Del2infinity provides renewable energy forecasting and scheduling services using proprietary machine learning algorithms. Their forecasting system leverages artificial intelligence techniques like artificial neural networks and statistical pattern recognition. It integrates historical and real-time wind/solar power and weather data to generate day-ahead forecasts in 15-minute time blocks with up to 16 revisions. Del2infinity aims to provide highly accurate forecasts even with limited available data by intelligently using available data inputs. They have tested their forecasting solutions on over 3 GW of renewable energy capacity in India and other countries.

1. Del2infinity Energy Consulting
(An accurate Wind Energy & Solar Energy Forecasting &
Scheduling Company)
How to do accurate RE Forecasting?

2. OUR BIG IDEA
2
• del2infinity Energy Consulting Private Limited (‘del2infinity’)
works in the domain of Energy analytics and performs IT
integrated and solution oriented approach for every energy
analytics problems. del2infinity serves AAAS (Analytics as A
Service) to its different clients in different energy related
problems specifically concentrated on renewable energy.

3. 3
about del2infinity
Methodology
del2infinity Renewable Energy Forecasting System is developed considering India-specific situations where access to possible and necessary wind/solar power prediction
data is a major issue. The system has been created considering the non-availability of some parameter data in many cases. However, the system is capable of accommodating
and optimizing on additional data when available.del2infinity believes in using the available data intelligently instead of asking for more data that may not be available.
del2infinity Renewable Energy Forecasting System leverages on artificial intelligence and uses proprietary algorithms based on statistical machine learning. The machine
learning using statistical pattern recognition and autonomous algorithms composing of Artificial Neural Network (ANN) with multiple layers of connections plays a crucial
role in the forecasting system. The major two parts of the del2infinity’s RenewableEnergy forecasting system are (1) Data Integration & (2) Analytics Engine.
Data Integration
del2infinity uses modern data architecture to store, arrange and integrate data from the Supervisory Control and Data Acquisition (SCADA) systems and/or
other customer specific servers. Additionally, significant effort has been put into automating the pre-processing of data so as to minimize time lag between
obtaining real-time raw data and converting it to meaningful data that can be input into the Analytics Engine to generate forecast. Historical data (1-3 years)
is used to train the forecasting system. Real-time data is used to do forecast revisions for improved accuracy in intra-day wind power forecasts.

4. 4
about del2infinity
Analytics Engine
• del2infinity uses Power to Power (P2P) algorithm as the basic functional module of the analytics engine. In this module the input is only the wind/solar-power time series data
(aggregated or plant level). The analytics engine uses different statistical pattern matching and ANN algorithms to fetch different statistical information related to the
wind/solar power data. The P2P algorithm generates different patterns in wind/solar power data as well as information on wind/solar power ramping events. The system
forecasts the next day’s wind/solar power output in 15-minute time blocks. The P2P algorithm does not use wind speed/radiation (like DNI) or weather-related data to ensure
system performance when there is lack of such data.
• The power to power (‘P2P’) forecast methodology developed by del2infinity is used to forecast the solar and wind power data points. The P2P forecast methodology works
better if there is no curtailment issue. But the curtailment issue of one day creates erroneous forecast in next day and affects next few days if only P2P Artificial Intelligence
(‘AI’) system is used. Hence though the proposed methodology uses P2P AI system, two parallel weather feedback loops are introduced to adapt the system. One feedback
loop is used for pattern matching using support vector machine (‘SVM’) and the other loop is used to measure the stochastic variation of power input. The analytics engine
also has a number of feedback loops that use wind/solar and weather data. Data used in these feedback loops can include numerical weather prediction variables (e.g.
forecasted wind speed, direction, and pressure / DNI, ambient temperature etc) and weather observations. These parallel feedback loops are activated only when these data are
available.
• del2infinity Renewable Energy Forecasting analytics tools and services are capable of doing 24 hours day-ahead power forecast with maximum 16 revisions, weekly ahead
forecast as well as monthly and yearly forecast. The proprietary solution of del2infinity’s forecasting and scheduling is tested over 3 GW of RE generations in India and other
parts of the globe.

