Artificial intelligence in Power Stations
- 1. ARTIFICIAL INTELLIGENCE IN POWER STATIONS PRESENTED BY SOMARLAPATI CHAITANYA AVINASH [email protected]
- 2. DRAFT AI as a Revolution. Load Forecasting definition and process involved in it. Types of load forecasting. Techniques used to calculate load forecasting. Conclusion. Bibliography.
- 3. ARTIFICIAL INTELLIGENCE Artificial Intelligence is the science of exhibiting intelligence by machine currently achieved by humans.
- 4. John McCarthy is one of the "founding fathers" of artificial intelligence, together with Marvin Minsky, Allen Newell, and Herbert A. Simon.
- 5. TIMELINE OF AI AI is coined as a proposal for the first time by John McCarthy 1955 DARPA and NASA’s exploration and Research begun 2004 MNC’s takeover very important and crucial projects. 2019 The future prediction says that AI is going to occupy 70% jobs. 2030 First National Conference of the American Association for Artificial Intelligence (AAAI) held at stanford. 1980
- 6. 2000 ~ 2018 Observation 1 : The Artificial Intelligence gained it’s initial boost. 2022 ~ 2025 Observation 2 : The Machine is dominating the work of human. 2025 2022 2018 2000 82% 70% 64% 48% WORLD STATISTICS AI as a Revolution Machine Human
- 7. NEED FOR AI ARTIFICIAL INTELLIGENCE o With increased competitiveness in power generation industries, more resources are directed in optimizing plant operation, including fault detection and diagnosis. o One of the most powerful tools in faults detection and diagnosis is artificial intelligence (AI).
- 8. HYDRO POWER PLANT THERMAL POWER PLANT GAS POWER PLANT NUCLEAR POWER PLANT POWER STATIONS
- 9. LOAD FORECASTING Why load forecasting is calculated frequently by the Power Stations ? The Load Forecasting helps in planning the future in terms of the size, location and type of the future generating plants to meet all the needs of future generations. It helps in deciding and planning for maintenance of the power systems.
- 10. PROCESS IN L.F Collect Information Choose the Forecasting Model Verify Model Performance Identify the problem Perform a Preliminary Analysis Data analysis SIX STEPS PROCEDURE
- 11. LOAD FORECASTING MEDIUM TERM FORECASTING SHORT TERM LOAD FORECASTING LONG TERM FORECASTING
- 12. SHORT TERM LOAD FORECASTING ᴥ Short term load forecasting (STLF) refers to forecasts of electricity demand (or load), on an hourly basis, from one to several days ahead. In the daily operations of a power utility, the short term load forecasting is of vital importance. ᴥ It is required for unit commitment, energy transfer scheduling and load dispatch. ᴥ The short term load forecasting has played a greater role in utility operations with the emergence of load management strategies.
- 13. ∞ Medium-term Load forecasting (MTLF) becomes an essential tool for today power systems, mainly in those countries whose power systems operate in a deregulated environment. ∞ This kind of load forecast has many applications like maintenance scheduling, mid-term hydro thermal coordination, adequacy assessment, management of limited energy units, negotiation of forward contracts, and development of cost efficient fuel purchasing strategies. MEDIUM TERM LOAD FORECASTING
- 14. LONG TERM LOAD FORECASTING ᴥ Long-term load forecasting is an important component for power system energy management and reliable power system operation. ᴥ Long - term load forecasting span is within the period of one year to more than one year.
- 15. Types of Forecasting Techniques Artificial Neural Networks Fuzzy Logic Approach Fuzzy logic or Fuzzy systems are logical systems for standardization and formalisation of approximate reasoning. NEURO - FUZZY APPROACH Neuro - fuzzy refers to combinations of artificial neural networks and fuzzy logic. Regression Method Regression method is a conventional approach. 01 02 03 04 Biologically inspired systems which convert a set of inputs into a set of outputs by a network of neurons.
