Mobile QR Code QR CODE : The Transactions P of the Korean Institute of Electrical Engineers

  1. (Dept. of Electrical Engineering at Chungnam National University, Korea.)



Distribution line, Operation limit, Energy storage system, ARIMA model, Load forecasting, QP optimization, Day-ahead scheduling, Real-time compensation

1. ์„œ ๋ก 

์—๋„ˆ์ง€์ €์žฅ์‹œ์Šคํ…œ(Energy Storage System, ESS)์€ ๊ณ ๊ฐ๋‹จ, ๋ฐœ์ „๋‹จ, ๊ทธ๋ฆฌ๊ณ  ์ „๋ ฅ๊ณ„ํ†ต์— ๋‹ค์–‘ํ•˜๊ฒŒ ์„ค์น˜๋˜์–ด ํ™œ์šฉ๋˜๊ณ  ์žˆ๋‹ค. ESS๋Š” ์‹œ๊ฐ„๋Œ€๋ณ„์ฐจ๋“ฑ์š”๊ธˆ์ œ(TOU)์— ๋”ฐ๋ผ ์š”๊ธˆ์ด ๋‚ฎ์€ ์‹œ๊ฐ„๋Œ€์— ์ถฉ์ „ํ•˜๊ณ  ์š”๊ธˆ์ด ๋†’์€ ์‹œ๊ฐ„๋Œ€์— ๋ฐฉ์ „ํ•จ์œผ๋กœ์จ ๊ณ ๊ฐ์˜ ์š”๊ธˆ์„ ์ ˆ๊ฐํ•  ์ˆ˜ ์žˆ๋‹ค[1]. ESS๋Š” ์žฌ์ƒ์—๋„ˆ์ง€ ์ถœ๋ ฅ์˜ ๊ฐ„ํ—์„ฑ๊ณผ ๋ณ€๋™์„ฑ์„ ์™„ํ™”ํ•จ์œผ๋กœ์จ ๋ณด๋‹ค ์•ˆ์ •์ ์ธ ์ถœ๋ ฅ์„ ์ „๋ ฅ๊ณ„ํ†ต์— ์ œ๊ณตํ•  ์ˆ˜ ์žˆ๊ธฐ ๋•Œ๋ฌธ์— ์ „๋ ฅ๊ณ„ํ†ต์˜ ์žฌ์ƒ์—๋„ˆ์ง€ ์ˆ˜์šฉ๋ ฅ์„ ์ฆ๋Œ€์‹œํ‚ฌ ์ˆ˜ ์žˆ๋‹ค[2]. ESS๋Š” ์ „๋ ฅ๊ณ„ํ†ต์˜ ์ฃผํŒŒ์ˆ˜ ์กฐ์ •๊ณผ ์ „์•• ์กฐ์ •์„ ์œ„ํ•œ ๋ณด์กฐ์„œ๋น„์Šค๋ฅผ ์ œ๊ณตํ•จ์œผ๋กœ์จ ์ „๋ ฅ๊ณ„ํ†ต์˜ ์•ˆ์ •๋„์™€ ์‹ ๋ขฐ๋„ ํ–ฅ์ƒ์— ๊ธฐ์—ฌํ•œ๋‹ค[3]. ๋˜ํ•œ, ESS๋Š” ๋ฐฐ์ „์„ ๋กœ, ๋ณ€์••๊ธฐ ๋“ฑ์˜ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜๋Š” ์ˆ˜์š”๋ฐœ์ƒ ์ง€์—ญ์— ์„ค์น˜ํ•˜์—ฌ ์„ค๋น„์˜ ์‹ ์„ค์„ ์–ต์ œํ•จ์œผ๋กœ์จ ํˆฌ์ž๋น„ ์ ˆ๊ฐ ๋˜๋Š” ์ง€์—ฐ ํšจ๊ณผ๋ฅผ ์ œ๊ณตํ•œ๋‹ค[4].

๋ณธ ๋…ผ๋ฌธ์€ ์ „๋ ฅํšŒ์‚ฌ๊ฐ€ ๋ฐฐ์ „์„ ๋กœ์˜ ๊ณผ๋ถ€ํ•˜๋ฅผ ์กฐ์ ˆํ•˜์—ฌ ์šด์ „์šฉ๋Ÿ‰์„ ์œ ์ง€ํ•˜๊ธฐ ์œ„ํ•œ ๋ชฉ์ ์œผ๋กœ ์„ค์น˜ํ•œ ESS์˜ ์šด์ „ ๋ฐฉ๋ฒ•์„ ์ œ์‹œํ•˜๊ณ  ์žˆ๋‹ค. ESS๋ฅผ ๊ณผ๋ถ€ํ•˜๊ฐ€ ๋ฐœ์ƒํ•˜๋Š” ๋ฐฐ์ „์„ ๋กœ์— ์„ค์น˜ํ•˜์—ฌ ํ‰์ƒ์‹œ ์ถฉ์ „ํ•˜๊ณ  ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜๋Š” ๊ณผ๋ถ€ํ•˜๊ฐ€ ๋ฐœ์ƒํ•˜๋Š” ์‹œ๊ฐ„๋Œ€์— ๋ฐฉ์ „ํ•จ์œผ๋กœ์จ ๋ฐฐ์ „์„ ๋กœ์˜ ์šด์ „์šฉ๋Ÿ‰์„ ์œ ์ง€ํ•˜๊ณ , ๋ฐฐ์ „์„ ๋กœ์˜ ์‹ ์„ค๋„ ์–ต์ œํ•˜๋Š” ๋ฐฉ์•ˆ์„ ๋ชจ์ƒ‰ํ•˜๊ณ  ์žˆ๋‹ค.

์„ ํ–‰ ์—ฐ๊ตฌ์—์„œ๋Š” ํ”ผํฌ ์ €๊ฐ์„ ์œ„ํ•œ ESS์˜ ์ถฉ๋ฐฉ์ „์„ ์œ„ํ•ด ๋กœ์ง ๊ธฐ๋ฐ˜์˜ ์šด์ „ ๋ฐฉ๋ฒ•๊ณผ ๋ถ€ํ•˜์˜ˆ์ธก์„ ํ†ตํ•œ ์ตœ์ ํ™” ๊ธฐ๋ฐ˜์˜ ์šด์ „ ๋ฐฉ๋ฒ•์„ ์ œ์‹œํ•˜๊ณ  ์žˆ๋‹ค[5-6]. ๊ทธ๋Ÿฌ๋‚˜, ๋กœ์ง ๊ธฐ๋ฐ˜์˜ ์šด์ „๋ฐฉ๋ฒ•์€ ์ถฉ๋ถ„ํ•œ ๋งˆ์ง„์„ ๋‘๊ณ  ์šด์ „ํ•˜๊ธฐ ๋•Œ๋ฌธ์— ํšจ์œจ์ด ์ข‹์ง€ ์•Š๊ณ , ๋ถ€ํ•˜์˜ˆ์ธก์„ ํ†ตํ•œ ์ตœ์ ํ™” ๊ธฐ๋ฐ˜์˜ ์šด์ „ ๋ฐฉ๋ฒ•์€ ์‹ค์ œ๋ถ€ํ•˜๊ฐ€ ์˜ˆ์ธก๋ถ€ํ•˜์™€ ์ฐจ์ด๊ฐ€ ์žˆ์„ ๊ฒฝ์šฐ ์ ์ ˆํ•œ ์ถฉ๋ฐฉ์ „ ๋™์ž‘์„ ์ˆ˜ํ–‰ํ•˜์ง€ ์•Š์„ ์ˆ˜ ์žˆ๋‹ค.

๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ๊ทธ๋ฆผ 1๊ณผ ๊ฐ™์ด ๊ณผ๋ถ€ํ•˜ ๋ฐœ์ƒ์ด ์žฆ์€ ๋ฐฐ์ „์„ ๋กœ์— ESS๋ฅผ ์„ค์น˜ํ•˜์—ฌ ํ‰์ƒ์‹œ ์ถฉ์ „ํ•˜๊ณ  ๊ณผ๋ถ€ํ•˜๊ฐ€ ๋ฐœ์ƒํ•  ๊ฒฝ์šฐ ๋ฐฉ์ „ํ•˜์—ฌ ์„ ๋กœ๋ถ€ํ•˜๊ฐ€ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜์ง€ ์•Š๋„๋ก ๊ด€๋ฆฌํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์ œ์‹œํ•˜์˜€๋‹ค. ์ผ์ •๊ธฐ๊ฐ„์˜ ๋ถ€ํ•˜์ด๋ ฅ์„ ํ™œ์šฉํ•˜์—ฌ ํ•™์Šต๋œ ARIMA(Auto-Regressive Integrated Moving Average) ๋ชจ๋ธ์„ ํ†ตํ•˜์—ฌ ๋ฐฐ์ „๋ง๊ด€๋ฆฌ์‹œ์Šคํ…œ(Distribution Management System, DMS)์ด ํ•˜๋ฃจ์ „ ํ•ด๋‹น์„ ๋กœ์˜ ๋ถ€ํ•˜๋ฅผ ์˜ˆ์ธกํ•˜์—ฌ ์ต์ผ ๋ฐฐ์ „์„ ๋กœ์— ๊ณผ๋ถ€ํ•˜๊ฐ€ ๋ฐœ์ƒํ•˜๋Š” ์ง€ ํ™•์ธํ•˜๊ณ , ์ต์ผ ๊ฐ ์‹œ๊ฐ„๋Œ€๋ณ„ ์˜ˆ์ธก๋ถ€ํ•˜๋ฅผ ๊ธฐ์ค€์œผ๋กœ ESS์˜ ์šด์ „์Šค์ผ€์ค„์„ ์ตœ์ ํ™” ๊ธฐ๋ฒ•์„ ํ†ตํ•˜์—ฌ ๊ตฌํ•˜๊ณ , ESS๋กœ ์ง€๋ นํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์ œ์‹œํ•˜์˜€๋‹ค. ๋˜ํ•œ, ์ต์ผ ์‹ค์ œ๋ถ€ํ•˜๊ฐ€ ์˜ˆ์ธก๋ถ€ํ•˜์™€ ์ฐจ์ด๊ฐ€ ์žˆ์„ ๊ฒฝ์šฐ, DMS๊ฐ€ ์‹œ๊ฐ„๋Œ€๋ณ„ ๋ถ€ํ•˜ ์ธก์ •๊ฐ’์„ ๋ฐ”ํƒ•์œผ๋กœ ESS์˜ ์ถฉ๋ฐฉ์ „ ๋ณด์ƒ๊ฐ’์„ ๊ตฌํ•˜๊ณ  ๋งค ์‹œ๊ตฌ๊ฐ„์—์„œ ESS๋กœ ๋ณด์ƒ๊ฐ’์„ ์ง€๋ นํ•˜๋Š” ์‹ค์‹œ๊ฐ„ ๋ณด์ƒ ๋ฐฉ์•ˆ์„ ์ œ์‹œํ•˜์˜€๋‹ค. ์‹ค์ œ ๋ฐฐ์ „์„ ๋กœ์— ์ ์šฉํ•˜์—ฌ ์‹ค์ œ๋ถ€ํ•˜๊ฐ€ ์˜ˆ์ธก๋ถ€ํ•˜์™€ ์ฐจ์ด๊ฐ€ ์žˆ๋Š” ๊ฒฝ์šฐ์— ๋Œ€ํ•˜์—ฌ ์‚ฌ๋ก€์—ฐ๊ตฌ๋ฅผ ์ˆ˜ํ–‰ํ•˜์—ฌ ์ œ์•ˆ๋œ ๋ฐฉ๋ฒ•์˜ ์œ ํšจ์„ฑ์„ ํ™•์ธํ•˜์˜€๋‹ค.