5. Custom solutions
Del2infinity have been appointed as independent Forecasting Consultant for understanding the Generation Forecasting for
remaining life of WTG for Inox - Leap Green Transaction for 260 MW in of windfarms across Rajasthan, Maharashtra, MP
and TN.
Energy Efficiency & Reliability Study
• Energy efficiency, reliability & life-span study for Wind Turbine & PV modules
• Degradation calculation for Power generation
• Uncertainty calculation in power production
• Energy forecast in Pxx-levels
• Techno-economic analysis
5
OUR SERVICES
Energy Forecasting & Scheduling services
• Forecasting & Scheduling of Wind & Solar Energy generation in day-ahead, week ahead, monthly and yearly forecast;
• Load forecasting;
• Highly accurate and minimum deviation penalty;
• Techno-economic analysis

6. Wind Energy Study
• Wind Energy Forecast & Scheduling services
• Reliability, life-span and yearly energy yield projection
• Degradation effects
• Spatio-temporal analysis of plausible sites
• Offshore & Onshore wind statistics
• Region based analysis
• Techno-economic & Techno-commercial analysis
6
OUR SERVICES
Smart Grid Study
• Demand-supply characteristics
• Effectiveness of Smart metering technology
• Techno-economic & Techno-commercial analysis

7. Models & Simulation Study
• System Dynamics models
• Scenario Based simulation
• Fuzzy logic models
• DNN (Deep Neural Network), ANN (Artificial Neural Network) & GA (Genetic Algorithm) models
• Physical, Semi-physical & Phenomenological Analytical & Computational models
7
OUR SERVICES
Solar Energy Study
• Solar Energy Forecast & Scheduling services
• Reliability, life-span and yearly energy yield projection
• Degradation effects–Spatio-temporal analysis of plausible sites
• Roof-Top analysis
• Region based analysis
• Techno-economic & Techno-commercial analysis

8. 8
Why del2infinity
• Best Accuracy for all most All States at All Seasons; (State wise tentative accuracy is mentioned at the end of the profile, as
per data, available with del2infinity)
• Minimum Number of Revision, saves Cost of Labour/Man-Hour, no need for deployment of work force for 24x7;
• Minimum Number of Revision, saves Cost of SLDC Charges; SLDC may charge on revision for each time of Revisions in
future;
• del2infinity is the Only Indian F&S Service Provider developed software and SOS for Plant Specific Customized Model for
Better Accuracy;
• Proficient for Indian scenario considering Indian critical weather;
• Proficient for Indian scenario considering Critical Grid Conditions;
• Intelligent Data Integration: del2infinity works with available data intelligently;
• del2infinity can work in those cases where Historical Data are not available.
• Very useful for new projects;
• Can work in Offline Mode if Real-time data is not available hence even useful for small projects.

9. A Comparative Study of
RE Forecasting &
Scheduling (DSM)
Penalty for day ahead
forecasting &
scheduling of RE
Energy plants of various
QCA/Service Provider

11. How to do RE Forecasting

12. Without going complex structure of grid network let define a simple structure where we have three variable generation grid
nodes say G1, G2 and G3 for simplicity let define three Load dispatch points L1, L2 and L3 such that L1 lies between G1 and
G2, L2 lies between G2 and G3, and L3 lies between G3 and G1. The structure is made as simple as possible for the energy
flow such that a generation station can distribute its generation in its two nearest Load dispatch points. For simplification,
this analysis considers only energy flow in the network to find the stability of the network.

13. min 𝐺1 𝑡 , 𝐿3 𝑡 , 𝑇13, 𝑇32 − 𝐺3 𝑡 − 𝐿3 𝑡 ≥ 𝑋(𝑡) ≥ max 0, 𝐺1 𝑡 − 𝑇11, 𝐿3 𝑡 − 𝑇33
min 𝐺2 𝑡 , 𝐿1 𝑡 , 𝑇21 ≥ 𝑌(𝑡) ≥ max 0, 𝐺2 𝑡 − 𝑇22
𝐺3 𝑡 − 𝐿3 𝑡 + 𝐺2 𝑡 − 𝐿2 𝑡 ≥ 𝑌 𝑡 − 𝑋 𝑡 ≥ − 𝐺1 𝑡 − 𝐿1 𝑡
A(t) depends on the each value of G1, G2 and G3 but not on the sum of its values i.e. G1 + G2 + G3. Interestingly since it is a
variable generation and A(t) is not constant but to get the +ve value of A(t) we need a prediction of G1, G2 and G3 separately but
not as a sum or aggregation of those values.

