d Load Curves ts opuCTION consumers. Load curves help is in understan 4 he needs of. nae i tricity is See factor, capacity factor sod ato ie play an ia th changes in load Se one classification of power plants is base peak load aa role in generation P ENTS 5 aa ce types of consumers—domestic, commercial, industrial, Agricul feeds ss stations invariably feed a grid which delivers power to the load etc. The present Se esa or system must take into account the future increase in load. Forth Ce ae Pan studies have to be made to predict the increase in load in the next x pee eral power station takes 5 to 10 years for completion and some hydro-electy, aa ae even longer time. As such the power system planning must be done about 29 ean e . . : 5 : : peace The estimate of power and the increase in generation capacity depend on ma: demand, the distribution and variation in demand and the energy requirements. 2.2. ENERGY A power station 2.3. MAXIMUM DEMAND Each device at the consumer terminals has its rated capacity. The connected load of a consume means the sum of the continuous ratings of all the devices and outlets installed on his distributiot circuit. The maximum demand of a consumer means the maximum power that his circuit is likely to draw at any time. If all the devices and outlets were used simultaneously to the full extent, the maximum demand of the consumer would equal his connected load. Experience has shown thé consumers do not use all the devices at full load simultaneously. The maximum demand of eact consumer is, therefore, less than his connected load. The maximum demand and the connectet load are related by Demand Factor = Maximum Demand (lt Connected Load yaad ula ee 100 watt, his connected load is 1000 watt (assum simultaneously. At Seri fe peices). However he would not use all the 10 “a am een 3 'y he may use only 3 bulbs, at another time only 2 and st! the maximum number of im si eously is six, contrat : rn er of bulbs used by him simultaneously is demand is 600 watt and d mand factor i: Each device will er ictor is 0.6, indicates the contribution A denny & Sometime during its operation. The demaot levice towards the maximum demand of the = 18{OADS AND LOAD CURVES 19 LOADS At el! pemand factors fr various types of loads vary within very wide limits, Some lypleal values are fever" Table 2.1, ‘Table 2.1, Demand factors Tipe of Load Pemand Factor Residential thw and less, 07 aw to LkW 06 1 KW and above 05 Commercial Restaurants, shops and offices, 07 Educational institutions, 05 Theatre, Cinema, 0.5 Hotels 05 Industrial Cottage (less than 5 kW) 08 Small Scale (5 kW to 25 kW) 07 Medium (25 kW to 100 kW) 06 Large (More than 100 kW) 0.5 Example 2.1. A residential consumer has the following connected load: 8 bulbs of 100 W exch, 2 fans of 60 W each and 2 light plug points of 100 W each. His use of electricity during a day is as under: 12 midnight to 5 am ‘one fan Samto7 am 2 fans and one light point - 7am to9am NIL : 9am to 6 pm 2 fans 6 pm to midnight 2 fans and 4 bulbs Find (a) connected load (5) maximum demand (c) demand factor (d) energy consumed 24 hours (e) energy consumed in 24 hours if all devices are used all the day. Solution. (a) Connected load = 8 x 100 + 2 x 60 +2 x 100 = 1120 W (0) Total wattage at different times is : : ce 6oW / 2x 60+ 1x 100= 220 W-GENERATION OF ELECTRICAL ENER, = 0 + 1080 + 3120 ing 24 hours = 300 + 44 Tova energy consumed during 24 hours = 300% he energy consumed is = 1120 x 24 = 26880 Wh = 26.88 kWh (©) Ifal devices are used throughout the day, # ‘The load actually used by a consumer at any time dur that time. Since the requirements at different times an nt during the day. 2.3.1. Group Diversity Factor. a day depends on his requirements at different, the load varies from instant to instal ‘ - i a "The electric supply authorities have to make provision for generation, transmission anj distribution facilities to supply loads to different consumers. The investment in all these facilitie is proportional to kVA ratings of these equipments. Therefore every electric supply authority ha to make a reasonable estimate of the maximum kVA demand of the area supplied by it. In every area there are a number of consumers. Let the number of consumers served by on particular distribution transformer be 8. This group of 8 consumers will have different electricg gadgets at their premises and each consumer will use the gadgets as per his requirement. Each om of the consumers will have a certain maximum demand, However, it is likely that the maximua demands of different consumers occurs at different times. This is due to different habits ani requirements of these 8 consumers. If this happens, the maximum demand of the group will be lex ‘than the sum of their maximum demands. The contribution of the maximum demand of a consume to the power