estimated. With such information and the material parameters, the numerical method based on shear strength reduction is applied to analyzing the overall stability of the waste rock dump and the tailings ponds over a number of typical cross sections under both current and designed conditions, where the complex geological profiles exposed by site investigation are considered. Through numerical experiments, the influence of soft lenses in the tailings and possible loose packing areas in the waste rock is examined. Although large displacements may develop due to the soft tailings foundation, the results show that the waste rock dump satisfies the safety requirements under both present and designed conditions.

Real-time transient stability assessment in power system based on improved SVM

Due to the strict requirements of extremely high accuracy and fast computational speed, real-time transient stability assessment(TSA) has always been a tough problem in power system analysis.Fortunately, the development of artificial intelligence and big data technologies provide the new prospective methods to this issue, and there have been some successful trials on using intelligent method, such as support vector machine(SVM) method.However, the traditional SVM method cannot avoid false classification, and the interpretability of the results needs to be strengthened and clear.This paper proposes a new strategy to solve the shortcomings of traditional SVM,which can improve the interpretability of results, and avoid the problem of false alarms and missed alarms.In this strategy, two improved SVMs, which are called aggressive support vector machine(ASVM) and conservative support vector machine(CSVM), are proposed to improve the accuracy of the classification.And two improved SVMs can ensure the stability or instability of the power system in most cases.For the small amount of cases with undetermined stability, a new concept of grey region(GR) is built to measure the uncertainty of the results, and GR can assessment the instable probability of the power system.Cases studies on IEEE 39-bus system and realistic provincial power grid illustrate the effectiveness and practicability of the proposed strategy.

Improved Deep Belief Network and Model Interpretation Method for Power System Transient Stability Assessment

The real-time transient stability assessment(TSA)and emergency control are effective measures to suppress accident expansion,prevent system instability,and avoid large-scale power outages in the event of power system failure.However,real-time assessment is extremely demanding on computing speed,and the traditional method is not competent.In this paper,an improved deep belief network(DBN)is proposed for the fast assessment of transient stability,which considers the structural characteristics of power system in the construction of loss function.Deep learning has been effective in many fields,but usually is considered as a black-box model.From the perspective of machine learning interpretation,this paper proposes a local linear interpreter(LLI)model,and tries to give a reasonable interpretation of the relationship between the system features and the assessment result,and illustrates the conversion process from the input feature space to the high-dimension representation space.The proposed method is tested on an IEEE new England test system and demonstrated on a regional power system in China.The result demonstrates that the proposed method has rapidity,high accuracy and good interpretability in transient stability assessment.

Data-driven Transient Stability Assessment Based on Kernel Regression and Distance Metric Learning

Transient stability assessment(TSA) is of great importance in power systems. For a given contingency, one of the most widely-used transient stability indices is the critical clearing time(CCT), which is a function of the pre-fault power flow.TSA can be regarded as the fitting of this function with the prefault power flow as the input and the CCT as the output. In this paper, a data-driven TSA model is proposed to estimate the CCT. The model is based on Mahalanobis-kernel regression,which employs the Mahalanobis distance in the kernel regression method to formulate a better regressor. A distance metric learning approach is developed to determine the problem-specific distance for TSA, which describes the dissimilarity between two power flow scenarios. The proposed model is more accurate compared to other data-driven methods, and its accuracy can be further improved by supplementing more training samples.Moreover, the model provides the probability density function of the CCT, and different estimations of CCT at different conservativeness levels. Test results verify the validity and the merits of the method.

Data-driven Transient Stability Assessment Model Considering Network Topology Changes via Mahalanobis Kernel Regression and Ensemble Learning

