Towards an effective power outage planning ann approach

  • 206 views
Uploaded on

The International Institute for Science, Technology and Education (IISTE). Science, Technology and Medicine Journals Call for Academic Manuscripts …

The International Institute for Science, Technology and Education (IISTE). Science, Technology and Medicine Journals Call for Academic Manuscripts

More in: Business
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
206
On Slideshare
0
From Embeds
0
Number of Embeds
0

Actions

Shares
Downloads
3
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 49 Towards An Effective Power Outage Planning: Ann Approach Adewuyi Philip Adesola 1* Aborisade David Olugbemiga 2 1. Department of Mechatronics Engineering, Bells University of Technology, Nigeria 2. Department of Electronic and Electrical Engineering, Ladoke Akintola University of Technology, Nigeria * E-mail of the corresponding author: solaadewuyi@gmail.com Abstract Utility companies in developing nations are battling the problem of incessant power outages without giving prior warning to the consumers. This has caused questions to be raised as touching the integrity of the so called utility companies. Inability to give precise forecast is another problem that confronts these companies. Most forecast data are purely statistical which could be enhanced by the use of artificial intelligence tool like Artificial Neural Network. Outage planning could therefore be well analyzed making both the utility company and consumers schedule their activities without untold damage to machine, equipment, and minimization of economic losses. Keywords: utility, outage, predictability, ANN, trainoss 1. Introduction Developing nations are daily seeking effective ways of bringing stable electricity to their citizenry. Construction of more power generating stations, spring up all across developing nations without appreciable improvements in the power available at the consumer end. Several works by researchers have been aimed at forecasting power generation and load demands [1]. This trend has been ongoing for sometimes now [2]. Countries like South Africa, Ghana, and Libya have experienced remarkable improvement in terms of stability of power. Nigeria, the focus of this work, is struggling even in the attainment of MW of power. Recently, the power has dropped to a disappointing level. Therefore, there is a difficulty in forecasting power that would be generated or available in the national grid due to factors ranging from gas pipeline vandalisation, unstable operational policy, political undertone, nepotism and so on. In the last five years, several companies have closed shops in Nigeria due to severe power outages experienced which is not good for the survival of any business outfit. For instance, Dunlop PLC closed shop in 2011, Michelin in 2007 while Unilever PLC, PZ Cussons Nigeria, Afprint Nigeria Limited have considered relocating to other neighbouring countries that have stable power supply for their productions. From the above discussions, the place of proper power outage planning cannot be overemphasized. Proper power outage planning would be helpful to industries, parastatals, small businesses and individuals to plan their activities ahead without much emergency situations. Artificial neural network has played a significant role in the forecast and predictability of load demand in power sector [3, 4, 5]. The backpropagation training algorithm is adopted with “trainoss” being the drive vehicle. This algorithm is settled for since its implementation is made possible by inherent parallelism in the network architecture due to repeated use of simple neuron processing element [6]. ANN’s characteristics of learning information by example, ability to generalize to new inputs, robustness to noisy data that occurs in real world applications, and fault tolerance make it an important tool to use for this work. The rest of this paper examines types of outages, outage planning, model of artificial neuron, methodology, results and discussions, and conclusion. 2. Types of Outages i. Forced outages ii. Planned outages iii. Urgent outages iv. Emergency outages 2.1 Outage Planning Utility company in Nigeria requires a comprehensive outage forecast details from all power generating stations across the country. Downtimes, strictly arisen from equipment breakdown, schedule maintenance periods as well as grid integrity and security are factors generally considered in forecasting periods of power outages likely to be experienced in a given year. The National Control Centre’s data of outages released in 2009 was used in the training of ANN, testing, and validation of the results generated. Outage planning is therefore a key component in the overall operation of the utility company. The transmission lines considered are 330kV line and 132kV line for different types of outages experienced in 2008 and 2009 [7].
