While recent years have witnessed the high potential of various machine learning based schemes in wind turbine fault prognosis (WTFP), none of them effectively tackle the challenging yet practical issue of severe operational data loss during WTFP. To address this issue, this paper develops an intelligent time series (TS) data analytics approach for missing data-tolerant WTFP. Instead of struggling to impute missing data, a powerful TS subsequence nearest neighbor (SNN) profiling technique is introduced to directly characterize TS data with missing values and transform them into discerning SNN profiles. Taking such profiles as inputs, the long short-term memory based recurrent network is further leveraged to perform in-depth faulty feature learning from temporal evolution trends. This eventually results in a discriminative multi-fault classification model, which can implement reliable online WTFP in severe missing data contexts. Experimental test results with field data collected from an actual wind turbine demonstrate the efficacy of the proposed approach.

With the increasing penetration of distributed photovoltaic generation and energy storage systems in the demand side of the power system, new demand side model structures are necessary in order to better describe the dynamic performance of the power system. In this paper, a composite demand side model structure with load, distributed photovoltaic generation, and energy storage system together with a model parameter identification method are proposed to improve the traditional load model identification. The structure of the demand side model is proposed first and is further simplified so as to be identified at a high voltage level bus. The model parameter identifiability analysis is conducted based on the sensitivity method. The ambient signal data and disturbance data based model parameter identification method is proposed for the new demand side model structure using the differential evolution optimization method. The case study results for the WSCC 9 bus system show the effectiveness of the proposed model structure. Then, the case study results in a simplified 500-kV network of the Guangdong Power Grid show the effectiveness of the parameter identification method.

This paper investigates the impact of network structures on small-disturbance voltage stability of power systems in the long-term. A differential-algebraic model is established in terms of network topology characteristics and the aggregate nonlinear recovery load model. We define the weighted load connectivity (WLC) as an index to indicate the stability of network structures. Then robust stability assessment is used to eliminate the impact of uncertain recovery time of loads. The Monte Carlo method is used to study the relationship between WLC and small-disturbance voltage stability (SDVS) for a broad range of network topologies and parameters. We discover from the simulation results that the network with a larger value of WLC has a stronger ability to maintain SDVS. To further study the impact of renewable energy sources (RESs), we improve the system model by including an inverter-based distributed generator in parallel with each aggregated nonlinear dynamic load. The simulation results of the Hong Kong power grid show that the WLC can still indicate the stability of network structures effectively, and the systems with more stable network structures are less impacted by the penetration of RESs. The proposed WLC appears useful in providing guidance for power network planning and operation.

The approach of ambient signals-based load modeling (ASLM) was recently proposed to better track the time-varying changes of load models. To improve computation efficiency and model structure complexity, a hierarchical framework for ASLM is proposed in this paper. Through this framework, the hidden quasi-convexity of the load modeling problem is explored for the first time. This allows more complicated static load model structures and gradient or Hessian-based optimization algorithms to be used. In the upper level, the identification of dynamic load parameters is regarded as an optimization problem. In the lower level, the optimal static load parameters are obtained through linear regression for a given group of dynamic load parameters. Afterward, the regression residuals are regarded as the objective function (OF) of the upper level optimization problem. The proposed method is validated by the case study results on the Guangdong Power Grid. The results show that the OF is mostly quasi-convex after the transformation of the induction motor model, which provides the basis for the application of gradient or Hessian-based optimization algorithms. The case study results also validate that the proposed approach has better computation efficiency and model structure complexity compared with the previous ASLM approaches.

The results of ambient signals‐based load modelling are more sensitive to phasor measurement unit (PMU) measurement errors due to the relatively smaller disturbance magnitudes of ambient signals. To analyse the impact of PMU measurement errors, the stochastic distributions and frequency spectra of PMU measurement errors are required. In this study, the measurement errors of Sifang CDS361 PMU are experimentally tested, and then their impacts on load modelling results are analysed. The following contributions have been made in this study. Firstly, the stochastic distributions and frequency spectra of PMU measurement errors are experimentally tested, from which it can be concluded that systematic errors occupy a larger proportion compared with accidental errors. Secondly, the impacts of PMU measurement errors on load modelling results are analysed. It can be concluded that accidental errors have more severe impacts on the accuracy of both dynamic and static load models. Besides, systematically errors also impact the accuracy of static load models. Thirdly, the function of low‐pass filters in recovering measurement signals from PMU measurement errors and enhancing the accuracy of load modelling results are validated. It can be further concluded that load model parameters can be accurately identified with current PMU measurement accuracy.