๊ทธ๋ฆผ 1. ๋ฐฐ์ „์„ ๋กœ์šฉ ESS ์šด์˜ ๊ฐœ๋…๋„

Fig. 1. ESS operation concept for a distribution line

../../Resources/kiee/KIEEP.2023.72.3.139/fig1.png

2. ๋ฐฐ์ „์„ ๋กœ ์šด์ „์šฉ๋Ÿ‰

ํ•œ์ „ ๋ฐฐ์ „๊ณ„ํ†ต์„ ๊ตฌ์„ฑํ•˜๋Š” ACSR-OC 160ใŽŸ ๊ฐ€๊ณต์„ ๋กœ์˜ ๊ฒฝ์šฐ, ํ—ˆ์šฉ์ „๋ฅ˜๋Š” ์—ฐ์†์‚ฌ์šฉ์˜จ๋„ 90โ„ƒ ์ด๋‚ด ์กฐ๊ฑด์—์„œ ์ตœ๋Œ€ 395[A] ์ด๊ณ , ํ—ˆ์šฉ์šฉ๋Ÿ‰์€ ์ตœ๋Œ€ 15,700[kW] ์ด๋‹ค. ์ด๋ฅผ ๊ธฐ์ค€์œผ๋กœ ๋ฐฐ์ „์„ ๋กœ์˜ ์šด์ „์šฉ๋Ÿ‰์€ ์ฃผ๋ณ€์˜จ๋„, ๊ฒฝ๋…„์—ดํ™”, ๋ถˆํ‰ํ˜• ๋“ฑ์„ ๊ณ ๋ คํ•œ ์šด์ „๋งˆ์ง„์„ ์ ์šฉํ•˜์—ฌ ๋น„์ƒ์‹œ 14,000[kW], ์ƒ์‹œ 10,000[kW]๋กœ ์ •ํ•ด์ ธ ์žˆ๋‹ค[7]. ๋น„์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์— ๋น„ํ•˜์—ฌ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์ด ์ž‘์€ ์ด์œ ๋Š” ๊ทธ๋ฆผ 2์™€ ๊ฐ™์ด ๋ฐฐ์ „์„ ๋กœ๊ฐ€ ์ตœ์†Œ 3๋ถ„ํ•  3์—ฐ๊ณ„๋ฅผ ๊ธฐ์ค€์œผ๋กœ ๊ตฌ์„ฑ๋œ๋‹ค๋Š” ๊ฐ€์ •ํ•˜์— ์—ฐ๊ณ„์„ ๋กœ์— ๊ณ ์žฅ์ด ๋ฐœ์ƒํ•  ๊ฒฝ์šฐ ์—ฐ๊ณ„์„ ๋กœ์˜ ๊ฑด์ „๊ตฌ๊ฐ„ ๋ถ€ํ•˜๋ฅผ ๋„˜๊ฒจ ๋ฐ›๊ธฐ ์œ„ํ•˜์—ฌ ์—ฌ์œ ๋ฅผ ๋‘๊ธฐ ๋•Œ๋ฌธ์ด๋‹ค.

๊ทธ๋ฆผ 2. ์ƒ์‹œ ๋˜๋Š” ๋น„์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰

Fig. 2. The normal or emergency operation limit

../../Resources/kiee/KIEEP.2023.72.3.139/fig2.png

๋”ฐ๋ผ์„œ, ์ผ๋ฐ˜์ ์œผ๋กœ ๋ฐฐ์ „์„ ๋กœ์˜ ๋ถ€ํ•˜๊ฐ€ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜๋Š” ๊ฒฝ์šฐ ์ผ์ •๋Ÿ‰์˜ ๋ถ€ํ•˜๋ฅผ ์—ฌ์œ ๊ฐ€ ์žˆ๋Š” ์—ฐ๊ณ„์„ ๋กœ๋กœ ์ „ํ™˜ํ•˜๊ฑฐ๋‚˜ ์„ ๋กœ๋ฅผ ์‹ ์„คํ•˜์—ฌ ๋ถ€ํ•˜๋ฅผ ์กฐ์ •ํ•˜์—ฌ์•ผ ํ•œ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜, ๋Œ€๋„์‹œ ๋„์‹ฌ์˜ ๊ฒฝ์šฐ ์ „๊ธฐํ™”์— ๋”ฐ๋ผ ๋ถ€ํ•˜๊ฐ€ ๊ณ„์† ์ฆ๊ฐ€ํ•˜๋Š” ์ถ”์„ธ์—๋„ ๋ถˆ๊ตฌํ•˜๊ณ  ๊ธฐ์„ค ๋ฐฐ์ „์„ ๋กœ์˜ ํฌํ™”๋กœ ๋ถ€ํ•˜์ „ํ™˜ ๋˜๋Š” ์„ ๋กœ์˜ ์‹ ์„ค์ด ์šฉ์ดํ•˜์ง€ ์•Š๋‹ค. ๊ทธ๋ž˜์„œ, ESS, ์ˆ˜์š”๋ฐ˜์‘ ๋“ฑ์„ ํ™œ์šฉํ•˜์—ฌ ๋ถ€ํ•˜๋ฅผ ์œ ์—ฐํ•˜๊ฒŒ ์กฐ์ •ํ•˜๋Š” ๋น„์ „์„ ๋Œ€์•ˆ(Non-Wires Alternative, NWA)์ด ๊ด€์‹ฌ์„ ๋ฐ›๊ณ  ์žˆ๋‹ค[4].

3. ๋ฐฐ์ „์„ ๋กœ ๋ถ€ํ•˜์˜ˆ์ธก

๋ฐฐ์ „์„ ๋กœ์˜ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰ ์œ ์ง€๋ฅผ ์œ„ํ•œ ESS์˜ ์šด์˜์ „๋žต์€ ํ•˜๋ฃจ์ „ ๋ถ€ํ•˜์˜ˆ์ธก์„ ํ†ตํ•ด ํ•ด๋‹น์„ ๋กœ์— ์‹œ๊ฐ„๋Œ€๋ณ„๋กœ ๊ณผ๋ถ€ํ•˜๊ฐ€ ๋ฐœ์ƒํ•˜๋Š” ์ง€ ์ ๊ฒ€ํ•˜๊ณ  ์‹œ๊ฐ„๋Œ€๋ณ„๋กœ ์ ์ ˆํ•œ ์ถฉ๋ฐฉ์ „๊ณ„ํš์„ ๋งˆ๋ จํ•˜๋Š” ๊ฒƒ์ด๋‹ค. ์ด๋ฅผ ์œ„ํ•ด ์ •ํ™•ํ•œ ๋ถ€ํ•˜์˜ˆ์ธก์ด ์ค‘์š”ํ•˜๋‹ค.

๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ์ข…์†๋ณ€์ˆ˜์ธ ๋ฏธ๋ž˜ ๋ถ€ํ•˜ ์˜ˆ์ธก๊ฐ’์„ ๊ตฌํ•˜๊ธฐ ์œ„ํ•ด ๊ณผ๊ฑฐ ์‹œ๊ณ„์—ด ๋ถ€ํ•˜ ์‹ค์ธก๊ฐ’์„ ๋…๋ฆฝ๋ณ€์ˆ˜๋กœ ์‚ฌ์šฉํ•˜๋Š” ์ž๊ธฐํšŒ๊ท€(Auto- Regressive Integrated Moving Average, ARIMA) ๋ชจ๋ธ์„ ๋„์ž…ํ•œ๋‹ค[8]. ํŠนํžˆ, ํ•ด๋‹น์„ ๋กœ์˜ ๊ณผ๊ฑฐ ๋ถ€ํ•˜ ์‹ค์ธก๊ฐ’์€ 1์‹œ๊ฐ„ ๋‹จ์œ„ ๋ฐ์ดํ„ฐ์˜ ์ง‘ํ•ฉ์œผ๋กœ์„œ 24์‹œ๊ฐ„์„ ์ฃผ๊ธฐ๋กœ ๋ฐ˜๋ณต๋˜๋Š” ๊ณ„์ ˆ์„ฑ(Seasonality) ํŒจํ„ด์„ ๋ฐ˜์˜ํ•œ Seasonal ARIMA ๋ชจ๋ธ์„ ์ ์šฉํ•œ๋‹ค. ๊ทธ๋ฆผ 3์€ ARIMA ๋ถ€ํ•˜์˜ˆ์ธก ๋ชจ๋ธ์„ ์ˆ˜๋ฆฝํ•˜๋Š” ์ ˆ์ฐจ๋ฅผ ๋‚˜ํƒ€๋‚ธ๋‹ค.

