1. Gabatarwa
Gine-gine suna da gagarumin gudunmawa ga yawan amfani da makamashi a duniya, kamar yadda aka nuna a Hoto na 1 wanda ke nuna babban rabonsu a ƙasashe kamar Indiya, Amurka, Sin, Koriya, da Ostiraliya. Ƙoƙarin ci gaba mai dorewa da yaduwar na'urori masu wayo, Tsarin Gudanar da Gine-gine (BMS), da na'urori masu auna yanayi sun haifar da tarin bayanai na makamashi na gine-gine. Waɗannan bayanan sun ba da dama da ba a taɓa samun irinta ba don ƙetare binciken makamashi na al'ada, da takardun kuɗi na wata-wata zuwa fahimta mai ci gaba, mai zurfi, da ingantaccen amfani da makamashi na gine-gine. Babban ƙalubalen shine yin amfani da waɗannan bayanan yadda ya kamata a duk faɗin fanni daga fahimta zuwa yanke shawara da haɗa mazauna.
Gudunmawar Makamashi na Gine-gine (Hoto na 1)
Bayanai daga PDF sun nuna gine-gine suna da babban rabo a cikin amfani da makamashi na ƙasa, misali, kusan kashi 30-40% a cikin ƙasashe masu ci gaba kamar Amurka da Ostiraliya, suna nuna mahimmancin buƙatar matakan inganci.
2. Tsarin I 5 don Amincewa da Bayanai don Ingantacciyar Amfani da Makamashi
Takardar ta ba da shawarar cikakken tsari wanda aka ƙunsa ta hanyar manyan fannoni guda biyar, wanda ake kira "I 5," don tabbatar da ingantacciyar amfani da makamashi ta hanyar bayanai a gine-gine.
2.1 Yi Amfani da Kayan Aiki da Kyau
Wannan ya ƙunshi sanya na'urori masu auna yanayi (na'urori masu wayo, wuraren BMS, na'urori masu auna yanayi) da dabara don tattara bayanai masu zurfi, masu dacewa ba tare da maimaitawa ko farashi mara amfani ba. Manufar ita ce cimma isasshen ingancin bayanai don bincike na ƙasa yayin la'akari da matsalolin tattalin arziki.
2.2 Haɗa Tsarin Ƙananan Tsarin
Gine-gine na zamani sun ƙunshi tsarin da ba a haɗa su ba (HVAC, haske, tsaro). Wannan ginshiƙi yana jaddada karya waɗannan rukunonin ta hanyar ba da damar kwararar bayanai da sadarwa tsakanin ƙananan tsarin daban-daban (misali, BMS, na'urori masu wayo, na'urorin mazauna) don ƙirƙirar cikakkiyar hangen nesa na ayyukan ginin.
2.3 Yanke Shawara ta Hanyar Fahimta
Bayanai marasa tsari ba su da amfani idan babu fahimta. Wannan ya ƙunshi amfani da binciken bayanai, koyon inji, da ƙirar ƙididdiga ga bayanan da aka tattara kuma aka haɗa su don fahimtar tsari, gano kurakurai, hasashen buƙatu, da gano damar ingantawa. Wannan yana canza bayanai zuwa hankali mai aiki.
2.4 Haɗa Mazauna
Halin mazauna shine babban abin da ke haifar da amfani da makamashi. Wannan yanki yana mai da hankali kan yin amfani da bayanai don ba da ra'ayi ga mazauna (ta hanyar allunan bayanai, faɗakarwa, wasa) da yuwuwar ba su damar shiga cikin shirye-shiryen amsa buƙatu, ƙirƙirar madauki na ra'ayi wanda ke haɓaka halin kula da makamashi.
2.5 Ayyuka na Hankali
Ƙarshen I 4 da suka gabata. Yana nufin sarrafa tsarin gine-gine ta atomatik ko rabin atomatik bisa ga fahimta da ra'ayin mazauna. Misalai sun haɗa da inganta jadawalin HVAC da ƙarfi, sarrafa haske bisa ga zama da hasken rana, da shiga cikin abubuwan amsa buƙatan grid da kansu.
Mahimman Fahimta
- Tsarin yana maimaitawa kuma an haɗa shi; nasara a cikin "I" ɗaya sau da yawa ya dogara da ci gaba a wasu.
- Yana canza hankali daga tattara bayanai keɓaɓɓu zuwa tsarin rufaffiyar ma'auni, bincike, da aiki.
- Haɗa mazauna yana da mahimmanci amma galibi shine mafi wahala a aiwatar da shi yadda ya kamata.
3. Nazarin Hali: Binciken Amfani da Wutar Lantarki ba tare da Katsalandan ba (NILM)
Takardar tana amfani da NILM a matsayin matsala da aka yi nazari sosai wacce ta shafi duk I 5. NILM tana nufin raba jimillar amfani da wutar lantarki na gini (daga na'urar aunawa mai wayo guda ɗaya) zuwa nauyin kayan aiki ɗaya.
- Kayan Aiki: Ya dogara da na'urar aunawa mai wayo guda ɗaya, wacce aka sanya ta da kyau.