  • 2. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 50 2.2 Model of Artificial Neuron The most common model of artificial neuron was proposed by McCulloch and Pitts in 1943. The model has each neuron input, xi, and corresponding weight, wi. An offset or better still, a bias with a constant input of 1 weighted by the value, wo. Output is given as y which is obtained by summing the weighted inputs to the neuron and passing the result through a non-linear activation function, f [8]. Axon Synapses Dendrites X1 Body Axon X2 . y Non-linearity Output Xn Inputs Weights 1 Bias Figure 1: McCulloch-Pitts neuron model 3. Methodology Backpropagation algorithm is used for this work. This algorithm is derived following the simple gradient principle with a start point containing a typical weight wpq which is updated as follows:pq (1) = the learning rate is the error function to be minimized. The implementation is realized with these steps: 1. Initialize the weights to small random values 2. Choose a pattern xk; k=1, 2, ..., n and apply it to the input layer. 3. Propagate the signal forwards through the network using (2) (3) (4) (5) 4. Compute the deltas for the output layer (6) 5. Compute the deltas for the hidden layers by propagating the errors backwards (7) 6. Use (8) (9) to update all connections. 7. Go back to step 2 and repeat for the next pattern. W 1 Σ W 2 W n f wo
  • 3. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 51 4. Results and Discussions 4.1 Results Figure 2: Performance training curve 55 60 65 70 75 80 85 90 95 55 60 65 70 75 80 85 90 95 Targets T OutputsA,LinearFit:A=(0.86)T+(9.6) Outputs vs. Targets, R=0.92572 Data Points Best Linear Fit A = T Figure 3: Regression curve of the training parameters 0 2 4 6 8 10 12 20 25 30 35 40 45 50 55 60 65 Months Frequency 330kV Forced Outage Profile for 2009 Actual ANN Target Figure 4: 330kV forced outage profile for 2009
  • 4. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 52 0 2 4 6 8 10 12 15 20 25 30 35 40 45 50 55 Months Frequency 330kV Forced Outage Profile for 2008 Actual ANN Target Figure 5: 330kV forced outage profile for 2008 0 2 4 6 8 10 12 0 5 10 15 20 25 Months Frequency 330kV planned Outage Profile for 2008 Actual ANN Target Figure 6: 330kV planned outage profile for 2008 0 2 4 6 8 10 12 0 2 4 6 8 10 12 14 Months Frequency 330kV planned Outage Profile for 2009 Actual ANN Target Figure 7: 330kV planned outage profile for 2009
  • 5. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 53 0 2 4 6 8 10 12 2 4 6 8 10 12 14 16 18 20 Months Frequency 330kV Urgent Outage Profile for 2009 Actual ANN Target Figure 8: 330kV urgent outage profile for 2009 0 2 4 6 8 10 12 0 5 10 15 20 25 30 Months Frequency 330kV Urgent Outage Profile for 2008 Actual ANN Target Figure 9: 330kV urgent outage profile for 2008 0 2 4 6 8 10 12 0 5 10 15 20 25 30 35 Months Frequency 330kV Emergency Outage Profile for 2009 Actual ANN Target Figure 10: 330kV emergency outage profile for 2009
  • 6. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 54 0 2 4 6 8 10 12 0 5 10 15 20 25 30 35 40 Months Frequency 330kV Emergency Outage Profile for 2008 Actual ANN Target Figure 11: 330kV emergency outage profile for 2008 0 2 4 6 8 10 12 40 50 60 70 80 90 100 110 Months Frequency 132kV Forced Outage Profile for 2009 Actual ANN Target Figure 12: 132kV forced outage profile for 2009 0 2 4 6 8 10 12 40 50 60 70 80 90 100 110 120 Months Frequency 132kV Forced Outage Profile for 2008 Actual ANN Target Figure 13: 132kV forced outage profile for 2008
  • 7. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 55 0 2 4 6 8 10 12 0 2 4 6 8 10 12 Months Frequency 132kV Planned Outage Profile for 2009 Actual ANN Target Figure 14: 132kV planned outage profile for 2009 0 2 4 6 8 10 12 0 2 4 6 8 10 12 14 16 Months Frequency 132kV Planned Outage Profile for 2008 Actual ANN Target Figure 15: 132kV planned outage profile for 2008 0 2 4 6 8 10 12 0 50 100 150 200 250 300 350 Months Frequency 132kV Urgent Outage Profile for 2009 Actual ANN Target Figure 16: 132kV urgent outage profile for 2009