๊ทธ๋ฆผ 3. ARIMA ๋ถ€ํ•˜์˜ˆ์ธก ๋ชจ๋ธ ์ˆ˜๋ฆฝ ์ ˆ์ฐจ

Fig. 3. ARIMA modeling procedure for load forecast

../../Resources/kiee/KIEEP.2023.72.3.139/fig3.png

๊ทธ๋ฆผ 4๋Š” ๋‚˜์ฃผ๋ณ€์ „์†Œ ์‹œ๋‚ด ๋ฐฐ์ „์„ ๋กœ์— ๋Œ€ํ•˜์—ฌ 2023.1.5 ~ 2.4 ๊ธฐ๊ฐ„ ๋™์•ˆ ์ธก์ •ํ•œ ์‹ค์ œ ๋ถ€ํ•˜ ๋ฐ์ดํ„ฐ ์ค‘์—์„œ ์œ ํšจํ•œ ์‹œ๊ณ„์—ด ๋ฐ์ดํ„ฐ 17์ผ์น˜๋ฅผ ๋‚˜ํƒ€๋‚ธ๋‹ค.

๊ทธ๋ฆผ 4. ํ•ด๋‹น ๋ฐฐ์ „์„ ๋กœ์˜ ์‹œ๊ณ„์—ด ๋ถ€ํ•˜๋ฐ์ดํ„ฐ

Fig. 4. Time-series load data of the distribution line

../../Resources/kiee/KIEEP.2023.72.3.139/fig4.png

ํ†ต๊ณ„์ ์œผ๋กœ ์˜๋ฏธ์žˆ๋Š” ARIMA ๋ชจ๋ธ์ด ์ˆ˜๋ฆฝ๋˜๋ ค๋ฉด ์‹œ๊ณ„์—ด ๋ถ€ํ•˜ ๋ฐ์ดํ„ฐ๊ฐ€ ์ผ์ •ํ•œ ํ‰๊ท (mean)๊ณผ ์ผ์ •ํ•œ ๋ถ„์‚ฐ(variance)์„ ๊ฐ–๋Š” ์ •์ƒ์„ฑ(stationality)์„ ๋‚˜ํƒ€๋‚ด๋Š”์ง€ ์ž๊ธฐ์ƒ๊ด€์„ฑ(Auto-correlation Function, ACF) ๋ถ„์„๊ณผ ํŽธ์ž๊ธฐ์ƒ๊ด€์„ฑ(Partial Auto-correlation Function, PACF) ๋ถ„์„์„ ํ†ตํ•ด ํ™•์ธํ•œ๋‹ค. ๋งŒ์•ฝ ์ •์ƒ์„ฑ์ด ๋‚˜ํƒ€๋‚˜์ง€ ์•Š์„ ๋•Œ๋Š” ๋ฐ์ดํ„ฐ๋ฅผ ์ฐจ๋ถ„(differenciation)ํ•˜์—ฌ ๊ณ„์† ๋ถ„์„ํ•œ๋‹ค. 1ํšŒ ์ฐจ๋ถ„ํ•œ ์‹œ๊ณ„์—ด ๋ถ€ํ•˜๋ฐ์ดํ„ฐ๋ฅผ ์ด์šฉํ•˜์—ฌ ๊ทธ๋ฆผ 5์™€ ๊ฐ™์ด ACF ๋ฐ PACF ๋ถ„์„์„ ์‹ค์‹œํ•˜์˜€๋‹ค.

๊ทธ๋ฆผ 5. 1์ฐจ๋ถ„ ๋ฐ์ดํ„ฐ์˜ ์ž๊ธฐ์ƒ๊ด€์„ฑ ๋ฐ ํŽธ์ž๊ธฐ์ƒ๊ด€์„ฑ ๋ถ„์„

Fig. 5. ACF and PACF analysis for 1st defferenciated data

../../Resources/kiee/KIEEP.2023.72.3.139/fig5.png

AR, MA, SAR, SMA ๊ณ„์ˆ˜์˜ ์ฐจ์ˆ˜(order)๋ฅผ ์„ ์ •ํ•˜์—ฌ Seasonal ARIMA(1,1,1)ร—(1,1,1)24 ๋ชจ๋ธ์„ ํ™•์ •ํ•˜์˜€์œผ๋ฉฐ ์‹ (1)๊ณผ ๊ฐ™์ด ํ‘œํ˜„๋œ๋‹ค.

(1)
$ (1-\phi_{1}L)(1-\Phi_{24}L^{24})(1-L)(1-L^{24})y_{t}\\ =(1+\theta_{1}L)(1+\Theta_{24}L^{24})\epsilon_{t} $

์—ฌ๊ธฐ์„œ, $L$์€ 1์‹œ์  ์ง€์—ฐ ์—ฐ์‚ฐ์ž, $1-L$ ์€ 1ํšŒ ์ฐจ๋ถ„ ์—ฐ์‚ฐ์ž, $1-L^{24}$์€ 24์‹œ์  ๊ณ„์ ˆ์„ฑ ์—ฐ์‚ฐ์ž, $\phi_{1}$์€ AR ๊ณ„์ˆ˜, $\theta_{1}$์€ MA ๊ณ„์ˆ˜, $\Phi_{1}$์€ SAR ๊ณ„์ˆ˜, $\Theta_{1}$์€ SMA ๊ณ„์ˆ˜์ด๋‹ค.

MATLAB Statistics Toolbox๋ฅผ ์ด์šฉํ•˜์—ฌ ํ•ด๋‹น์„ ๋กœ์˜ ์‹œ๊ณ„์—ด ๋ถ€ํ•˜ ์ด๋ ฅ 12์ผ์น˜๋ฅผ ๊ธฐ์ค€์œผ๋กœ ์‹ (1)์˜ SARIMA ๋ชจ๋ธ์— ๋Œ€ํ•œ ๊ณ„์ˆ˜์ถ”์ •์„ ์‹ค์‹œํ•˜์˜€๋‹ค. ํ™•์ •๋œ ๋ชจ๋ธ์˜ ์˜ค์ฐจ๋Š” ๊ทธ๋ฆผ 6๊ณผ ๊ฐ™์ด ์ •๊ทœ๋ถ„ํฌ๋ฅผ ๋”ฐ๋ฅด๊ณ  ์ผ์ •๋ฒ”์œ„ ์ด๋‚ด๋กœ ์ˆ˜๋ ด๋˜๋ฏ€๋กœ ๋ชจ๋ธ์ด ํ†ต๊ณ„์ ์œผ๋กœ ์œ ํšจํ•จ์„ ๋‚˜ํƒ€๋‚ธ๋‹ค.

๊ทธ๋ฆผ 6. ARIMA ๋ชจ๋ธ์˜ ์œ ํšจ์„ฑ ๊ฒ€์ฆ

Fig. 6. Validation of the ARIMA model

../../Resources/kiee/KIEEP.2023.72.3.139/fig6.png

๋˜ํ•œ, ์ดํ›„ 3์ผ์น˜ ์‹ค๋ถ€ํ•˜์™€ ๋ชจ๋ธ์„ ํ†ตํ•œ ์˜ˆ์ธก๊ฐ’์„ ๋น„๊ตํ•œ ๊ฒฐ๊ณผ ๊ทธ๋ฆผ 7๊ณผ ๊ฐ™์ด ์˜ˆ์ธก์˜ค์ฐจ(Mean Average Percentage Error, MAPE) 8.48% ์ˆ˜์ค€์˜ ์„ฑ๋Šฅ์„ ํ™•๋ณดํ•˜์˜€๋‹ค.

๊ทธ๋ฆผ 7. ๋ถ€ํ•˜์˜ˆ์ธก ์„ฑ๋Šฅ (MAPE = 8.48)

Fig. 7. Load forecasting performance (MAPE = 8.48)

../../Resources/kiee/KIEEP.2023.72.3.139/fig7.png

4. ํ•˜๋ฃจ์ „ ESS ์ตœ์  ์Šค์ผ€์ค„๋ง

๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ํ•˜๋ฃจ์ „ ์˜ˆ์ธก๋ถ€ํ•˜๋ฅผ ํ™œ์šฉํ•˜์—ฌ ์ต์ผ ESS์˜ ์šด์ „๊ณ„ํš์„ ์ˆ˜๋ฆฝํ•˜๊ธฐ ์œ„ํ•˜์—ฌ ์ตœ์ ํ™” ๊ธฐ๋ฒ•(Optimization Method)์„ ๋„์ž…ํ•œ๋‹ค[9]. ์ตœ์ ํ™” ๊ธฐ๋ฒ•์€ ๊ตฌํ•˜๊ณ ์ž ํ•˜๋Š” ๋ณ€์ˆ˜๋“ค์˜ ๊ด€๊ณ„์‹์œผ๋กœ ๊ตฌ์„ฑ๋œ ๋ชฉ์ ํ•จ์ˆ˜(Objective Function)์™€ ์ œ์•ฝ์กฐ๊ฑด(Constraints)์„ ์„ค์ •ํ•˜๊ณ , ์ œ์•ฝ์กฐ๊ฑด์„ ๋งŒ์กฑํ•˜๋ฉด์„œ ๋ชฉ์ ํ•จ์ˆ˜์˜ ๊ฒฐ๊ณผ๊ฐ’์„ ์ตœ์†Œํ™”ํ•˜๋Š” ๋ณ€์ˆ˜๋“ค์˜ ํ•ด๋ฅผ ๊ตฌํ•˜๋Š” ๊ฒƒ์ด๋‹ค. ๊ทธ๋ฆผ 8์€ ํ•˜๋ฃจ์ „์— ์‹œํ–‰ํ•˜๋Š” ์ต์ผ ESS์˜ ์šด์ „ ๊ณ„ํš์„ ์ˆ˜๋ฆฝํ•˜๊ธฐ ์œ„ํ•œ ์ ˆ์ฐจ๋ฅผ ๋‚˜ํƒ€๋‚ธ๋‹ค.

