UNCERTAINTY TREATMENT IN ECONOMIC DISPATCH...

UNCERTAINTY TREATMENT IN ECONOMIC DISPATCH

WITH RENEWABLE ENERGYPRESENTED BY:

KIPKEMOI KIRUI GEOFFREYSUPERVISOR: MR. P.M MUSAU

EXAMINER: PROF. M. MANG’OLIDATE: 19TH MAY, 2016

IntroductionDefinition of Terms

Uncertainty is the (changefulness) unpredictability, inaccuracy, variability. It is a state for power system components, where it is impossible to exactly describe a future outcome, or more than one possible outcome due to limited knowledge [1].

Economic Dispatch is the determination of the optimal output of a

number of electricity generating facilities to meet the system load

at the lowest possible cost subject to generation, transmission and

operational constraints [39].

Objectives

i. To study uncertainties present in a power system.

ii. To study the effect of introducing renewable energy sources on the power system uncertainties.

iii. To determine how to deal with uncertainties in a power system.

Uncertainties present in a power system

• Generation availability• Load requirements• Market forces • Fuel prices • Forces of nature such as extreme weather• Technological developments• Regulatory uncertainties. New reliability

standards (environmental policies )

Effects of uncertainties introduced by RE to power system

• Stability: is the ability of a power system to maintainsynchronism when subjected to severe disturbance. Sudden disconnection of an entire solar or wind farm at full generation, the power system will lose the production capacity. Unless the remaining power plants have enough ‘spinning reserve’ to compensate for the loss in a short time, large frequency and voltage drops will occur and possibly followed by complete loss of power. To avoid this new generator technology is being developed that can ‘ride through’ and use of reactive power compensating devices.

Conti’ of Effects of uncertainties introduced by RE to a power system

• Security: is the ability of the system to withstand disturbances without causing a breakdown of the power system. Breakdowns leads to interruption of power to consumers. This can occur due to:

1. Insufficient active power reserve leading to load shedding. 2. Grid congestion (overloaded lines) that require disconnection

of loads to avoiding cascading faults. 3. Bus bar voltages getting out of permitted ranges 4. System running into stability problems (frequncy,voltage)

Conti’ of Effects of uncertainties introduced by RE to a power system

RE improves system securityi. Increases the diversity of a power system (no

sole dependence on one source)ii. Its resources are continuously replenished on

human timescales (fossil get depleted)However, variability and the unpredictability of wind and solar power can cause a power imbalance on the grid.

Conti’ of Effects of uncertainties introduced by RE to a power system• Power Quality: Degree of deviation from the

normal sinusoidal voltage and current waveforms.

To achieve integration of RE converters are used, these converters introduce harmonics.

Conti’ of Effects of uncertainties in a power system

Effects of harmonics 1. Excessive heating of equipment decreases

their lifetime.2. Increase line losses3. Cause flickers that result in an uncomfortable

visual effect on the eyes

Conti’ of Effects of uncertainties introduced by RE to a power system

• Phase ImbalanceMajority of PV sources are connected in the form of single-phase units. Unbalanced voltage profiles among phases can shift the neutral point to an unacceptable value.Unbalanced three-phase condition can lead to instability problems and higher network losses

How to reduce uncertainty in RE• Weather forecasts (RE is a function of weather)

However, has errors upto 20%This can be further improved by on‐site monitoring

Conti’ of How to reduce uncertainty of RE

A case study was conducted for 11 sites in the U.S. using modelled and measured data [14].A comparison of the results shows confidence can be increased by incorporating data from on‐site instrumentation. On‐site monitoring reduced uncertainty from 9.2% to 5.7% as shown in figure 2.3 below

Conti’ of How to reduce uncertainty of RE

Problem Formulation

Conti’ of Problem Formulation

Constraints

Uncertainty Treatment(Generation availability)