14. Here the red area is actual requirement and the
area of A(t) decreases due to the uncertainty of
the generation.
Suppose schedule generation of G1, G2 and
G3 are not known separately, then the
above situation may arise:
The aggregated forecast creates instability when A(t) is not positive.

15. Data support by CSTEP, India
The Complex Network

16. For M Generating point and N Load point of a Complex Network, A(t) is
approximately (M+N – 4) dimensional space.
If A(t) is not a connected space then it creates the instability
Hence what is the minimum temporal and spatial granularity of Forecasting?
Minimum Temporal Granularity = 15 min already fixed
Minimum Spatial Granularity ?
Minimum Capacity of Generation?

17. Statistical Forecasting

18. Statistical Forecasting problem is an error minimization problem
min
𝑃
න
Ω
𝐹(𝐱, 𝑃(𝐱), 𝛻𝑃(𝐱))𝑑𝐱

19. min
𝑃
න
Ω
𝐹(𝐱, 𝑃(𝐱), 𝛻𝑃(𝐱))𝑑𝐱
𝛻
𝜕𝐹(𝐱, 𝑃(𝐱), 𝛻𝑃(𝐱))
𝜕 𝛻𝑃(𝐱)
=
𝜕𝐹(𝐱, 𝑃(𝐱), 𝛻𝑃(𝐱))
𝜕𝑃(𝐱)
Solve it using Euler-Lagrange:
And get an Iterative (or Dynamic) equation:
𝑃𝑡+1 𝐱 = 𝜙(𝑃𝑡 𝐱 )

20. Few Computational Techniques in
Solving the Dynamic Problem for Wind
/ Solar Forecasting
• Markov -> Hidden Markov
• ANN -> DNN
• PDE -> Stochastic PDE
• Single Hypothesis -> Multiple Hypothesis -
> Scenario based Analysis
• Fractional Calculus !

21. Fractional
World
•
𝑑
𝑑𝑥
,
𝑑2
𝑑𝑥2 exists
• What about
𝑑0.5
𝑑𝑥0.5 or
𝑑0.9
𝑑𝑥0.9 or
𝑑1.1
𝑑𝑥1.1 ?

22. Fractional Derivative
Riemann fractional derivative (Left) is defined as
1( ) 1
( ) ( ) ( )
( )
xn
n
L x n
L
d f x d
D f x x s f s ds
ndx dx

 


− −
= = −
 − 
( ) 11( )
( ) ( ) ( )
( )
n Ln
n
x L n
x
d f x d
D f x x s f s ds
ndx dx

 


− −−
= = −
 − 
Riemann fractional derivative (Right) is defined as
Other forms of Fractional derivative also exist like Riesz or Caputo fractional derivative

23. Why Fractional
Derivative ?
• It is non-local convolution type

24. L=0 => Riemann-Liouville Form of fractional derivative
min
𝑃
න
Ω
𝐹(𝐱, 𝑃(𝐱), 𝛻 𝛼,𝛽
𝑃(𝐱))𝑑𝐱
Define the error minimization problem as
And solve using Fractional Euler Lagrange as
𝛻 𝛽,𝛼
𝜕𝐹(𝐱, 𝑃(𝐱), 𝛻 𝛼,𝛽
𝑃(𝐱))
𝜕 𝛻𝑃(𝐱)
=
𝜕𝐹(𝐱, 𝑃(𝐱), 𝛻 𝛼,𝛽
𝑃(𝐱))
𝜕𝑃(𝐱)