๊ทธ๋ฆผ 8. ํ•˜๋ฃจ์ „ ๊ณ„ํšํ•˜๋Š” ์ต์ผ ESS ์ถฉ๋ฐฉ์ „ ์Šค์ผ€์ค„๋ง

Fig. 8. Day-ahead ESS scheduling for next day

../../Resources/kiee/KIEEP.2023.72.3.139/fig8.png

ESS์˜ ์ต์ผ ์ถฉ๋ฐฉ์ „ ๊ณ„ํš์€ ๋‘ ๊ฐ€์ง€๋กœ ๊ตฌ๋ถ„ํ•  ์ˆ˜ ์žˆ๋‹ค. ์ฒซ์งธ, ๋ฐฉ์ „์— ํ•„์š”ํ•œ ์—๋„ˆ์ง€๋ฅผ ํ™•๋ณดํ•˜๋Š” ์ถฉ์ „๊ณ„ํš์ด ํ•„์š”ํ•˜๋‹ค. ์ถฉ์ „์— ํ•„์š”ํ•œ ์ „๋ ฅ์€ ์ „์›์ธก์œผ๋กœ๋ถ€ํ„ฐ ์ถฉ๋‹น๋˜๋ฏ€๋กœ SMP๋ฅผ ์ ์šฉํ•˜๊ณ  ์ถฉ์ „์™„๋ฃŒ์‹œ์ ์€ ํ†ต์ƒ SMP๊ฐ€ ์ฆ๊ฐ€ํ•˜๊ธฐ ์‹œ์ž‘ํ•˜๋Š” ์‹œ์ ์ธ 8์‹œ๋กœ ์ •ํ•œ๋‹ค. ์ฆ‰ 0์‹œ๋ถ€ํ„ฐ 8์‹œ๊นŒ์ง€ ESS์˜ SOC๊ฐ€ ์ดˆ๊ธฐ๊ฐ’์œผ๋กœ๋ถ€ํ„ฐ SOC ์ตœ๋Œ€๊ฐ’๊นŒ์ง€ ์™„์ „ ์ถฉ์ „ํ•˜์—ฌ ์ดํ›„ ๊ณผ๋ถ€ํ•˜ ์‹œ์ ์—์„œ ๋ฐฉ์ „ํ•  ์ˆ˜ ์žˆ๋„๋ก ๊ด€๊ณ„์‹์„ ์„ค์ •ํ•œ๋‹ค.

๋‘˜์งธ, ๋ถ€ํ•˜ ์˜ˆ์ธก๊ฐ’์ด ์„ ๋กœ์˜ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜๋Š” ๊ฒฝ์šฐ ๊ณผ๋ถ€ํ•˜๋ฅผ ํ•ด์†Œํ•˜๊ธฐ ์œ„ํ•œ ๋ฐฉ์ „๊ณ„ํš์ด ํ•„์š”ํ•˜๋‹ค. ์ฆ‰ 9์‹œ๋ถ€ํ„ฐ 23์‹œ๊นŒ์ง€ ๋งค ์‹œ์ ๋ณ„๋กœ ๋ถ€ํ•˜ ์˜ˆ์ธก๊ฐ’๊ณผ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์˜ ์ฐจ์ด๋งŒํผ ๋ฐฉ์ „ํ•˜์—ฌ ๋ถ€ํ•˜๋ฅผ ์šด์ „์šฉ๋Ÿ‰ ์ด๋‚ด๋กœ ์œ ์ง€ํ•œ๋‹ค. ๋งŒ์•ฝ ์ต์ผ ๋ชจ๋“  ์‹œ์ ์—์„œ ๋ถ€ํ•˜ ์˜ˆ์ธก๊ฐ’์ด ์„ ๋กœ์˜ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜์ง€ ์•Š๋Š”๋‹ค๋ฉด ๋ฐฉ์ „๊ณ„ํš์€ ํ•„์š”์น˜ ์•Š์œผ๋ฏ€๋กœ ๊ทธ ๋‹ค์Œ๋‚ ์˜ ๋ฐฉ์ „๊ณ„ํš ์ˆ˜๋ฆฝ ์‹œ์ ๊นŒ์ง€ ๋Œ€๊ธฐํ•œ๋‹ค.

ESS์˜ ์ต์ผ ์ถฉ๋ฐฉ์ „ ๊ณ„ํš ์ˆ˜๋ฆฝ์„ ์œ„ํ•œ ์ตœ์ ํ™” ๋ชฉ์ ํ•จ์ˆ˜๋Š” ์‹ (2)~(3)๊ณผ ๊ฐ™๋‹ค. ์‹ (2)๋Š” $k = 0,\: \cdots ,\: 8$ ์‹œ์ ์—์„œ ์ถฉ์ „๊ณ„ํš ์ˆ˜๋ฆฝ์„ ์œ„ํ•œ ๋ชฉ์ ํ•จ์ˆ˜์ด๋‹ค. ๋ชฉ์ ํ•จ์ˆ˜์˜ ์ฒซ์งธํ•ญ์€ ๊ฐ ์‹œ์ ๋ณ„ ์ถฉ์ „์š”๊ธˆ์— ์ƒ์‘ํ•˜๋Š” ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ์˜ ์ตœ์†Œํ™”๋ฅผ ์˜๋ฏธํ•˜๊ณ , ๋‘˜์งธํ•ญ์€ ESS์˜ ์ถฉ๋ฐฉ์ „์ „๋ ฅ ์ตœ์†Œํ™”๋ฅผ ์˜๋ฏธํ•˜๋ฉฐ, ์…‹์งธํ•ญ์€ ์—๋„ˆ์ง€์ €์žฅ์žฅ์น˜๊ฐ€ ์ตœ๋Œ€ ์ถฉ์ „๋  ์ˆ˜ ์žˆ๋„๋ก k ์‹œ์  SOC์™€ SOC ์ตœ๋Œ€๊ฐ’ ์‚ฌ์ด์˜ ์˜ค์ฐจ ์ตœ์†Œํ™”๋ฅผ ์˜๋ฏธํ•œ๋‹ค. ๊ฐ ํ•ญ์— ๊ฐ€์ค‘์น˜๋ฅผ ๋‘์–ด ๋ฐ˜์‘๋„๋ฅผ ์กฐ์ ˆํ•  ์ˆ˜ ์žˆ๋‹ค. ์‹ (3)์€ ๋ฐฉ์ „๊ณ„ํš ์ˆ˜๋ฆฝ์„ ์œ„ํ•œ $k = 9,\: \cdots ,\: 23$ ์‹œ์ ์—์„œ์˜ ๋ชฉ์ ํ•จ์ˆ˜์ด๊ณ , ESS์˜ ์ถฉ๋ฐฉ์ „์ „๋ ฅ ์ตœ์†Œํ™”๋ฅผ ์˜๋ฏธํ•œ๋‹ค. ๊ฒฐ๊ตญ, ESS์˜ ์ต์ผ ์ถฉ๋ฐฉ์ „ ๊ณ„ํš ์ตœ์ ํ™” ๋ฌธ์ œ๋Š” ์ต์ผ์˜ ๋ชจ๋“  ์‹œ์  $k =[0,\: 1,\: \cdots ,\: 23]$์—์„œ ๋ชฉ์ ํ•จ์ˆ˜ $F$๋ฅผ ์ตœ์†Œํ™”ํ•˜๋˜ ์ œ์•ฝ์กฐ๊ฑด์„ ๋งŒ์กฑํ•˜๋Š” ๋ณ€์ˆ˜ $x=[P_{k}^{g},\: P_{k}^{e},\: SOC_{k}^{e}]$๋ฅผ ๊ตฌํ•˜๋Š” ๋ฌธ์ œ๋กœ ๊ท€๊ฒฐ๋œ๋‹ค.

(2)
$$ F=\min _x \sum_{k=0}^8\left\{\omega_1 \gamma_k P_k^g+\omega_2\left(P_k^e\right)^2+\omega_3\left(S O C_{\max }^e-S O C_k^e\right)^2\right\} $$
(3)
$$ F=\min _x \sum_{k=9}^{23}\left(P_k^e\right)^2 $$

์—ฌ๊ธฐ์„œ, $P_{k}^{g}$๋Š” k ์‹œ์  ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ[kW], $P_{k}^{e}$๋Š” k ์‹œ์  ESS ์ถฉ๋ฐฉ์ „์ „๋ ฅ(-: ์ถฉ์ „, +: ๋ฐฉ์ „)[kW], $SOC_{k}^{e}$๋Š” k ์‹œ์  ESS SOC[%], $\gamma_{k}$๋Š” k ์‹œ์  SMP[์›], $\omega_{1},\: \omega_{2},\: \omega_{3}$๋Š” ๊ฐ€์ค‘์น˜์ด๋‹ค.

๋˜ํ•œ, ๊ตฌํ•˜๊ณ ์ž ํ•˜๋Š” ๋ณ€์ˆ˜ $x=[P_{k}^{g},\: P_{k}^{e},\: SOC_{k}^{e}]$์˜ ์ œ์•ฝ์กฐ๊ฑด์€ ์‹ (4)~(7)๊ณผ ๊ฐ™๋‹ค.

(4)
$P_{k}^{g}+ P_{k}^{e}= P_{k}^{l}$
(5)
$-P_{k}^{e}\le\dfrac{E_{\max}^{e}\delta_{k}}{100\eta}(SOC_{\max }^{e}- SOC_{k-1}^{e})$
(6)
$P_{k}^{e}\le -\dfrac{E_{\max }^{e}\delta_{k}}{100\eta}(SOC_{"\min "}^{e}- SOC_{k-1}^{e})$
(7)
$ \begin{align*} SOC_{k}^{e}& =SOC_{k-1}^{e}- P_{k}^{e}\dfrac{100\eta}{E_{\max }^{e}\delta_{k}} \end{align*} $