requirements of the group depends on the group diversity factor Group Diversity Factor = Su of individual maximum demands = pees, Maximum demand of the group ~(Q2 divided Beaten aca es than unity. The consumers maximum demas | ee termine his effective demand at the distribute | E: we 2 A group of 2 consumers has the following electricity demand pattem on! ‘Consumer A: Connected load Load from 12 midnight a Eee icon ees a to 8 am LL kW from 8 am to $ pm ae som 5 Pm to 12 midnight NIL nf load 500 W 3 kW NIL 300 Ww Vkw 200 wcee ee ity factor It is evident i A ; the capacity facto! and lod Tactor Becomes ths 2d plant capacity equals the maximum load, the 2.9. UTILISATION FACTOR Iris defined as the ratio of the maximum demand to the rated capacity of plant. Utilisation factor < Maximum Load on (2.13) Rated Plant Capacity The ubisation factor fora plant depends on the use to which the plant is put. A low utilisation factor means that the plant is either a standby plant or has been installed to take into account the future increase in the load. For a power plant, in an integrated system, a high utilisation aoe shows that the plant is probably the most efficient in the system. Since loads in excess of rate capacity are also carried, sometimes, utilisation factor may be more than unity. It is evident that Capacity Factor = (Load Factor) x (Utilisation Factor) va (2.14) 2.10. BASE LOAD AND PEAK LOAD PLANTS The system load varies from time to time. A typical chronological curve is as shown in Fig. 2.11. The maximum demand is M. If the total demand is supplied from one power plant havingan installed capacity M (or somewhat higher to keep some reserve capacity), the plantwill be running under loaded most of the time, thus making the operation uneconomical. A better method is to divide the load into two portions, one below the line AB and the other above the line AB. These loads are referred to as base load and peak load respectively. These two loads are supplied from separate plants called the base load plant and the peak load plant. A base load plant operates at a high load factor and should be one which has low operating costs. The peak load 12 p 2S aa 4 plant operates at a low load factor. Sometimes the AM TIME—> PM load curve can be divided into three portions ie. Fig, 2.11. Base load and peak load. base load, intermediate load and peak load. Example 2.4. The load on a power plant on.a typical day is as under: M LOAD MW —> BASE LOAD Time 125am $-9am = 9-6 pm") Spm-10pm 10 pm-12 am Load (MW) 20 40 80 100 20 Plot the chronological load curve and load duration curve. Find the load factor and the energy supplied by the plant in 24 hours. | Solution. The chronological load curve is plotted in Fig. 2.12. The duration of 1 80 and 40 and Load (MW: 100 ang above above Duration (Hours) 4 13 7 “The load duration curve is plotted in Fig. 2.13. The energy produced byLOADS AND LOAD CURVES 31 100 1500, 80 1200 ha |» 3 40 3 600 20 300 “5 ab: ay 2 je) eee ENERGY — eae ™ME—> Fig. 2.14. Energy load curve for Example 2.4. Fig 2.15. Mass curve for Example 2.4. _ Example 2.7. The maximum demand of a power plant is 40 MW. The capacity factor is 0.5 and the utilisation factor is 0.8. Find (a) load factor (b) plant capacity (c) reserve capacity (d) annual energy production. [PU. 1991} Capacity factor Utilisation factor Solution. (a) Load factor = Max.demand _ 40 Utilisation factor 0.8 () Reserve capacity = 50-40 = 10 MW (@) Annual Energy Production = 40 0.625 x 8760 = 219000 MWh Example 2.8. The load curve of an electrical system is linear with the following different times of the day: Time 12 Sam 9 6 8pm 10 Load (MW) 50 50 100 100 150 80 Plot chronological load curve. Find the energy required by the system in one day and the system daily load factor. [PU. 1991} Solution. The chronological load curve is ploued in Fig. 2.16. Energy required by the system in 24 hours t = 50354100 410049410 x2 : (6) Plant capacity = =50 MW + Hearsa nee 19%, Lighting an usually consists of: Induction motors 50-70%, Lig 10%, Transmission losses etc. 10-20%. ‘ power. Since the resistang 2.11.1. Effect of Voltage. ny reactive PY gy? but approxi (4) Lighting load. Lighting loads do not ae oat vary as (voltage) ae of lighting devices is not constant, the active power dO°S as (voltage)!-®, active (6) Heating. The resistance is almost constant ae : (€) Induction motors. The performance of an In aie in a rather comp! mane equivalent circuit. The active and reactive power vary with VOT ith speed and also whethe effect also depends on whether the shaft torque is constant OF VANE" age causes an increas the motor is running at full load or less than full load. The a is constant, a 10% drop in voltag in current so thatthe effect on power is smal. Ifthe shaft torque is COMM tse descend at full load condition may cause about 0.5% decrease in Oe eduction in voltage. appreciably (to about 0.8 pu) the power increases with further " gjaWoltage and remains marl (@) Synchronous motors. The active power does not cay i the leading direction. less, constant. A reduction in voltage causes the vars to increas rly Constarl sopeaa In most of the analysis the composite sub-station loads are rept it is only an approximate so that both P and Q are assumed to vary as (voltage) However representation. 