Robert Craven

Multiclass support vector machines for power system disturbances classification based on wide-area frequency measurements

The intelligent, robust and fast multi-class classification of power system disturbances is very important to improve control algorithms for ensuring power system security and reliability, an essential function for smart grid infrastructure. Moreover, in a future power system mostly consisting of distributed generators and renewable energy resources on which the disturbance has more impact, the analysis of disturbances by classifying and categorizing real-time frequency data is rather critical. Fortunately, wide area frequency data from a nation-wide frequency monitoring network (FNET) provides a means by which disturbances can be detected. However, so far none of strategies reported to date has good performance at classifying the disturbances although many of them are used currently in on-line analysis. The complex and irregular pattern characteristics of each kind of disturbance are the main reason. Artificial intelligence methods could be one of the solutions, but the large number of input values and an insufficient number of training examples has slowed the reduction of artificial intelligence methods to practice. Therefore, a mathematical model of common disturbances is proposed to generate a training database for artificial intelligence method and feature extraction by computing the wavelet coefficients, parameterizing the results and computer generating the data. This paper uses a multi-class support vector machine model to be trained on the extracted features to discern the otherwise hard-to-classify disturbances pattern and upon testing, yields good performance.

An island detection demonstration on a laboratory sized power Grid (LabGrid)

A trend away from large conventional electrical power generation stations such as hydro, coal-fired, and nuclear toward smaller Distributed Generation (DG) has both benefits and complications. While DG can capitalize on renewable energy and can reduce transmission losses by being closer to local loads, the intermittent availability of some power sources coupled with the current lack of a “smart grid” creates conditions when small sections of the grid need to be disconnected, or islanded, from the main grid due to disturbances or outages. Published techniques for detecting the conditions for islanding exist and are briefly discussed in this paper. This paper introduces a Laboratory sized power Grid (LabGrid) as a physical model for testing islanding techniques, together with a first demonstration of one of these techniques.

Multi-Agent System algorithm for preventing cascading failures in smart grid systems

In this work, a technique based on an adaptive Multi-Agent System algorithm is implemented to solve the complex problem of cascading failure events which lead to total blackout. This method proposes a solution to a variant of cascading failure events and is unique as previous literature focuses on identifying the possibility of occurrence of the cascading failures and then mitigates the failures. The proposed solution which utilizes pre-stated mathematical combinations that aim to redispatch the power from the generators is dynamically and experimentally applied in real-time, therefore it considers all the active factors and constraints involved as it halts the occurrence of cascading failures after an N-1 contingency. The distributed and intelligent algorithm is modeled to suit power system applications and then implemented on an experimental set up of the generation and transmission side of the IEEE 30-bus system utilizing a reconfigurable Smart Grid Laboratory hardware developed for testing distributed algorithms requiring two way communication capabilities.

Modeling of the Performance of a Coal-Fired Power Plant in Real-Time

Efficient operation of a coal-fired power plant was not a serious consideration for the electric power industry until recently. The main reason for this was that until recently the power industry was regulated and as a consequence the customer base for a given generating station was almost predetermined. The fuel costs were passed on to the customer and there was no incentive for the power industry to improve operating efficiency which would have resulted in reduced fuel consumption. Deregulation of the electric power industry which is quickly spreading all over the world has brought about a big change in the thinking of the power industry. In a deregulated market the customer in principle can choose the power supplier. Obviously the customer would choose a supplier with the lowest cost of electricity. Thus there is an incentive for every power producer to generate power at the lowest cost which means improved efficiency. In most coal-fired units fuel cost accounts for nearly 60-70% of the total cost of generation which implies that there is an immediate need to reduce fuel cost by improving efficiency. As mentioned in the foregoing paragraph fuel costs account for nearly 60-70% of the total cost of generation. Therefore, if the fuel flow rate for a given unit were to be monitored accurately in real-time that would enable the power producer to declare a more realistic cost of electric power. It should be noted that before deregulation was contemplated, utilities used to measure fuel consumption on a semi-annual or annual basis. While some units are equipped with belt scales and gravimetric feeders for coal flow rate measurement, these devices need constant maintenance and calibration and are in general considered unreliable. Several developments have taken place in the last two decades due to which it is now possible to model the performance of a coal-fired unit in real-time. The key parameters that can be calculated by the model are: boiler efficiency, steam cycle heatrate, unit heatrate, and coal flow rate. In what follows, a brief review of the relevant literature will be given. This will be followed by a description of the model for performance monitoring. Finally some field results will be presented.