์—ฌ๊ธฐ์„œ, $P_{k}^{l}$๋Š” k ์‹œ์  ์„ ๋กœ๋ถ€ํ•˜[kW], $\eta$๋Š” ESS ์ถฉ๋ฐฉ์ „ ํšจ์œจ[pu], $E_{\max }^{e}$๋Š” ESS ์ •๊ฒฉ์šฉ๋Ÿ‰[kWh], $\delta_{k}$๋Š” ์‹œ์ ๊ฐ„ ์‹œ๊ฐ„๊ฐ„๊ฒฉ[h] ์ด๋‹ค. ์‹ (4)๋Š” ์ „๋ ฅ์ˆ˜๊ธ‰ ๊ท ํ˜• ์ œ์•ฝ์กฐ๊ฑด์œผ๋กœ์„œ ํ•ด๋‹น์„ ๋กœ์˜ ์ด๋ถ€ํ•˜๋Š” ์ „์›์ธก์œผ๋กœ๋ถ€ํ„ฐ ์„ ๋กœ๋กœ ๊ณต๊ธ‰๋˜๋Š” ์ „๋ ฅ๊ณผ ESS์—์„œ ๋ฐฉ์ „๋˜๋Š” ์ „๋ ฅ์˜ ํ•ฉ์œผ๋กœ ํ‘œํ˜„๋œ๋‹ค. ์‹ (5)๋Š” ESS์˜ ์ตœ๋Œ€ ์ถฉ์ „๊ฐ€๋Šฅ์ „๋ ฅ ์ œ์•ฝ์กฐ๊ฑด์œผ๋กœ์„œ ์ตœ๋Œ€ SOC์—์„œ ํ˜„์žฌ ์‹œ์ ์˜ SOC์˜ ์ฐจ๋กœ ํ‘œํ˜„๋œ๋‹ค. ์‹ (6)์€ ESS์˜ ์ตœ๋Œ€ ๋ฐฉ์ „๊ฐ€๋Šฅ์ „๋ ฅ ์ œ์•ฝ์กฐ๊ฑด์œผ๋กœ์„œ ์ตœ์†Œ SOC์—์„œ ํ˜„์žฌ ์‹œ์ ์˜ SOC์˜ ์ฐจ๋กœ ํ‘œํ˜„๋œ๋‹ค. ์‹ (7)์€ ESS์˜ ๋ฐฉ์ „์ „๋ ฅ๊ณผ SOC์˜ ์ƒ๊ด€์„ฑ ์ œ์•ฝ์กฐ๊ฑด์œผ๋กœ์„œ k ์‹œ์  ๋ฐฉ์ „์ „๋ ฅ์€ k ์‹œ์  SOC์—์„œ k-1 ์‹œ์  SOC์˜ ์ฐจ๋กœ ํ‘œํ˜„๋œ๋‹ค.

๋˜ํ•œ, ๊ตฌํ•˜๊ณ ์ž ํ•˜๋Š” ๋ณ€์ˆ˜ $x=[P_{k}^{g},\: P_{k}^{e},\: SOC_{k}^{e}]$์˜ ๋ฒ”์œ„์— ๊ด€ํ•œ ์ œ์•ฝ์กฐ๊ฑด์€ ์‹ (8)~(11)๊ณผ ๊ฐ™๋‹ค.

(8)
$-P_{limit}\le P_{k}^{g}\le P_{limit}$
(9)
$-P_{\max }^{e}\le P_{k}^{e}\le P_{\max }^{e}$
(10)
$SOC_{"\min "}^{e}\le SOC_{k}^{e}\le SOC_{"\min "}^{e}$
(11)
$SOC_{0}^{e}=SOC_{initial}$

์—ฌ๊ธฐ์„œ, $P_{limit}$๋Š” ์„ ๋กœ์˜ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰[kW], $SOC_{0}^{e}$๋Š” SOC ์ดˆ๊ธฐ๊ฐ’[%]์ด๋‹ค. ์‹ (8)์€ ์ „์›์ธก์œผ๋กœ๋ถ€ํ„ฐ ์„ ๋กœ๋กœ ๊ณต๊ธ‰๋˜๋Š” ์ „๋ ฅ์€ ์„ ๋กœ์˜ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰ ์ด๋‚ด๋กœ ์ œ์•ฝ๋จ์„ ๋‚˜ํƒ€๋‚ธ๋‹ค. ์‹ (9)๋Š” ESS์˜ ์ถฉ๋ฐฉ์ „ ์ „๋ ฅ์€ PCS์˜ ์ •๊ฒฉ์ถœ๋ ฅ ์ด๋‚ด๋กœ ์ œ์•ฝ๋จ์„ ๋‚˜ํƒ€๋‚ธ๋‹ค. ์‹ (10)์€ ESS์˜ SOC๋Š” ๋ฐฉ์ „์‹ฌ๋„ ์ด๋‚ด๋กœ ์ œ์•ฝ๋จ์„ ๋‚˜ํƒ€๋‚ธ๋‹ค. ์‹ (11)์˜ SOC ์ดˆ๊ธฐ๊ฐ’์€ ์ด์ „ ์‹œ์ ์— ๋‚จ์•„ ์žˆ๋Š” SOC๊ฐ’์„ ์˜๋ฏธํ•œ๋‹ค.

์‹ (2)~(11)์— ๋”ฐ๋ฅธ ์ตœ์ ํ™” ๋ฌธ์ œ๋Š” ๋ชฉ์ ํ•จ์ˆ˜์— ์ œ๊ณฑํ•ญ์ด ์กด์žฌํ•˜๋Š” ์ „ํ˜•์ ์ธ Quadratic Programing ์ตœ์ ํ™” ๋ฌธ์ œ์ด๋ฉฐ MATLAB Optimization Toolbox๋ฅผ ์ด์šฉํ•˜์—ฌ ๊ตฌํ•  ์ˆ˜ ์žˆ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜, ARIMA ๋ชจ๋ธ์„ ํ†ตํ•œ ๋ฐฐ์ „์„ ๋กœ์˜ ์ต์ผ ๋ถ€ํ•˜ ์˜ˆ์ธก๊ฐ’์€ ์•ฝ 10% ์ด๋‚ด์˜ ์˜ค์ฐจ๋ฅผ ํฌํ•จํ•˜๊ณ  ์žˆ๋‹ค. ๋”ฐ๋ผ์„œ, ํ•˜๋ฃจ์ „์— ์ˆ˜๋ฆฝํ•œ ESS์˜ ์ถฉ๋ฐฉ์ „ ๊ณ„ํš์€ ์ต์ผ ์‹ค์ œ ๋ถ€ํ•˜์˜ ์ฆ๊ฐ€์— ๋”ฐ๋ผ ์„ ๋กœ ์šด์ „์šฉ๋Ÿ‰์„ ์œ ์ง€ํ•˜๋Š” ๋ฐ ์ถฉ๋ถ„ํ•˜์ง€ ์•Š์„ ์ˆ˜ ์žˆ์œผ๋ฏ€๋กœ ๋ถ€ํ•˜์˜ˆ์ธก ์˜ค์ฐจ๋ฅผ ๊ณ ๋ คํ•œ ์‹œ์ ๋ณ„ ์ถฉ๋ฐฉ์ „ ๊ณ„ํš์„ ๋ณด์™„ํ•  ํ•„์š”๊ฐ€ ์žˆ๋‹ค[10].

5. ๋‹น์ผ ESS ์‹ค์‹œ๊ฐ„ ๋ณด์ƒ

๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ์ต์ผ ๋งค ์‹œ์ ๋ณ„ ์‹ค์ œ ๋ถ€ํ•˜์˜ ์ธก์ •๊ฐ’๊ณผ ํ•˜๋ฃจ์ „ ์˜ˆ์ธก ๋ถ€ํ•˜์˜ ์ฐจ์ด๋ฅผ ๊ณ ๋ คํ•˜์—ฌ ํ•˜๋ฃจ์ „ ์ˆ˜๋ฆฝํ•œ ์ถฉ๋ฐฉ์ „ ๊ณ„ํš์„ ๋ณด์ƒํ•˜์—ฌ ๋ถ€ํ•˜๋ฅผ ์„ ๋กœ ์šด์ „์šฉ๋Ÿ‰ ์ด๋‚ด๋กœ ํ™•์‹คํžˆ ์กฐ์ •ํ•  ์ˆ˜ ์žˆ๋„๋ก ์‹ค์‹œ๊ฐ„ ์ถฉ๋ฐฉ์ „ ๋ณด์ƒ ๊ธฐ๋ฒ•์„ ์ œ์•ˆํ•œ๋‹ค.

๊ทธ๋ฆผ 9๋Š” ์‹ค์‹œ๊ฐ„ ์ถฉ๋ฐฉ์ „ ๋ณด์ƒ์„ ์œ„ํ•œ ์ˆœ์„œ๋„์ด๋‹ค. ๋จผ์ €, ์‹ค์‹œ๊ฐ„ ๋ณด์ƒ์„ ์œ„ํ•˜์—ฌ ๋‹น์ผ ๋งค ์‹œ์ ์—์„œ ๋ฐฐ์ „์„ ๋กœ์˜ ๋ถ€ํ•˜๋ฅผ ์‹ค์ธกํ•˜์—ฌ ํ•˜๋ฃจ์ „ ๋ถ€ํ•˜ ์˜ˆ์ธก๊ฐ’ ๋Œ€๋น„ ์˜ค์ฐจ๋ฅผ ํŒŒ์•…ํ•˜์—ฌ์•ผ ํ•œ๋‹ค. ๋ถ€ํ•˜์„ ๋กœ์˜ ๊ฒฝ์šฐ ๋ฐฐ์ „์„ ๋กœ์˜ ์ด๋ถ€ํ•˜๋Š” ์†์‹ค์„ ๊ณ ๋ คํ•˜์ง€ ์•Š๋Š”๋‹ค๋ฉด ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ๊ณผ ๋™์ผํ•˜๋‹ค.

๊ทธ๋ฆผ 9. ์‹ค์‹œ๊ฐ„ ์ถฉ๋ฐฉ์ „ ๋ณด์ƒ

Fig. 9. Real-time charging/discharging compensation

../../Resources/kiee/KIEEP.2023.72.3.139/fig9.png

๋”ฐ๋ผ์„œ, ์‹ค๋ฌด์ ์œผ๋กœ ๋ฐฐ์ „์„ ๋กœ์˜ ๋ถ€ํ•˜๋ฅผ ์‹ค์ธกํ•˜๊ธฐ ์œ„ํ•ด, ์ต์ผ ๋งค ์‹œ์ ์—์„œ ์„ ๋กœ๋กœ ์œ ์ž…๋˜๋Š” ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ์„ ์ธก์ •ํ•œ๋‹ค. ์ต์ผ ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ์ธก์ •๊ฐ’์ด ํ•˜๋ฃจ์ „ ๋งˆ๋ จํ•ด๋‘” ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ๊ณ„ํš๊ฐ’๋ณด๋‹ค ํฌ๋ฉด ์„ ๋กœ์˜ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•  ๊ฐ€๋Šฅ์„ฑ์ด ์žˆ๊ธฐ ๋•Œ๋ฌธ์— ์‹ (12)์™€ ๊ฐ™์ด ESS์˜ ๋ฐฉ์ „์ „๋ ฅ ์ž ์ •๋ณด์ƒ๊ฐ’์€ ์ต์ผ ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ์ธก์ •๊ฐ’์˜ ์šด์ „์šฉ๋Ÿ‰ ์ดˆ๊ณผ๋ถ„ ์ „๋ ฅ๊ณผ ํ•˜๋ฃจ์ „ ESS์˜ ๋ฐฉ์ „์ „๋ ฅ ๊ณ„ํš๊ฐ’ ์ค‘ ์ตœ๋Œ€๊ฐ’์„ ์„ ํƒํ•˜์—ฌ ์กฐ์ •ํ•œ๋‹ค.