2.11.2. Effect of Frequency. The power consumed by a resistive load does as eas frequency. However most of the loads are inductive in nature. The impedance of an inductive load depends on the frequency and therefore the power drawn by an inductive load depends on frequency. The analysis of the effect of frequency on motor load is rather complex. Since only very small changes in frequency are permissible, the effect of frequency on active and reactive Power is neglected in calculations. However in load frequency control studies it is generally as sumed that a one per cent decrease in frequency causes a one per cent decrease in active power. aeey ower varies as (voltage) motor can be analysed from jg 2.12. LOAD FORECASTING Load forecasting can be divided into two categories: Long term load forecasting and Short term load forecasting. 2.12.1. Long term Load Forecasting. It takes a pretty long time to plan, install and commission additional generating capacity. Generation system expansion planning starts with a forecast of anticipated future load requirements. Proper long term load forecasting is necessary for optimal generation capacity expansion, In preparing a forecast, the system planner is confronted with the following questions: 1. Should the maximum demand be forecast using forecasted energy and load factors of should it be forecast separately. 2. Should the total forecast be obtained by combining the forecast appropriate load components or should the total forecast be directly obtained from etait load cana Bs Should simple forecasting method be ueed or should morbiformal mathemsatioal provedures be employed. Each of the above methodologies has its own advantages and limitations. approach is used by all the utilities consistently. Choosing the best method for a given system seauien, oodLOADS AND LOAD CURVES Straight line yeA+Br Parabola yrA+Br+Ce s curve yeAt Br + Cr + De Exponential y= Ce Gompertz yedn' (A+ Ce) Combination y=A+Br+ Co™ where y is the load in year x (with base year zero) and A, B, C, D are constants computed, generally, by the method of least squares. If the uncertainty of extrapolated results is to be quantified using statistical entities like mean and variance, the basic technique becomes probabilistic extrapolation. The use of stochastic models to generate a forecast from random inputs derived from historical data has also been suggested but is not used in actual practice. Another technique for load forecasting is correlation. This technique relates system loads to various demographic and economic factors. Typically factors like population, employment, industrial licenses, appliance saturation, weather data etc. are used in correlation i forecasting of the demographic and economic factors is rather difficult. 2.12.2. Short term Load Forecasting. A precise short term load forecast is essentially for ‘monitoring and controlling power system operation. The hourly lond forecast with lead time upto ‘one week in advance is necessary for on line solution of scheduling problems. A 24 hour load forecast is needed for successful operation of the power system. One hour forecast is important for on line real time control and security evaluation of a large power system. Short term load forecasting techniques generally involve physical decomposition of load into components. The load is decomposed into a daily patter reflecting the difference in activity level during the day, a weekly pattem representing the day of the week effect on load, a trend ‘component concerning the seasonal growth in load and a weather sensitive component reflecting the deviations in load due to weather fluctuations. The random error can be statistically analysed ‘to obtain a stochastic model for error estimation. ‘Thus the expected hourly load forecast is divided into five components and written as YG.) = ADP() + AWP (kf) + WSC (i, ) + TR@ + SEC GD ~~ 21S) ‘Where Y(i, j) = load forecast for j th hour of i th day ADP (/) = average daily load pattern at j th hour AWP (k, f) = average weekly load increment pattern at j the hour and & th day of the week (k = 1, 2... 7) WSC (i. f) = weather sensitive component at j th hour of { th day TR (i) = trend component of load on i th day SEC) = ‘stochastic error component which is assumed to be normally distributed. the hour of the day effect, It is an average of the daity “a