Transient stability assessment(TSA)is of great importance in power system operation and control.One of the usual tasks in TSA is to estimate the critical clearing time(CCT)of a given fault under the given network topology and pre-fault power flow.Data-driven methods try to obtain models describing the mapping between these factors and the CCT from a large number of samples.However,the influence of network topology on CCT is hard to be analyzed and is often ignored,which makes the models inaccurate and unpractical.In this paper,a novel data-driven TSA model combining Mahalanobis kernel regression and ensemble learning is proposed to deal with the problem.The model is a weighted sum of several sub-models.Each sub-model only uses the data of one topology to construct a kernel regressor.The weights are determined by both the topological similarity and numerical similarity between the samples.The similarities are decided by the parameters in Mahalanobis distance,and the parameters are to be trained.To reduce the model complexity,sub-models within the same topology category share the same parameters.When estimating CCT,the model uses not only the sub-model which the sample topology belongs to,but also other sub-models.Thus,it avoids the problem that there may be too few data under some topologies.It also efficiently utilizes information of data under all the topologies.Moreover,its decision-making process is clear and understandable,and an effective training algorithm is also designed.Test results on both the IEEE 10-machine 39-bus and a real system verify the effectiveness of the proposed model.

Multi-indicator Inference Scheme for Fuzzy Assessment of Power System Transient Stability

A multi-indicator inference scheme is proposed in this paper to achieve an intuitive assessment of post-fault transient stability of power systems.The proposed scheme uses the fuzzy inference technique to classify the stability level as“safe,”“low-risk,”“high-risk,”and“danger.”A multi-criteria quality assessment method is first introduced.Several transient indicators are then proposed as assessment criteria.To select the effective indicators for assessment,correlation mining using univariate regression analysis is performed between each indicator and a critical clearance time(CCT)-based stability index.The fuzzy sets of indicators for different stability levels are then determined according to their correlations with the stability index.The weighting factors of indicators are also allocated according to their regression error in correlation mining.The proposed inference scheme is further demonstrated and its effectiveness is validated in case studies on IEEE 68-bus system and a 756-bus transmission system in China.

Data-driven Transient Stability Assessment Using Sparse PMU Sampling and Online Self-check Function

Artificial intelligence technologies provide a newapproach for the real-time transient stability assessment (TSA)of large-scale power systems. In this paper, we propose a datadriven transient stability assessment model (DTSA) that combinesdifferent AI algorithms. A pre-AI based on the time-delay neuralnetwork is designed to locate the dominant buses for installingthe phase measurement units (PMUs) and reducing the datadimension. A post-AI is designed based on the bidirectionallong-short-term memory network to generate an accurate TSAwith sparse PUM sampling. An online self-check function of theonline TSA’s validity when the power system changes is furtheradded by comparing the results of the pre-AI and the post-AI.The IEEE 39-bus system and the 300-bus AC/DC hybrid systemestablished by referring to China’s existing power system areadopted to verify the proposed method. Results indicate that theproposed method can effectively reduce the computation costswith ensured TSA accuracy as well as provide feedback forits applicability. The DTSA provides new insights for properlyintegrating varied AI algorithms to solve practical problems inmodern power systems.

Post-disturbance transient stability assessment of power systems towards optimal accuracy-speed tradeoff

The recent development of phasor measurement technique opens the way for real-time post-disturbance transient stability assessment(TSA).Following a disturbance,since the transient instability can occur very fast,there is an urgent need for fast TSA with sufficient accuracy.This paper first identifies the tradeoff relationship between the accuracy and speed in post-disturbance TSA,and then proposes an optimal self-adaptive TSA method to optimally balance such tradeoff.It uses ensemble learning and credible decision-making rule to progressively predict the post-disturbance transient stability status,and models a multi-objective optimization problem to search for the optimal balance between TSA accuracy and speed.With such optimally balanced TSA performance,the TSA decision can be made as fast as possible while maintaining an acceptable level of accuracy.The proposed method is tested on New England 10-machine 39-bus system,and the simulation results verify its high efficacy.

Power system transient stability assessment based on the multiple paralleled convolutional neural network and gated recurrent unit

In order to accurately evaluate power system stability in a timely manner after faults,and further improve the feature extraction ability of the model,this paper presents an improved transient stability assessment(TSA)method of CNN+GRU.This comprises a convolutional neural network(CNN)and gated recurrent unit(GRU).CNN has the feature extraction capability for a micro short-term time sequence,while GRU can extract characteristics contained in a macro long-term time sequence.The two are integrated to comprehensively extract the high-order features that are contained in a transient process.To overcome the difficulty of sample misclassification,a multiple parallel(MP)CNN+GRU,with multiple CNN+GRU connected in parallel,is created.Additionally,an improved focal loss(FL)func-tion which can implement self-adaptive adjustment according to the neural network training is introduced to guide model training.Finally,the proposed methods are verified on the IEEE 39 and 145-bus systems.The simulation results indicate that the proposed methods have better TSA performance than other existing methods.