26. 2626
Forecast Accuracy in Wind (R12) & Solar (R1) Forecast Plant Level
Absolute Error
Margin
Probability (%)
Wind
Probability (%)
Solar
< 15% 93.36 +/- 5 98.69 +/- 2.5
15%-25% 4.37 +/- 5 1.27+/-2.5
25%-35% 1.16 +/- 5 0.04+/- 2.5
>35% 1.11 +/- 5 0

27. Plant wise -
del2infinity’s accuracy
/ finding for various
states
Sr
No
Renewable Energy
Plants
Location State Capacity (MW) Del2infinity’s energy
accuracy / finding (%)
1 Solar Plant Durg Chhattisgarh 30 99.17
2 Solar Plant Khambat Gujarat 15 98.32
3 Solar Plant Bitta Gujarat 40.2 98.37
4 Solar Plant Porbandar Gujarat 15 99.69
5 Solar Plant Pavgada, Karnataka 10 99.41
6 Solar Plant Chalkere Karnataka 10 99.53
7 Solar Plant Hiriyur Karnataka 10 99.44
8 Solar Plant Hiriyur Karnataka 10 99.27
9 Solar Plant Hiriyur Karnataka 10 99.13
10 Solar Plant Dharmapur (newly commisioned < 2
month)
Karnataka 40 99.21
11 Solar Plant Hiriyur (newly commisioned < 2 month) Karnataka 40 99.24
12 Solar Plant Hiriyur (newly commisioned < 2 month) Karnataka 50 99.36
13 Wind Plant Bijapur Karnataka 165.2 93.56
14 Solar Plant Rajgarh Madhya Pradesh 25 98.86
15 Solar Plant Kapeli Madhya Pradesh 10 99.92
16 Solar Plant Muktsar Punjab 2.1 99.23
17 Solar Plant Badisid Rajasthan 10 98.63
18 Wind Plant Kaladongar Rajasthan 75.6 87.73
19 Solar Plant Sivgangi Tamil Nadu 5 98.63
20 Solar Plant Permabalur Tamil Nadu 10 98.73

28. Del2infinity’s
technical requirement
for RE forecasting &
scheduling solution
•Minimum 1 computer preferably Linux/Windows OS.
•Internet access with minimum 128 kbps rate.
•FTP based access service
Plant characteristics:
•Plant location and layout
•Plant installed and available capacity
•Wind Turbine / PV panel information (Manufacturing dataset if available)
•Weather station and/or sensor location (if available)
Historical (3-5 years) & Real time Data (15-minute Time Interval updated in 2
hours):
•Aggregated and/or Turbine level and/or Array and/or Panel level generation power
•Wind Speed / Solar Irradiance
•Temperature
•(Maintain a fixed file structure as del2infinity AI system is fully automated and
different file structure may create erroneous results.)
IPP/OEM should provide del2infinity:
•Curtailment information
•O&M schedule
•FTP / web based access (login & password, file structure) or specific Email id(s)
•Forecast & Scheduling format (specific file structure)
•After getting all the information, we need at least 7- 10 working days (depending on the nature of
information provided) to process and to analyze the information required for initial performance
analysis of forecasting before starting the forecast service.

29. del2infinity’s
deliverables
1.Plant characteristics
Analysis Report
before starting the
forecast service
2.Day-ahead forecast
(R0) in 15-minute
time-block (A sample
scheduling structure
is attached. It differs
for different SLDC)
3.Revision in every 2
hours (if applicable)
up to R12 (as decided
by del2infinity team)
4.Daily report on
Forecast accuracy and
one monthly report.