  • 8. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 56 0 2 4 6 8 10 12 15 20 25 30 35 40 Months Frequency 132kV Urgent Outage Profile for 2008 Actual ANN Target Figure 17: 132kV urgent outage profile for 2008 0 2 4 6 8 10 12 0 50 100 150 200 250 300 350 400 450 500 Months Frequency 132kV Emergency Outage Profile for 2009 Actual ANN Target Figure 18: 132kV emergency outage profile for 2009 0 2 4 6 8 10 12 0 50 100 150 200 250 300 350 400 450 500 Months Frequency 132kV Emergency Outage Profile for 2008 Actual ANN Target Figure 19: 132kV emergency outage profile for 2008
  • 9. Journal of Energy Technologies and Policy www.iiste.org ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online) Vol.4, No.7, 2014 57 4.2 Discussion The results obtained for the outage periods shows that ANN could be used to predict the power outage periods in the country as attested to by the regression curve value of 0.9 obtained. This would enable adequate plans to be made and the consumers should be provided with such data in order to plan their activities. The training performance goal is met with one of the fastest training algorithms other than the ‘trainlm’, that is, ‘trainoss’ as shown on the matlab window below: TRAINOSS-srchbac-calcgrad, Epoch 0/300, MSE 1.76984/0.001, Gradient 5.33559/1e-006 TRAINOSS-srchbac-calcgrad, Epoch 21/300, MSE 0.000936266/0.001, Gradient 0.0469698/1e-006 TRAINOSS, Performance goal met. The “trainoss” stands for one step secant (OSS) method and is an attempt to bridge the gap between the conjugate gradient algorithms and the quasi-Newton (secant) algorithms. This algorithm does not store the complete Hessian matrix. It assumes that at each iteration the previous Hessian was the identity matrix. This has the additional advantage that the new search direction can be calculated without computing a matrix inverse. It therefore requires less storage space and computational time. 5. Conclusion Power outage forecasting helps to evaluate grid system outage times and serve in some way to represent or show the pattern of the present and future grid downtime. It also enhances adequate planning on both consumer side as well as the supply authority side. Furthermore, improvement as regards ways to lessen outage periods of whatever type could be fashioned since research activities would have shown the causes of such outages before they occur. The result obtained from this research work shows the efficiency of neural network and the accuracy of short term method to forecast outage periods. The Network was able to determine the nonlinear relationships that exist between the historical outages experienced that is presented to it during training phase. References [1] Adams S. O., Akano R. O., and Asemota O. J. (2011) “Forecasting Electricity Generation in Nigeria using Univariate Time Series Models”, European Journal of Scientific Research, Vol. 58, No. 1, pp. 30- 37. [2] Kayode O., and Nyanapfenen A. (2011) “Urban Residential Energy Demand Modeling in Developing Countries: A Nigerian Case Study”, The Pacific Journal of Science and Technology, Vol. 12, No. 2, pp.152-159. [3] Adepoju G. A., Ogunjuyigbe S. O. A., and Alawode K. O. (2007) “Application of Neural Network to Load Forecasting i Nigerian Electrical Power System”, Pacific Journal of Science and Technology, Vol. 8, No. 1, pp.68-72. [4] Adewuyi P. A., and Okelola M. O. (2012) “Reducing Peak Power Demand of Ladoke Akintola University of Technology, Nigeria-Using Artificial Neural Network”, International Journal of Innovative Research and Development, Vol. 1, Iss. 11, pp. 576-588. [5] Afolabi A. O., Olatunji B. O., and Ajayi A. O. (2008) “Electricity Load Forecasting Using Artificial Neural Networks”, Journal of Engineering and Applied Sciences, Vol. 3, Iss. 2, pp. 149-154. [6] Hagan M. T., Demuth H. B., and Beale M. H. (1996) “Neural Network Design”, MA: PWS Publishing, Boston. [7] Annual Technical Report on Generation and Transmission Grid Operations 2009 by Power Holding Company of Nigeria PLC, National Control Centre, Osogbo-Nigeria. [8] Nagrath I. J., and Gopal M. (2012) “Control Systems Engineering”, New Age International Publishers, New Delhi, India.