Experimental Transient Stability Analysis of MicroGrid systems: Lessons learned

A Transient Stability Analysis of a low inertia microgrid system is presented in this paper. The analysis is performed on experimental data collected for different case scenarios of grid disturbances. The test bed consists of a reconfigurable 100kW MicroGrid system with more than 10 rotating generators, PV, and Wind penetration capabilities. Both islanded and non-islanded situations are considered for a solid three phase fault analysis and comparison. The paper also presents the results of fault scenario when the rotating machines are pushed to an unstable region. The experimental results and system behavior observed show interesting phenomena. The MicroGrid showed higher resiliency for faults than expected even if the relative angles are more than 160° and even if the fault is sustained for a long time.

Development of a real time wind turbine emulator based on RTDS using advanced perturbation methods

This paper presents the development of a fixed speed wind turbine emulator using the Real Time Digital Simulator (RTDS). The wind turbine emulator consists of a wind turbine modeled in an RSCAD software environment. The software model for the wind turbine generates appropriate torque signals based on wind speed profiles using auto-regressive moving average and perturbation methods. These torque signals were fed to a servo motor emulating the wind turbine and were connected to an induction machine that emulated the generator. The response of the emulator shows that it closely matches the performance of the software model under various wind speed profiles. Thus, it can be used as an effective research and experimental tool in several wind energy based renewable energy projects.

Real-time measurement of frequency using affordable rotary encoder and LabVIEW

The accuracy of measurements from a rotary encoder is of high importance and the level of accuracy desired will depend on the nature of the application in which the encoder is used. Due to the various errors in the rotary encoder measurement during real-time usage, this work has designed a LabVIEW program that exempts erroneous readings that are not within the chosen range, forms an array of readings within the acceptable range and finds the mean of the array once the size of the array is equal to the chosen size of the array stipulated in the program. This mean value is given as the actual measurement. This gives a more accurate real-time measurement than is the case without the program. A motor-generator set frequency measurements is used to validate the effectiveness of this program and frequency readings on the inverter driving the motor are compared with that displayed on the system through the program. Some analyses are carried out in order to ascertain the dependency of the accuracy of the measurements acquired on the array size utilized in the program.

Parametric Studies of Power Plant Performance Monitoring

Real-time performance monitoring of coal-fired power plants is becoming very important due to the impending deregulation of the electric power industry. Performance testing is made to be real-time by changing the traditional output loss method to include an estimation of coal composition based on the Continuous Emission Monitoring System (CEMS) data. This paper illustrates the robustness of the calculations by introducing a variance into each of the calculation inputs to access its effect on the final outputs of heatrate, boiler efficiency and coal flow. Though the original study was over five power plants this paper presents results for the two most diverse coals.Copyright

A novel evaporative cooling tower constructed from an inflatable fabric duct

This paper describes a prototype of a novel vertical cooling tower constructed from an inflatable fabric duct that incorporated a water spray nozzle mounted at the top. The device was constructed using two concentric impermeable fabric tubes sealed at either end. It was inflated using a small pressurization fan located at the tower base. A numerical performance model was developed to estimate the downstream conditions of the evaporative cooling tower, and the length necessary for complete evaporation. To test the predictive capability the model of the evaporative cooling tower, experiments were performed under different ambient conditions. The prototype tower did exhibit downdraft cooling. However, the ambient conditions under which the validation measurements were performed were not conducive to achieving a moist air flow that could be measured accurately, and further investigation is recommended.

Adaptive Immune System reinforcement Learning-Based algorithm for real-time Cascading Failures prevention

Abstract Artificial intelligent algorithms have found a wide-range of applications in power systems, especially in solving long-existing problems immune to non-intelligent algorithms. Cascading Failures (CF), one of such problems, require load shedding as a current industrial solution. Load shedding results in losses to all power system stakeholders. This work proposes the use of an Artificial Immune System (AIS) algorithm to intelligently adjust the power output of the generators in the power system relative to one another in real time to prevent CF. AIS gives the artificial intelligent algorithm reinforcement learning capability by enabling it to pick the appropriate combination(s) for a particular system state; hence, the algorithm is called Immune System Reinforcement Learning-Based (ISRL-Based) algorithm. The algorithm was trained offline using both static and dynamic power equations and the effectiveness of both approaches was evaluated through statistical deviation. Analyses showed that using dynamic equations resulted in a more accurate solution than the static equations. CF was dynamically simulated on the IEEE 118-Bus system after an N-2 contingency, the results obtained agrees with the results from the analysis of the 2003 Northeast USA CF event. The effectiveness of the algorithm and online training were also experimentally validated after an N-1-1 contingency in a nine-bus system.