Information Entropy Based Prioritization Strategy for Data-driven Transient Stability Batch Assessment

Transient stability batch assessment(TSBA)is es-sential for dynamic security check in both power system planning and day-ahead dispatch.It is also a necessary technique to generate sufficient training data for data-driven online transient stability assessment(TSA).However,most existing work suffers from various problems including high computational burden,low model adaptability,and low performance robustness.Therefore,it is still a significant challenge in modern power systems,with numerous scenarios(e.g.,operating conditions and"N-k"contin-gencies)to be assessed at the same time.The purpose of this work is to construct a data-driven method to early terminate time-domain simulation(TDS)and dynamically schedule TSBA task queue a prior,in order to reduce computational burden without compromising accuracy.To achieve this goal,a time-adaptive cas-caded convolutional neural networks(CNNs)model is developed to predict stability and early terminate TDS.Additionally,an information entropy based prioritization strategy is designed to distinguish informative samples,dynamically schedule TSBA task queue and timely update model,thus further reducing simulation time.Case study in IEEE 39-bus system validates the effectiveness of the proposed method.

基于Relief-FT的时间自适应TSA方法研究

传统暂态稳定评估方法难以同时保证评估过程的实时性和准确性,在提升准确率的情况下降低实时性能,不利于系统快速排除故障。提出了一种时间自适应暂态稳定评估方案,首先对特征提取算法Relief-F进行了改进,克服对时间序列等形式特征较为敏感的缺陷,考虑时间变化因素的影响提出新的特征提取方案Relief-FT,计算得到多变量时间特征的重要性;然后将Relief-FT方案与长短期记忆网络结合,并对网络模型进行训练;最后通过新英格兰39节点系统进行实验分析,仿真结果表明,上述方法在保证准确率的情况下能够有效降低模型复杂度和加快训练速度。

Two-stage Transient-stability-constrained Optimal Power Flow for Preventive Control of Rotor Angle Stability and Voltage Sags

In practice,an equilibrium point of the power system is considered transiently secure if it can withstand a specified contingency by maintaining transient evolution of rotor angles and voltage magnitudes within set bounds.A novel sequential approach is proposed to obtain transiently stable equilibrium points through the preventive control of transient stability and transient voltage sag(TVS)problems caused by a severe disturbance.The proposed approach conducts a sequence of non-heuristic optimal active power re-dispatch of the generators to steer the system toward a transiently secure operating point by sequentially solving the transient-stability-constrained optimal power flow(TSC-OPF)problems.In the proposed approach,there are two sequential projection stages,with the first stage ensuring the rotor angle stability and the second stage removing TVS in voltage magnitudes.In both projection stages,the projection operation corresponds to the TSC-OPF,with its formulation directly derived by adding only two steady-state variable-based transient constraints to the conventional OPF problem.The effectiveness of this approach is numerically demonstrated in terms of its accuracy and computational performance by using the Western System Coordinated Council(WSCC)3-machine 9-bus system and an equivalent model of the Mexican 46-machine 190-bus system.

Transient Stability Analysis of Grid-connected Converters in Wind Turbine Systems Based on Linear Lyapunov Function and Reverse-time Trajectory

As the proportion of converter-interfaced renewable energy resources in the power system is increasing,the strength of the power grid at the connection point of wind turbine generators(WTGs)is gradually weakening.Existing research has shown that when connected with the weak grid,the stability of the traditional grid-following controlled converters will deteriorate,and they are prone to unstable phenomena such as oscillation.Due to the limitations of linear analysis that cannot sufficiently capture the stability phenomena,transient stability must be investigated.So far,standalone time-domain simulations or analytical Lyapunov stability criteria have been used to investigate transient stability.However,the time-domain simulations have proven to be computationally too heavy,while analytical methods are difficult to formulate for larger systems,require many modelling assumptions,and are often conservative in estimating the stability boundary.This paper proposes and demonstrates an innovative approach to estimating the transient stability boundary via combining the linear Lyapunov function and the reverse-time trajectory technique.The proposed methodology eliminates the need of time-consuming simulations and the conservative nature of Lyapunov functions.This study brings out the clear distinction between the stability boundaries with different post-fault active current ramp rate controls.At the same time,it provides a new perspective on critical clearing time for wind turbine systems.The stability boundary is verified using time-domain simulation studies.