30. del2infinity’s selected
success stories

31. del2infinity’s selected success stories

32. del2infinity’s selected success stories

33. del2infinity’s
selected success
stories

34. Knowledge Sharing
by Team del2infinity
(Journal Publication)
❑ Gaming or Forecasting, Climate Samurai's January 2018 [https://view.publitas.com/climatesamurai-com/climate-samurai-january-2018-
issue/page/1]
❑ Game of Forecast or Gaming in Forecast, Saur Energy International [http://www.saurenergy.com/solar-energy-articles/game-of-
forecast]
❑ 'Submitting Forecast & Schedule of Solar Power generation: Communication with SLDC’' Saur Energy International
(http://www.saurenergy.com/solar-energy-articles/submitting-forecast-and-schedule-solar-power-generation-communication-sldc).
❑ 'Generation Forecasting for remaining life of WTG’’ published in WindPro-IWPA Magazine-May 2017.
❑ ‘Forecasting Daily Power Generation of Solar Plant’ has published in Council of Power Utilities INDIA POWER, April - June 2017, Vol
XXV No 2, Pg. 12
❑ 'Forecasting Solar Power Generation: Revise the Forecast when it requires' has published in EQ International Magazine, 10th April
2017. [http://www.eqmagpro.com/forecasting-solar-power-generation-revise-the-forecast-when-it-requires/]
❑ ‘Applicability of Error Limit in Forecasting & Scheduling of Wind & Solar Power in India’' has published in WINDPRO, IWPA Journal,
March 2017 issue @ pg 11.
❑ 'Forecasting Solar Power Generation in India: A Path of Unnecessary Revisions' SaurEnergy, 1st March 2017.
[http://www.saurenergy.com/articles/forecasting-solar-power-generation-india-path-unnecessary-revisions]
❑ 'Wind Power Forecasting in India' Abhik Kumar Das, WINDINSIDER, January 2017. [http://www.windinsider.com/index.php/16-
industry/863-wind-power-forecasting-in-india]]
❑ 'Forecasting Solar Power Generation: Variability vs. Predictability' Abhik Kumar Das, SolarQuarter, 6th February 2017.
[http://www.solarquarter.com/index.php/technology/2166-forecasting-solar-power-generation-variability-vs-predictabilityforecasting-
solar-power-generation-variability-vs-predictability]
❑ 'Solar Power: Forecasting in India' Abhik Kumar Das, Saur Energy, December 2016 @ 60-62.
❑ “An analytical model for ratio based analysis of wind power ramp events”, Abhik Kumar Das, Sustainable Energy Technology and
Assessments, Elsevier vol. 9, pp.49-54, March 2015
❑ “An Empirical Model for Ramp Analysis of Utility-Scale Solar PV Power”, Bishal Madhab Mazumdar, Md. Saquib, Abhik Kumar Das,
Solar Energy, Elsevier, vol. 107, September 2014
❑ “Quantifying photovoltaic power variability using Lorenz curve”, Abhik Kumar Das, Journal of Renewable and Sustainable Energy 6,
June 2014
❑ “Analytical derivation of equivalent functional form of explicit J–V model of an illuminated solar cell from physics based implicit model”,
Abhik Kumar Das, Solar Energy, Elsevier, May 2014
❑ “An Empirical Model of Power Curve of a Wind Turbine”, Abhik Kumar Das, Energy Systems, March 2014
❑ “An Explicit J–V Model of a Solar Cell using Equivalent Rational Function Form for Simple Estimation of Maximum Power Point
Voltage”, Abhik Kumar Das, Solar Energy, Elsevier, vol. 98(C), pp. 400-403, December 2013
❑ “Statistical Model for Wind Power based on Ramp Analysis”, Abhik Kumar Das & Bishal Madhab Majumder. International Journal of
Green Energy, 2013
❑ “Analytical Expression of the Physical parameters of an Illuminated Solar Cell using Explicit J-V Model”, Abhik Kumar Das, Renewable
Energy, Elsevier vol. 52, issue 1, pp. 95-98, April 2013
❑ "A Simple Explicit Model Approximating the Relationship between Speed and Density of Vehicular Traffic on Urban Roads", Abhik
Kumar Das & Jai Asundi, Int. J. of Critical Infrastructures, Vol.8, No.2/3, pp.195 – 204, 2012
❑ "Using the Gini Index to Measure the Inequality in Infrastructure Services Provided within an Urban Region", Abhik Kumar Das & Jai
Asundi Int. J. of Critical Infrastructures, Vol.8, No.2/3, pp.178 – 186, 2012
❑ "Analytical Derivation of Explicit J–V Model of a Solar Cell from Physics based Implicit Model", Abhik Kumar Das, Solar Energy,
Elsevier, vol. 86, issue 1, pp 26-30, January 2012
❑ "An Explicit J–V Model of a Solar Cell for Simple Fill Factor Calculation", Abhik Kumar Das, Solar Energy, Elsevier, vol. 85, issue 9, pp
1906-1909, September 2011
❑ "Analytical investigation of parabolic trough receiver performance with outer Vacuum Shell", Premjit Daniel, Yashavant Joshi, Abhik K
Das,Solar Energy, Elsevier, vol. 85, issue 9, pp 1910-1914, September 2011
❑ "Analytical Derivation of the Closed-form Power Law J-V Model of an illuminated Solar Cell from the Physics Based Implicit Model",
Abhik Kumar Das and Shreepad Karmalkar, IEEE Transactions on Electron Devices, vol. 58, No 4, pp 1176-1181, April 2011