๋ฐ˜๋ฉด์— ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ์ธก์ •๊ฐ’์ด ํ•˜๋ฃจ์ „ ๋งˆ๋ จํ•ด๋‘” ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ๊ณ„ํš๊ฐ’๋ณด๋‹ค ์ž‘๊ณ , ESS๊ฐ€ ๋ฐฉ์ „(+) ์ƒํƒœ์ธ ๊ฒฝ์šฐ, ๋ฐฐ์ „์„ ๋กœ์˜ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•  ๊ฐ€๋Šฅ์„ฑ์ด ์—†๊ธฐ ๋•Œ๋ฌธ์— ์‹ (13)์˜ ์ƒ๋‹จ์‹๊ณผ ๊ฐ™์ด ESS์˜ ๋ฐฉ์ „์ „๋ ฅ ์ž ์ •๋ณด์ƒ๊ฐ’์€ ์ต์ผ ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ์ธก์ •๊ฐ’์˜ ์šด์ „์šฉ๋Ÿ‰ ์ดˆ๊ณผ๋ถ„ ์ „๋ ฅ๊ณผ ํ•˜๋ฃจ์ „ ESS์˜ ๋ฐฉ์ „์ „๋ ฅ ๊ณ„ํš๊ฐ’ ์ค‘ ์ตœ์†Œ๊ฐ’์„ ์„ ํƒํ•œ๋‹ค. ๋‹จ, ๋ฐฉ์ „์ƒํƒœ๋ฅผ ์ถฉ์ „์ƒํƒœ๋กœ ๋ณ€๊ฒฝํ•˜์ง€ ์•Š๋„๋ก 0๋ณด๋‹ค ํฐ๊ฐ’์„ ์„ ํƒํ•œ๋‹ค. ๋˜ํ•œ, ESS๊ฐ€ ์ถฉ์ „(-) ์ƒํƒœ์ธ ๊ฒฝ์šฐ, ์‹ (13)์˜ ํ•˜๋‹จ์‹๊ณผ ๊ฐ™์ด ESS์˜ ์ถฉ์ „์ „๋ ฅ ์ž ์ •๋ณด์ƒ๊ฐ’์€ ์ต์ผ ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ์˜ ์šด์ „์šฉ๋Ÿ‰ ์ดˆ๊ณผ๋ถ„ ์ „๋ ฅ๊ณผ ํ•˜๋ฃจ์ „ ESS์˜ ์ถฉ์ „์ „๋ ฅ ๊ณ„ํš๊ฐ’ ์ค‘ ์ตœ์†Œ๊ฐ’์„ ์„ ํƒํ•œ๋‹ค. ๋‹จ, ์ถฉ์ „์ƒํƒœ๋ฅผ ๋ฐฉ์ „์ƒํƒœ๋กœ ๋ณ€๊ฒฝํ•˜์ง€ ์•Š๋„๋ก 0๋ณด๋‹ค ์ž‘์€๊ฐ’์„ ์„ ํƒํ•œ๋‹ค. ESS์˜ ๋ฐฉ์ „์ „๋ ฅ ์ตœ์ข…๋ณด์ƒ๊ฐ’์€ ์‹ (15)์™€ ๊ฐ™์ด ESS์˜ ์ž ์ •๋ณด์ƒ๊ฐ’, ์ •๊ฒฉ์ถœ๋ ฅ ๋ฐ ์‹ (14)์˜ ๋ฐฉ์ „ํ•˜ํ•œ ์ œ์•ฝ๊ฐ’ ์ค‘ ์ตœ์†Œ๊ฐ’์œผ๋กœ ๊ฒฐ์ •๋œ๋‹ค.

(12)
$\widetilde{P}_{k}^{e}=\max[(\check{P}_{k}^{g}-P_{limit})/\eta ,\: \hat{P}_{k}^{e}],\: {if}\check{{P}}_{{k}}^{{g}}\ge\hat{{P}}_{{k}}^{{g}}$
(13)
$\widetilde{P}_{k}^{e}=\begin{cases} \max[0,\: \min[(\check{P}_{k}^{g}-P_{limit})/\eta ,\: \hat{P}_{k}^{e}]],\: {if}\check{{P}}_{{k}}^{{g}}<\hat{{P}}_{{k}}^{{g}}{WED}\ge \hat{{P}}_{{k}}^{{e}}\ge 0\\ \min[0,\: \max[(\check{{P}}_{{k}}^{{g}}-{P}_{limit})/\eta ,\: \hat{{P}}_{{k}}^{{e}}]],\: {if}\check{{P}}_{{k}}^{{g}}<\hat{{P}}_{{k}}^{{g}}{WED}\ge \hat{{P}}_{{k}}^{{e}}< 0 \end{cases}$
(14)
$P_{k}^{e_{lb}}=\dfrac{E_{\max }^{e}\delta_{k}}{100\eta}(SOC_{k-1}^{e}-SOC_{"\min "}^{e})$
(15)
$P_{k}^{e}=\min[P_{k}^{e_{lb}},\: P_{\max}^{e},\: \widetilde{P}_{k}^{e}],\: {if}\widetilde{{P}}_{{k}}^{{e}}\ge 0$

์—ฌ๊ธฐ์„œ, $\check{P}_{k}^{g}$๋Š” k์‹œ์  ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ์ธก์ •๊ฐ’[kW], $\hat{P}_{k}^{g}$๋Š” k์‹œ์  ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ๊ณ„ํš๊ฐ’[kW], $\hat{P}_{k}^{e}$๋Š” k์‹œ์  ESS ์ถฉ๋ฐฉ์ „์ „๋ ฅ ๊ณ„ํš๊ฐ’[kW], $\widetilde{P}_{k}^{e}$๋Š” k์‹œ์  ESS ์ถฉ๋ฐฉ์ „์ „๋ ฅ ์ž ์ •๋ณด์ƒ๊ฐ’[kW], $P_{k}^{e_{lb}}$๋Š” k์‹œ์  ESS ์ถฉ๋ฐฉ์ „์ „๋ ฅ ํ•˜ํ•œ์ œ์•ฝ๊ฐ’[kW], $P_{\max }^{e}$๋Š” ESS ์ •๊ฒฉ ์ถœ๋ ฅ์ „๋ ฅ[kW], $P_{k}^{e}$๋Š” k์‹œ์  ESS ์ถฉ๋ฐฉ์ „์ „๋ ฅ ์ตœ์ข…๋ณด์ƒ๊ฐ’[kW] ์ด๋‹ค.

6. ์‚ฌ๋ก€์—ฐ๊ตฌ

์‚ฌ๋ก€์—ฐ๊ตฌ์— ์ ์šฉํ•  ESS๋ฅผ ๊ทธ๋ฆผ 10๊ณผ ๊ฐ™์ด ๋‚˜์ฃผ๋ณ€์ „์†Œ ์‹œ๋‚ด ๋ฐฐ์ „์„ ๋กœ์— ๋ฐฐ์น˜ํ•˜์˜€๋‹ค. ์‚ฌ๋ก€์—ฐ๊ตฌ์— ์‚ฌ์šฉ๋œ ESS์™€ ๋ฐฐ์ „์„ ๋กœ์˜ ์ฃผ์š” ์šด์ „๋ณ€์ˆ˜๋Š” ํ‘œ 1๊ณผ ๊ฐ™์ด ์ •๋ฆฌํ•˜์˜€๋‹ค. 3์žฅ ์‹ (1)์˜ Seasonal ARIMA(1,1,1)ร—(1,1,1)24 ๋ชจ๋ธ์„ ๋ฐ”ํƒ•์œผ๋กœ ํ•ด๋‹น์„ ๋กœ์˜ ๋ถ€ํ•˜์˜ˆ์ธก์„ ์‹ค์‹œํ•˜๊ณ , 4์žฅ์˜ ํ•˜๋ฃจ์ „ ESS ์ตœ์  ์ถฉ๋ฐฉ์ „ ์Šค์ผ€์ค„๋ง ๋ฐฉ๋ฒ•๊ณผ 5์žฅ์˜ ์ต์ผ ์‹ค์‹œ๊ฐ„ ๋ณด์ƒ ๋ฐฉ๋ฒ•์„ ๋ฐ”ํƒ•์œผ๋กœ ESS๋ฅผ ๋ชจ์˜์šด์˜ํ•˜๊ณ  ๊ทธ ๊ฒฐ๊ณผ๋ฅผ ๋ถ„์„ํ•˜์˜€๋‹ค.

๊ทธ๋ฆผ 10. ํ•ด๋‹น์„ ๋กœ์˜ ๊ฐ„๋žต ๋‹จ์„ ๋„

Fig. 10. Single line diagram of the distribution line

../../Resources/kiee/KIEEP.2023.72.3.139/fig10.png

ํ‘œ 1 ๋ฐฐ์ „์„ ๋กœ ๋ฐ ESS์˜ ์šด์ „๋ณ€์ˆ˜

Table 1 The operation parameters of D/L and ESS

Parameters

Values

Etc.

