Solving data acquisition and processing problems in power systems using a pattern analysis approach

Abstract

The paper proposes a new methodology for the combined solution of the topological identification observability analysis and bad data processing problems in power systems. The solution is based on a pattern analysis approach. An efficient framework is suggested for solving data acquisition and processing problems, as well as joining pattern analysis and analytical procedures. Two Werent techniques of pattern analysis are combined to produce a classifier and an estimator with unique characteristics to deal with noisy environments. Unobservable network areas, multiple interacting bad data and bad critical measurements can be efficiently treated. The patterns required for the training process can be acquired from the SCADA system and/or from load-flow simulations. Test results are presented for the IEEE 24-busbar reliability test system. IntroductionNowadays, in the operation of power systems, the performance of application programs used for either security or economical purposes is highly dependent on the quality of data being used. The role of power system state estimation (PSSE) is to supply a reliable real-time data base.The general problem of PSSE can be divided into four parts. The first is concerned with the network topology identification. The second is related to the observability analysis; the set of measurements that allows the estimation of all busbar complex voltages is determined in this part. Bad data detection and identification are performed in the third part. In the fourth part, the system state is estimated from the network structure and parameters, and the prefiltered set of measurements. The topology of the system can be determined if the status of the switches is available error free. However, the SCADA (supervisory control and data acquisition) system does not monitor all switches. Even the status of the monitored ones can not be trusted because of noisecorrupted telemetered information. Solutions for the problem of converting busbar-section-circuit breaker topology into busbar-branch topology have been proPaper 8045C (C4, m), lirst posed using logical procedures. The first real-time methods [l, 23 were designed to be implemented in a central computer. Decentralised processing schemes have been proposed [3, 41 to speedup the connectivity determination. The inclusion of the linked Tables of adjacent network elements into the database is avoided in Reference 5. This is possible by considering the network connectivity determination problem as a special case of an abstract graph connectivity problem. More recently, a tracking topology processor [6] has been suggested to avoid reordering and/or refactorising the whole state estimation matrix from the previous solution. Besides these logical procedures, indirect methods for the detection of configuration changes have been developed to correct erroneous information on the switching equipment status. Based on the mathematical model for a known

Identification of topology changes in power grids using phasor measurements

“…Each bus provides two of the four power and voltage parameters P i , Q i , |V i |, θ i and the remaining two parameters are found by solving the nonlinear power flow equations (1). Unfortunately, the nonlinear equations can be difficult to solve.…”

“…Transmission grid operators have a long history of using state estimation for supervisory control and planning but in the distribution grid real-time measurements are sparsely available and much must be inferred. State estimation is a major research topic in the power engineering community [20,28,22,21,4,27,1,23]. The problem of estimating the state of the grid is often divided into two interrelated phases.…”

Distribution System State Estimation Using an Artificial Neural Network Approach for Pseudo Measurement Modeling

“…A probabilistic neural network approach for assigning load profiles to consumers is proposed in [11]. In [12] ANNs are used to solve data acquisition and processing problems, with training data obtained either from load flow or the supervisory control and data acquisition (SCADA) system. A pseudo measurement modeling approach based on the Gaussian mixture model (GMM) is presented in [13].…”

A pattern recognition approach for data estimation and debugging in power systems

“…A new approach for the combined solution of the observability analysis and bad-data processing problems is presented in [26] and [27]. Using a pattern analysis approach, it provides a reliable input data base for the state estimator in a very fast way (pre-estimation data validation).…”

“…This paper compares the event-covering method developed in [26] and [27] with an artificial neural network model. It also presents the event-covering performance for realistic test cases.…”