35. Conference
Proceedings
by Team
del2infinity
➢ “A Simple Functional Relationship of Error distribution of Day-Ahead Powerr Generation Forecast and the
Variability of Power Generation,’’ presented at 1st International Conference on "Large-Scale Grid Integration of
Renewable Energy in India endorsed by the Indian Ministry of New and Renewable Energy as well as the Indian
Ministry of Power and organized by Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), United
States Agency for International Development (USAID) and Energynautics, Germany.
➢ Probability based Scenario Analysis and Ramping Correction Factor in Wind Power Generation Forecasting at
Windergy 2017, New Delhi
➢ Higher Efficiency in O&M (Monitoring, forecasting and data management solutions) at “India Solar Conference,
April 6th & 7th-2017, BIEC, Bangalore”
➢ 'Applicability of Error Limit in Forecasting & Scheduling of Wind & Solar Power in India’ has published in Indian
Smart Grid Week 2017 Compendium of Technical Papers organized by India Smart Grid Forum & Government of
India at Manekshaw Centre, New Delhi.
➢ 'Forecasting and Scheduling of Wind and Solar Power generation in India' Abhik Kumar Das, NTPC's Third
International Technology Summit ‘Global Energy Technology Summit’ 2016.
[http://www.ntpcgets.com/abstract/finalpapers/619.pdf]
➢ ‘A Neural Network based approach to predict high voltage Li-ion battery cathode materials”, Tanmay Sarkar,
Alind Sharma, Abhik Kumar Das and Mridula Dixit Bharadwaj, Proceedings of IEEE International Conference on
Devices, Circuits & Systems, (IEEE-ICDCS-12), March 2014 (in press)
➢ “Developing a GIS based Plume Rose for Industrial Chemical Incident Preparedness and Response”, Abhik Kumar
Das, Subhrajit Debnath, Jai Asundi, 2013 IEEE International Conference on Technology for Homeland Security,
Waltham, Massachusetts, USA, Nov 12-14, 2013
➢ “An Empirical Geometric Model for City Expansion”, Abhik Kumar Das & Sujaya Rathi, International Conference
on Intelligent Infrastructure, 47th Annual Convention of the Computer Society of India, Science City, Kolkata, Dec
1-2, 2012
➢ "Determination of the Peak Power Voltage Using Explicit PLM of an Illuminated Solar Cell", Abhik Kumar Das,
IEEE International Conference on Devices, Circuits & Systems, (IEEE-ICDCS-12), March 2012
➢ "Edge Filtering with Orientation Entropy", Abhik Kumar Das et al.,Proceedings of the International Conference on
Computing: Theory and Applications (ICCTA-2007), Kolkata, India, IEEE Computer Society Press, March 2007.
➢ "Stochastic Spectral Density Analysis on Network Traffic Characterization", Abhik Kumar Das & S. K.
Ghosh,Lecture Notes in Computer Science, Distributed Computing and Networking, LNCS vol. 4308, pp. 276-281,
Springer, December, 2006.
➢ "A Bidirectional Linear Semi-Systolic Architecture for DCT-Domain Image Resizing Processor", Abhik Kumar Das &
S. K. Ghosh, Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS-2006), Kos, Greece,
May, 2006.

36. Thank you !
del2infinity Energy Consulting
Email: contact@del2infinity.xyz;
Phone: +91-7760989341 ||
9891770702 || 9990433149