1. D/L

- Operation limit

7,700 [kW]

(setpoint)

- Load condition

D-1 forecast < D-0 measure

'23.1.27 load

- Peak time slot

10, 11, 12 [h]

2. ESS

- Rated capacity

1,400 [kWh]

- Depth of charge

80 [%]

10~90 [%]

- Rated power

650 [kW]

- Efficiency

0.9 [p.u]

- Location

SW2 backward

์‚ฌ๋ก€์—ฐ๊ตฌ๋Š” ๊ทธ๋ฆผ 11๊ณผ ๊ฐ™์ด ํ•˜๋ฃจ์ „ ์˜ˆ์ธก๋ถ€ํ•˜($P_{F}^{l}$)๊ฐ€ ์ต์ผ ์‹ค์ธก๋ถ€ํ•˜($P_{M}^{l}$)๋ณด๋‹ค ์ž‘๊ฒŒ ์˜ˆ์ธก๋œ ๊ฒฝ์šฐ์ด๋‹ค. ์ฆ‰, ํ•˜๋ฃจ์ „ ์˜ˆ์ธก๋ถ€ํ•˜๊ฐ€ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰($P_{limit}$)์„ ์ดˆ๊ณผํ•˜๋Š” ๊ณผ๋ถ€ํ•˜ ๋ฐœ์ƒ๋Ÿ‰์€ ์ต์ผ ์‹ค์ธก๋ถ€ํ•˜๊ฐ€ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜๋Š” ๊ณผ๋ถ€ํ•˜ ๋ฐœ์ƒ๋Ÿ‰ ๋ณด๋‹ค ์ž‘์€ ๊ฒฝ์šฐ์ด๋‹ค.

๊ทธ๋ฆผ 11. ์‚ฌ๋ก€ (์ „์ผ ์˜ˆ์ธก๋ถ€ํ•˜ < ๋‹น์ผ ์‹ค์ œ๋ถ€ํ•˜)

Fig. 11. Case (D-1 load forecast < D-0 load meaesure)

../../Resources/kiee/KIEEP.2023.72.3.139/fig11.png

์ต์ผ ์šด์ „๊ณ„ํš์€ ํ•˜๋ฃจ์ „ ์˜ˆ์ธก๋ถ€ํ•˜๋ฅผ ๊ธฐ์ค€์œผ๋กœ ์‚ฐ์ •๋œ๋‹ค. ํ•˜๋ฃจ์ „ ์˜ˆ์ธก๋ถ€ํ•˜๋ฅผ ์‹ (2)~(11)์˜ ๋ชฉ์ ํ•จ์ˆ˜, ์ œ์•ฝ์กฐ๊ฑด ๋ฐ ๋ฒ”์œ„์กฐ๊ฑด์— ๋”ฐ๋ฅธ ESS ์ตœ์  ์Šค์ผ€์ค„๋ง ์‚ฐ์ •์‹์— ๋Œ€์ž…ํ•˜์—ฌ ์ต์ผ ์‹œ๊ฐ„๋Œ€๋ณ„๋กœ ESS์˜ ์ถฉ์ „ ๋˜๋Š” ๋ฐฉ์ „ ๊ณ„ํš($P_{k}^{e}$)๊ณผ ๊ทธ์— ๋”ฐ๋ผ ๊ฒฐ์ •๋˜๋Š” SOC ๊ณ„ํš($SOC_{k}^{e}$) ๋ฐ ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ ๊ณ„ํš($P_{k}^{g}$)์„ ์‚ฐ์ •ํ•œ๋‹ค. ESS๋Š” ๊ทธ๋ฆผ 12(a)์™€ ๊ฐ™์ด 0~8 ์‹œ๊ฐ„๋Œ€์—์„œ ์ถฉ์ „๋™์ž‘์„ ์ˆ˜ํ–‰ํ•˜๊ณ  10~12 ์‹œ๊ฐ„๋Œ€์—์„œ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜๋Š” ํ”ผํฌ๋ถ€ํ•˜๋ฅผ ์ €๊ฐํ•˜๋Š” ๋ฐฉ์ „๋™์ž‘์„ ์ˆ˜ํ–‰ํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๊ณ„ํš๋˜์—ˆ๋‹ค. ์ถฉ์ „๊ณ„ํš($P_{k}^{e}$<0)์€ ์ „๋ ฅ๊ฑฐ๋ž˜์†Œ๊ฐ€ ์ œ๊ณตํ•˜๋Š” ๊ทธ๋ฆผ 12(b)์˜ ์ต์ผ ์‹œ๊ฐ„๋Œ€๋ณ„ SMP๋ฅผ ๊ณ ๋ คํ•œ๋‹ค. SMP๋Š” 0~8 ์‹œ๊ฐ„๋Œ€์—์„œ ๋Œ€์ฒด์ ์œผ๋กœ ์ผ์ •ํ•œ ํŒจํ„ด์„ ๋‚˜ํƒ€๋‚ด๊ณ  ์žˆ๊ธฐ ๋•Œ๋ฌธ์— ์ถฉ์ „๊ณ„ํš๋„ ์ผ์ •ํ•˜๊ฒŒ ์‚ฐ์ •๋˜์—ˆ๊ณ  ์ด์— ๋”ฐ๋ผ SOC ๊ณ„ํš($SOC_{k}^{e}$)์€ 0์‹œ์  ์ดˆ๊ธฐ๊ฐ’ 50%์—์„œ 8์‹œ์  90%๊นŒ์ง€ ์™„์ „ ์ถฉ์ „๋˜๋„๋ก ์‚ฐ์ •๋˜์—ˆ๋‹ค. ๋˜ํ•œ, ๋ฐฉ์ „๊ณ„ํš($P_{k}^{e}$>0)์€ ์˜ˆ์ธก๋ถ€ํ•˜๊ฐ€ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜๋Š” 10~12 ์‹œ๊ฐ„๋Œ€์—์„œ ํ”ผํฌํ•ด์†Œ๋ฅผ ์œ„ํ•œ ๋ฐฉ์ „์„ ์‹œํ–‰ํ•˜์—ฌ SOC ๊ณ„ํš($SOC_{k}^{e}$)์€ 9์‹œ์  90%์—์„œ 12์‹œ์ ์—์„œ 73%๊นŒ์ง€ ์†Œ์ง„๋˜๊ณ  ์ดํ›„ ๋Œ€๊ธฐ์ƒํƒœ๋ฅผ ์œ ์ง€ํ•˜๋Š” ๊ฒƒ์œผ๋กœ ์‚ฐ์ •๋˜์—ˆ๋‹ค. ์ต์ผ ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ($P_{k}^{g}$)์€ ๊ทธ๋ฆผ 12(c)์™€ ๊ฐ™์ด 0~8 ์‹œ๊ฐ„๋Œ€์—์„œ ESS์˜ ์ถฉ์ „์ „๋ ฅ์ด ๋ถ€๊ฐ€๋˜๋ฏ€๋กœ ์†Œํญ ์ƒ์Šนํ•˜๊ณ  0~12 ์‹œ๊ฐ„๋Œ€์—์„œ ํ”ผํฌ๋ถ€ํ•˜ ์ €๊ฐ์„ ์œ„ํ•ด ๋ฐฉ์ „์ „๋ ฅ์ด ์ฐจ๊ฐ๋˜๋ฏ€๋กœ ์†Œํญ ๊ฐ์†Œํ•˜์—ฌ ํ•˜๋ฃจ์ „ ์˜ˆ์ธก๋ถ€ํ•˜๋ฅผ ๊ธฐ์ค€์œผ๋กœ ๊ณผ๋ถ€ํ•˜๊ฐ€ ํ•ด์†Œ๋˜๋Š” ๊ฒƒ์œผ๋กœ ๊ณ„ํš๋˜์—ˆ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜, ๊ทธ๋ฆผ 12(d)์™€ ๊ฐ™์ด ์ต์ผ ์‹ค์ธก๋ถ€ํ•˜๋Š” ํ•˜๋ฃจ์ „ ์˜ˆ์ธก๋ถ€ํ•˜๋ณด๋‹ค ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์„ ํ›จ์”ฌ ์ดˆ๊ณผํ•˜๊ธฐ ๋•Œ๋ฌธ์— ์ต์ผ ์‹ค์ธก๋ถ€ํ•˜๋ฅผ ๊ธฐ์ค€์œผ๋กœ ๊ณผ๋ถ€ํ•˜๊ฐ€ ์ถฉ๋ถ„ํžˆ ํ•ด์†Œ๋˜์ง€ ์•Š๋Š”๋‹ค. ๋”ฐ๋ผ์„œ, ๋ฐฉ์ „๊ณ„ํš์€ ์‹ค์งˆ์ ์œผ๋กœ ๋” ํฌ๊ฒŒ ์‚ฐ์ •๋˜์–ด์•ผ ํ•˜๋ฉฐ ์ด๋ฅผ ๋ณด์ถฉํ•˜๊ธฐ ์œ„ํ•ด ์ต์ผ ์‹ค์‹œ๊ฐ„ ๋ณด์ƒ ๋™์ž‘์ด ํ•„์š”ํ•œ ์ƒํ™ฉ์ด๋‹ค.

๊ทธ๋ฆผ 12. ํ•˜๋ฃจ์ „ ESS ์ถฉ๋ฐฉ์ „ ๊ณ„ํš (์‚ฌ๋ก€)

Fig. 12. Day-ahed ESS charge/discharge schedule (Case)