  • 10. The IISTE is a pioneer in the Open-Access hosting service and academic event management. The aim of the firm is Accelerating Global Knowledge Sharing. More information about the firm can be found on the homepage: http://www.iiste.org CALL FOR JOURNAL PAPERS There are more than 30 peer-reviewed academic journals hosted under the hosting platform. Prospective authors of journals can find the submission instruction on the following page: http://www.iiste.org/journals/ All the journals articles are available online to the readers all over the world without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself. Paper version of the journals is also available upon request of readers and authors. MORE RESOURCES Book publication information: http://www.iiste.org/book/ IISTE Knowledge Sharing Partners EBSCO, Index Copernicus, Ulrich's Periodicals Directory, JournalTOCS, PKP Open Archives Harvester, Bielefeld Academic Search Engine, Elektronische Zeitschriftenbibliothek EZB, Open J-Gate, OCLC WorldCat, Universe Digtial Library , NewJour, Google Scholar
  • 11. Business, Economics, Finance and Management Journals PAPER SUBMISSION EMAIL European Journal of Business and Management EJBM@iiste.org Research Journal of Finance and Accounting RJFA@iiste.org Journal of Economics and Sustainable Development JESD@iiste.org Information and Knowledge Management IKM@iiste.org Journal of Developing Country Studies DCS@iiste.org Industrial Engineering Letters IEL@iiste.org Physical Sciences, Mathematics and Chemistry Journals PAPER SUBMISSION EMAIL Journal of Natural Sciences Research JNSR@iiste.org Journal of Chemistry and Materials Research CMR@iiste.org Journal of Mathematical Theory and Modeling MTM@iiste.org Advances in Physics Theories and Applications APTA@iiste.org Chemical and Process Engineering Research CPER@iiste.org Engineering, Technology and Systems Journals PAPER SUBMISSION EMAIL Computer Engineering and Intelligent Systems CEIS@iiste.org Innovative Systems Design and Engineering ISDE@iiste.org Journal of Energy Technologies and Policy JETP@iiste.org Information and Knowledge Management IKM@iiste.org Journal of Control Theory and Informatics CTI@iiste.org Journal of Information Engineering and Applications JIEA@iiste.org Industrial Engineering Letters IEL@iiste.org Journal of Network and Complex Systems NCS@iiste.org Environment, Civil, Materials Sciences Journals PAPER SUBMISSION EMAIL Journal of Environment and Earth Science JEES@iiste.org Journal of Civil and Environmental Research CER@iiste.org Journal of Natural Sciences Research JNSR@iiste.org Life Science, Food and Medical Sciences PAPER SUBMISSION EMAIL Advances in Life Science and Technology ALST@iiste.org Journal of Natural Sciences Research JNSR@iiste.org Journal of Biology, Agriculture and Healthcare JBAH@iiste.org Journal of Food Science and Quality Management FSQM@iiste.org Journal of Chemistry and Materials Research CMR@iiste.org Education, and other Social Sciences PAPER SUBMISSION EMAIL Journal of Education and Practice JEP@iiste.org Journal of Law, Policy and Globalization JLPG@iiste.org Journal of New Media and Mass Communication NMMC@iiste.org Journal of Energy Technologies and Policy JETP@iiste.org Historical Research Letter HRL@iiste.org Public Policy and Administration Research PPAR@iiste.org International Affairs and Global Strategy IAGS@iiste.org Research on Humanities and Social Sciences RHSS@iiste.org Journal of Developing Country Studies DCS@iiste.org Journal of Arts and Design Studies ADS@iiste.org