- 16. ARTIFICIAL NEURAL NETWORKS ¥The application of neural networks in power utilities has been growing in acceptance over the years. ¥The main reason behind this is because the capability of the artificial neural networks in capturing process information in a black box manner. ¥When load forecasting is dealt by using neural networks, we must select one of the number of the available architectures (such as Hopfield, back propagation, Boltzmann, etc). ¥The number of layers and elements, the connectivity between them , usage of unilateral or bilateral links and the number format to be used by inputs and outputs.
- 17. FUZZY LOGIC APPROACH Step 1: Compile a tentative list of input and output variables using statistical analysis, engineering judgments and/or load forecasting. There are three input variables which are used to forecast electric loads an output are 1. Temperature. 2. Humidity. 3. Wind speed. Step 2: Normalization of the input and output variables is done by analyzing the input and the output behavior the input space is mapped to the membership value. Step 3: Select the shape of the fuzzy membership for each variable; namely the triangular, trapezoidal, Gaussian and bell shape membership function. Step 4: For each input and output variable, tentatively define the number of fuzzy membership functions. Step 5: Fuzzy logic rule base is each pair of input and output data, and it’s called training data. Example: If ‘temperature’ is hot and ‘humidity’ is humid and ‘wind speed’ is above average then ‘load’ is above average.
- 19. NEURO – FUZZY APPROACH A price - sensitive (PS) load forecaster is developed. This forecaster consists of two stages, an artificial neural network based PIS load forecaster followed by a fuzzy logic (FL) system that transforms the PIS load forecasts of the first stage into PS forecasts. The first stage forecaster is a widely used forecaster in the industry known as ANN. For the FL system of the second stage, a genetic algorithm based approach is developed to automatically optimize the number of rules and the number and parameters of the fuzzy membership functions. Another FL system is developed to simulate PS load data from the PIS historical data of a utility. This new forecaster termed NFLF is tested on three PS databases, and it is shown that it produces superior results to the PIS ANNLF.
- 22. FUTURE LOAD FORECASTING OF INDIA ֍India planned a lot in load forecasting related to electricity domain by 2030. ֍Firstly by 2020, India wants to succeed by replacing the LED’s for lighting purpose in all industrial usage.
- 23. LOAD FORECASTING 1. Seasonal Use of certain utility increases or decreases. 2. Trend On certain events. 3. Random No specific cause. DEPENDS ON TIME - SERIES
- 24. The load forecasting has both commercial and technical implications and if not done properly, it may lead to bad planning and inefficient operation of the electrical power systems. What happens if LOAD FORECASTING is not planned properly ?
- 25. CONCLUSION Precise load forecasting is very essential for electric utilities in a spirited environment created by the electric industry. In this presentation we appraise some statistical view of artificial intelligence growth and techniques that are used for electric load forecasting. Neural network alone cannot work for forecasting better. If neural network is combined with fuzzy logic then it can handle the forecasting problems well. The evolution in load forecasting will be achieved in two ways. One is getting excellence in statistics and artificial intelligence and the other is to have good understanding of the load dynamics.
- 26. AI revolves around the globe. ֍https://ieeexplore.ieee.org/document/76685 ֍Tomonobu Senjyu, Hitoshi Takara, Katsumi Uezato, and Toshihisa Funabashi, “OneHour-Ahead load forecasting using neural network” , IEEE Transactions on power systems, Vol. 17, No. 1, February 2002 ֍Jagadish H. Pujar, “Fuzzy ideology based long term load forecasting”, World Academy of Science, Engineering and Technology 64 2010. BIBLIOGRAPHY
- 27. TOP CLASS AI PROJECTS BLUE BRAIN PROJECT AI INFERENCE CHIP IBM NEW AI LAB Taken by US Funding by Swiss Govt $1.3 billion Taken by INTEL Chip runs Electronic Device Taken by IBM Collaboration with MIT $240 million $13 Million
- 28. THANK YOU For Listening my lecture Patiently