基于改进非对称三重训练的风电并网系统暂态稳定自适应评估

为进一步提高暂态稳定评估模型的特征提取能力和模型在系统运行工况改变后的适应性,构建具有注意力机制的双路径卷积网络,以判别暂态稳定情况,得到更好的暂态稳定评估效果。当拓扑结构和运行方式变化过大时,通过时域仿真生成大量无标签样本,以双路径卷积网络作为三重训练基分类器;通过融合非对称三重训练和主动查询策略自适应调整基分类器参数,逐步提取源域与目标域之间的公共特征,减少标签样本的需求。最后,算例分析验证了所提方法的有效性。

电力系统暂态电压稳定评估的混合智能特征双重筛选方法

含高比例新能源与直流接入的电力系统暂态电压稳定特征呈高维冗余性,影响基于数据驱动评估模型的效率和性能。为此,在构建一组适应含高比例新能源和直流接入场景的完备特征集合基础上,提出一种基于改进Relief算法和改进群智能优化算法双重筛选的混合智能特征选择方法,以降低原始特征维度,提高模型稳定评估的效率和准确率。首先,通过时序分层处理对原始Relief算法进行时序改进,并利用该改进算法进行特征的有效性度量,以消除分类低效特征,得到降维后的初筛特征子集;随后,融合特征有效性度量值对群智能优化算法进行搜索性能增强。再以此增强算法为寻优策略,并以时序分类模型卷积门控循环单元(convolution gated recurrent unit,ConvGRU)为分类器,构成基于群智能优化算法的封装式特征选择方案,进一步实现特征子集寻优。最后,通过算例对比分析,该方法下高维特征维度能压缩80%以上,且所选特征子集能有效提高评估模型的准确率,验证该方法对于高维时序特征筛选处理的有效性及必要性。

基于XGboost-DF的电力系统暂态稳定评估方法

针对现代互联电网扰动后失稳模式不再单一,多摆失稳频频发生的现象,文中提出一种基于极限梯度提升-深度森林的暂态稳定评估方法。利用母线电压轨迹簇构建人工特征集,通过极限梯度提升方法对特征集进行监督特征编码;利用深度森林对监督编码后的稀疏矩阵进行三分类,进而建立起大规模数据集和失稳模式的映射关系;在IEEE 39节点和IEEE 140节点系统上进行仿真分析,所提方法具有很高的准确率和抗噪性能,能有效降低多摆失稳的误判率,并且在同步相量测量单元缺失情况下仍有较强的鲁棒性。

考虑新能源随机性的新型电力系统图深度学习稳定指标概率分布评估模型

新型电力系统的新能源大规模随机性出力特点导致目前在线暂态稳定扫描结论面临失效风险,这一问题从元件建模精度、系统拓扑适应性和计算速度三个方面同时对现有方法提出挑战。因此提出了基于重概率化改进的置信带方法和图深度学习的稳定指标概率分布及置信带评估模型,通过图深度学习模型快速评估少量采样点,再由重概率化置信带方法扩增结果,利用标签共现知识引导时域仿真对准确率进行增强,最后使用置信带方法给出随机出力下的稳定概率分布和区间形式的评估结论。该方法优点在于利用了图深度学习的拓扑适应能力和快速计算特点,且不受元件建模精度限制,置信带计算结论具有可靠的理论背景,能够评估稳定概率分布。在IEEE-39和IEEE-300节点系统上的评估精度验证表明,所提方法能够高精度地预测指定方式暂态稳定指标,并给出可靠的概率评估结论。