Uncertainty treatment in wind generators

dw (absolute difference) Crwj Cpwj

<=5 1.5 3.0

<=10 2.0 4.0

<=15 3.0 6.0

<=20 4.5 6.5

>20 7.0 8.5

Uncertainty treatment in solar generators

• Solar_Fuel_Cost = (Cpvi *(solar_a)) + Cppvi * ds+ Crpvi *ds

ds(absolute difference) Cppvi Crpv

<=5 1.5 3.0

<=10 2.5 4.0

<=15 5.5 6.0

<=20 6.5 7.5

>20 8.0 8.5

Methodology

PSO method was chosen to implement (UTIED).PSO is population based optimization techniquebased on the movement and intelligence of swarms.

Conti’ of Methodology

Why PSOa. Its simple concept and coding

implementationb. Faster in convergencec. Generate high-quality solutionsd. Convergence is independent of initial

points

PSO Parameter Representation

• Swarm: All possible generation from power plant • Particle: An individual power generation (solution) • Velocity: Rate of change from one possible solution

to another in iteration• Position: Generation at every instant relative to the

global best• Dimension: Number of generating units against the

number of possible generations (possible gen from all units)

FLOW CHART

Results

Generator type Power Generated in (MW) Cost ($/Hr)

Thermal generator 1 200 550Thermal generator 2 80 252Thermal generator 3 31.488 93.4562Thermal generator 4 35 123.967Solar generator 28 70Wind generator 39 68.25Iterations taken to converge 90Power loses 13.4793Total 413.488 1157.67

Conti’ of Results

Generator type Power Generated in (MW) Cost ($/Hr)

Thermal generator 1 200 550Thermal generator 2 80 252Thermal generator 3 31.4821 93.4273Thermal generator 4 35 123.967Solar generator 28 224Wind generator 39 189.25Iterations taken to converge 156Power loses 13.4792Total 413.482 1432.64

Total cost of Generation against amount of RE available at a constant

discrepancy

Total value of RE (MW)

70 60 50 40 30 20 10 0

Total cost of power generation ($/Hr)

943.245 959.886 977.073 995.878 1018.09 1043.62 1075.52 1124.81

At a penetration level of 50% RE

Alpha(Available RE-predicted RE )

0 5 10 15 20

30% RE penetration cost of generation($/Hr)

766.243 891.257 1016.25 1141.21 1316.2

50% RE penetration cost of generation($/Hr)

735.15 915.108 1095.12 1275.15 1433.16

Graph of cost of generation against level of discrepancy(alpha)

Graph of a cost of generation against available RE at a discrepancy level of (0)

A graph of cost of generation for 30% and 50% penetration level of RE against

discrepancy

Discussion• The cost of generation is minimum when the difference

between the predicted and the available power is minimum. As the discrepancy between the predicted and the available power increases the cost also increases. This is therefore calls for precise prediction of RE power for efficient planning of operations of the power system. Increasing uncertainty increases costs incurred due to penalties and cost of reserves.

• Cost of generation reduces with increase in the amount of RE power under minimum uncertainty. This is because thermal generators are fuel dependant while for RE there is no cost of fuel, the resource is available in nature for free. The only costs incurred are maintenance and operating costs.

Conti’ of Discussion• At a penetration level of 50% of RE, and low

uncertainty levels the cost of generation is lower than that at 30% level of RE penetration. However, as uncertainty of RE increases the cost of generation with high penetration level increases as shown by figure 4.4 above. Integration of RE in power system can only be economical if uncertainty in the prediction of resources can be done accurately. Accurate forecasting of RE, minimizes costs of reserves and penalties incurred.

Conclusion

• This project has developed a model to successfully include wind and solar generators in the economic dispatch problem

• Has developed a model of treating uncertainty in generation availability of RE

Recommendation

• This project considered uncertainty on generation availability; in future it could be extended to consider other uncertainties like load requirements, fuel prices, regulatory uncertainties and uncertainties due to nature such as extreme weather conditions.

THANKYOU