../../Resources/kiee/KIEEP.2023.72.3.139/fig12.png

๊ทธ๋ฆผ 13(a)์™€ ๊ฐ™์ด ์ต์ผ 10~12 ์‹œ๊ฐ„๋Œ€์— ์„ ๋กœ๋กœ ์œ ์ž…๋˜๋Š” ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ์˜ ์ธก์ •๊ฐ’($\check{P}_{k}^{g}$)์€ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰์„ ์ดˆ๊ณผํ•˜๋ฉฐ ํ•˜๋ฃจ์ „ ๊ณ„์‚ฐ๋œ ๊ณต๊ธ‰์ „๋ ฅ ๊ณ„ํš๊ฐ’($\hat{P}_{k}^{g}$) ๋ณด๋‹ค ํฌ๋‹ค. ๋”ฐ๋ผ์„œ ์‹ (12)์— ๋”ฐ๋ผ ESS์˜ ๋ฐฉ์ „๋Ÿ‰์€ ํ•˜๋ฃจ์ „ ์‚ฐ์ •ํ•œ ๋ฐฉ์ „ ๊ณ„ํš๊ฐ’($\hat{P}_{k}^{e}$) ๋Œ€์‹ ์— ๋งค์‹œ์ ๋งˆ๋‹ค ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ์˜ ์ธก์ •๊ฐ’($\check{P}_{k}^{g}$)์—์„œ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰($P_{limit}$)์„ ์ฐจ๊ฐํ•œ ๊ฐ’์„ ์„ ํƒํ•˜์—ฌ ๊ณ„ํš๊ฐ’๋ณด๋‹ค ๋” ๋งŽ์ด ๋ฐฉ์ „ํ•˜๋„๋ก ์‹ค์‹œ๊ฐ„ ๋ณด์ƒ ์ง€๋ น์„ ์‹ค์‹œํ•œ๋‹ค. ๋”ฐ๋ผ์„œ, ๊ทธ๋ฆผ 13(b)์™€ ๊ฐ™์ด ์ต์ผ 0~8 ์‹œ๊ฐ„๋Œ€์—์„œ ์ถฉ์ „๋™์ž‘์€ ๊ณ„ํš๋Œ€๋กœ ์ˆ˜ํ–‰๋˜์ง€๋งŒ 10~12 ์‹œ๊ฐ„๋Œ€์—์„œ ํ”ผํฌ์ €๊ฐ์„ ์œ„ํ•œ ๋ฐฉ์ „๊ณ„ํš์€ ์ต์ผ ์‹ค์‹œ๊ฐ„ ์ธก์ •์„ ํ†ตํ•ด ์กฐ์ •๋˜๋Š” ์‹ค์‹œ๊ฐ„ ๋ณด์ƒ์ง€๋ น์— ๋”ฐ๋ผ ๋” ๋งŽ์€ ๋ฐฉ์ „๋Ÿ‰์ด ๋ฐœ์ƒํ•˜๊ณ  ์ด์— ๋”ฐ๋ผ SOC๋Š” 40%๊นŒ์ง€ ์†Œ์ง„๋จ์„ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋‹ค. ๋”ฐ๋ผ์„œ, ์„ ๋กœ์— ์œ ์ž…๋˜๋Š” ์ „์›์ธก ๊ณต๊ธ‰์ „๋ ฅ๋„ ์ƒ์‹œ ์šด์ „์šฉ๋Ÿ‰ ์ด๋‚ด๋กœ ์œ ์ง€๋จ์„ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋‹ค.

๊ทธ๋ฆผ 13. ์‹ค์‹œ๊ฐ„ ESS ์ถฉ๋ฐฉ์ „ ๋ณด์ƒ (์‚ฌ๋ก€)

Fig. 13. Real-time ESS charge/discharge compensation (Case)

../../Resources/kiee/KIEEP.2023.72.3.139/fig13.png

6. ๊ฒฐ ๋ก 

๋ณธ ๋…ผ๋ฌธ์—์„œ ๊ณผ๋ถ€ํ•˜ ํ•ด์†Œ๋ฅผ ํ†ตํ•ด ๋ฐฐ์ „์„ ๋กœ์˜ ์šด์ „์šฉ๋Ÿ‰์„ ์œ ์ง€ํ•˜๊ธฐ ์œ„ํ•œ ๋ชฉ์ ์œผ๋กœ ์„ค์น˜๋˜๋Š” ESS์˜ ์šด์ „ ๋ฐฉ๋ฒ•์„ ์ œ์‹œํ•˜์˜€๋‹ค. ARIMA ๋ชจ๋ธ์„ ์ด์šฉํ•˜์—ฌ ํ•ด๋‹น์„ ๋กœ์˜ ๋ถ€ํ•˜๋ฅผ ์˜ˆ์ธกํ•˜๋Š” ๊ธฐ๋ฒ•์„ ์ œ์‹œํ•˜์˜€๋‹ค. ํ•˜๋ฃจ์ „ ํ•ด๋‹น์„ ๋กœ์˜ ๋ถ€ํ•˜๋ฅผ ์˜ˆ์ธกํ•˜์—ฌ ์ต์ผ ํ”ผํฌ์˜ ๋ฐœ์ƒ์—ฌ๋ถ€๋ฅผ ํ™•์ธํ•˜๊ณ , ์˜ˆ์ธก๋œ ๋ถ€ํ•˜๋ฅผ ๊ธฐ์ค€์œผ๋กœ ์„ ๋กœ์— ์„ค์น˜๋œ ESS์˜ ์šด์ „์Šค์ผ€์ค„์„ QP ์ตœ์ ํ™” ๊ธฐ๋ฒ•์„ ์ด์šฉํ•˜์—ฌ ๊ตฌํ•˜๋Š” ๋ฐฉ๋ฒ•์„ ์ œ์‹œํ•˜์˜€๋‹ค. ๋˜ํ•œ ํ•˜๋ฃจ์ „ ๋ถ€ํ•˜์˜ˆ์ธก๊ฐ’์ด ์ต์ผ ๋ถ€ํ•˜์‹ค์ธก๊ฐ’๊ณผ ์ฐจ์ด๊ฐ€ ๋ฐœ์ƒํ•  ๊ฒฝ์šฐ์—๋„ ์ด๋ฅผ ๋ณด์ƒํ•˜๋Š” ์‹ค์‹œ๊ฐ„ ๋ณด์ƒ ๋ฐฉ์•ˆ์„ ์ œ์‹œํ•˜์˜€๋‹ค. ์‹ค์ œ ๋ฐฐ์ „์„ ๋กœ์™€ ์šด์ „์ด๋ ฅ์„ ํ†ตํ•œ ์‚ฌ๋ก€์—ฐ๊ตฌ๋ฅผ ์ˆ˜ํ–‰ํ•˜์—ฌ ์ œ์•ˆ๋œ ๋ฐฉ๋ฒ•์˜ ์œ ํšจ์„ฑ์„ ํ™•์ธํ•˜์˜€๋‹ค. ํ–ฅํ›„ ์ œ์•ˆ๋œ ๋ฐฉ์•ˆ์„ ํƒœ์–‘๊ด‘์ด ์ง€๋ฐฐ์ ์ธ ์„ ๋กœ์—๋„ ์ ์šฉํ•  ์ˆ˜ ์žˆ๋„๋ก ์ถ”๊ฐ€์ ์œผ๋กœ ์—ฐ๊ตฌํ•˜๊ณ ์ž ํ•œ๋‹ค.

Acknowledgements

This research was supported by the Korea Electric Power Corporation as a part of the R&D project No. R22DA08.

References

1 
M. J. Jang, H. J. Choi, C. G. Lim, B. W. An, and J. S. Sim, โ€œOptimization of ESS Scheduling for Cost Reduction in Commercial and Industry Customers in Korea,โ€ Sustainability, vol. 14, no. 6, pp. 3605, 2022.DOI
2 
J. H. Kong, F. H. Jufri, B. O. Kang, and J. S. Jung, โ€œDevelopment of ESS Scheduling Algorithm to Maximize the Potential Profitability of PV Generation Supplier in South Korea,โ€ JEET, vol. 13. no. 6, pp. 2227, 2018.URL
3 
S. M. Cho, and S. Y. Yun, โ€œOptimal Power Assignment of Energy Storage Systems to Improve the Energy Storage Efficiency for Frequency Regulation,โ€ Energies, vol. 10, no. 12, pp. 2092, 2017.DOI
4 
J. Deboever, J. Peppanen, A. Maitra, G. Damato, J. Taylor, and J. Patel, โ€œEnergy Storage as a Non-Wires Alternative for Deferring Distribution Capacity Investments,โ€ IEEE/PES Transmission and Distribution Conference and Exposition, 2018.DOI
5 
A. Ebrahimi, and M. Ziabasharhagh, โ€œIntroducing a novel control algorithm and scheduling procedure for optimal operation of energy storage systems,โ€ Elsevier Energy, no. 252, 2022.DOI
6 
D. Kodaira, B. G. Yu, W. W. Jung, and S. K. Han, โ€œOptimized ESS Operation for Peak Shaving based on Probabilistic Load Prediction,โ€ IEEE Innovative Smart Grid Technologies-Asia, pp. 1199, 2018.DOI
7 
N. H. Cho, K. J. Kim, and J. C. Kim, โ€œThe Optimal Number of Dividing and Connecting Per Feeder in Korea Distribution System,โ€ The Transactions of the Korean Institute of Electrical Engineers, vol. 51, no. 7, pp. 349, 2002.URL
8 
B. Nepal, M. Yamaha, A. Yokoe, and T. Yamaji, โ€œElectricity load forecasting using clustering and ARIMA model for energy management in buildings,โ€ Japan Architectural Review, vol. 3, no. 1, pp. 62, 2020.DOI
9 
S. Y. Choi, and S. W. Min, โ€œOptimal Scheduling and Operation of the ESS for Prosumer Market Environment in Grid-Connected Industrial Complex,โ€ IEEE Transactions on industry applications, vol. 54, no. 3, pp. 1949, 2018.DOI
10 
C. H. Shin, W. W. Jung, S. M. Cho, and H. J. Cha, โ€œDevelopment of UPS combined ESS operation method considering load forecasting error,โ€ The Transactions of the Korean Institute of Electrical Engineers, vol. 69P, no. 2, pp. 76, 2020.URL

์ €์ž์†Œ๊ฐœ

์‹ ์ฐฝํ›ˆ(Chang-hoon Shin)
../../Resources/kiee/KIEEP.2023.72.3.139/au1.png

He received B.S., and M.S. degree in Kyungpook National University, Daegu, Korea in 1992 and 1994, respectively. He is currently pursuing the Ph.D. degree in electrical engineering at Chungnam National University, Daejeon, Korea. Since 1994, he has been with the KEPCO Research Institute and his research interests include Advanced Distribution Management System, distributed energy resources and micro-grid.

์ฐจํ•œ์ฃผ(Hanju Cha)
../../Resources/kiee/KIEEP.2023.72.3.139/au2.png

He received his B.S. degree in Electrical Engineering from Seoul National University, Seoul, Korea, in 1988; his M.S. degree in Electrical Engineering from Pohang Institute of Science and Technology, Pohang, Korea, in 1990; and his Ph.D. degree in Electrical Engineering from Texas A&M University, College Station, TX, USA, in 2004. From 1990 to 2001, he was with LG Industrial Systems, Anyang, Korea, where he was engaged in the development of power electronics and adjustable speed drives. Since 2005, he has been with the Department of Electrical Engineering, Chungnam National University, Daejeon, Korea. He was a Visiting Professor in the United Technology Research Center, Hartford, CT, USA, in 2009. His current research interests include high-power converter, ac/dc, dc/ac, and ac/ac converter topologies, power quality, and utility interface issues for distributed energy systems and microgrids.