- Haɗin Kai: Na iya haɗawa da wasu rafukan bayanai (misali, lokacin rana, zafin yanayi) don inganta daidaito.
- Fahimta: Cibiyar NILM, ta amfani da sarrafa siginar da algorithms na koyon inji (kamar ƙirar Markov ɓoye) don fahimtar yanayin kayan aiki daga jimillar bayanai.
- Haɗawa: Yana ba da cikakken bayani na matakin kayan aiki ga mazauna, yana ba su ikon canza halayensu.
- Ayyuka na Hankali: Bayanan da aka raba za su iya haifar da ayyuka ta atomatik, kamar kashe kayan aiki mara kyau ko tsara manyan kayan aiki don lokutan da ba su da yawa.
4. Cikakkun Bayanai na Fasaha & Tsarin Lissafi
A tsakiyar fahimta (I3) da matsaloli kamar NILM akwai ƙirar lissafi. Tsarin gama gari don raba nauyi yana ɗaukar siginar ƙarfin gabaɗaya $P_{total}(t)$ a matsayin jimillar ƙarfin kayan aiki ɗaya tare da ƙara hayaniya:
$P_{total}(t) = \sum_{i=1}^{N} s_i(t) \cdot P_i(t) + \epsilon(t)$
Inda:
- $N$ shine adadin kayan aiki.
- $s_i(t)$ shine yanayin (KUNNE/KASHE ko yanayi da yawa) na kayan aiki $i$ a lokacin $t$.
- $P_i(t)$ shine amfani da wutar lantarki na kayan aiki $i$ lokacin da yake cikin yanayin da $s_i(t)$ ya nuna.
- $\epsilon(t)$ shine hayaniya ko nauyin da ba a ƙirƙira ba.
5. Sakamakon Gwaji & Bayanin Ginshiƙi
PDF ɗin ya haɗa da Hoto na 1, ginshiƙi mai taken "Gudunmawar gine-gine ga jimillar amfani da makamashi." Yana kwatanta kashi na jimillar amfani da makamashi na ƙasa da aka dangana ga gine-gine a cikin ƙasashe biyar: Indiya, Amurka, Sin, Koriya, da Ostiraliya. Ginshiƙin yana ƙarfafa ainihin jigon takardar, yana nuna cewa gine-gine suna da babban rabo mai ban sha'awa (mai yiwuwa daga ~20% zuwa sama da 40%) na amfani da makamashi, don haka kafa babban tasirin yuwuwar samun riba a wannan fanni. Duk da yake ba a bayyana takamaiman sakamakon ƙididdiga na NILM ko wasu aikace-aikace ba a cikin abin da aka ba da shi, tsarin yana nuna cewa nasarar amfani da I 5 zai haifar da sakamako masu aunawa kamar:
- Rage jimillar amfani da makamashi na gini (kWh).
- Ingantaccen daidaiton gano kuskure (misali, gano rashin ingancin HVAC da wuri).
- Ƙara ma'aunin haɗa mazauna (misali, amfani da allon bayanai, shiga cikin shirye-shiryen ceton kuɗi).
- Ceton kuɗi daga ingantaccen aiki da shiga cikin amsa buƙatu.
6. Tsarin Nazari: Misali ba na Lamba ba
Yi la'akari da ginin ofis na kasuwanci da ke nufin rage nauyin sanyaya. Nazari da ke bin tsarin I 5 zai ci gaba kamar haka:
- Kayan Aiki: Sanya na'urori masu auna zafi masu wayo a kowane yanki, na'urori masu auna wutar lantarki akan na'urorin HVAC, da na'urori masu auna zafin yanayi/ɗanɗano/zama.
- Haɗin Kai: Kwarara bayanai daga duk na'urori masu auna zuwa dandalin bayanai na tsakiya, danganta bayanan zama da yanayin zafi na yankin HVAC da jan ƙarfi.
- Fahimta: Aiwatar da bincike don gano tsari: misali, "Yankin A yana sanyaya koyaushe zuwa 22°C ko da lokacin da zama ya kasance ƙasa da 10% bayan 7 na dare," ko "Ingancin sanyaya yana raguwa da kashi 15% lokacin da zafin waje ya wuce 35°C."
- Haɗawa: Aika rahotanni na mako-mako ga masu gudanar da kayan aiki suna nuna ɓarnatar da sa'o'in sanyaya. Ai wata manufa da ke ba mazauna damar jefa ƙuri'a kan ɗan gyara wurin saitin zafin jiki a lokacin rani mai zafi, haɗa shi da gudummawar agaji don makamashin da aka adana.
- Ayyuka na Hankali: Sarrafa jadawalin HVAC ta atomatik don ɗaukar yanayin "koma baya" a yankuna masu ƙarancin zama bayan sa'o'i. Haɗa shi da siginar amsa buƙatu daga ma'aikatar don sanyaya ginin na ɗan lokaci kafin wani babban taron.
7. Aikace-aikace na Gaba & Jagororin Bincike
Juyin halittar tsarin I 5 yana nuna zuwa ga wasu jagorori na gaba:
- Sarrafa Hasashen AI: Matsawa daga ƙirar tushen ƙa'ida zuwa ƙirar koyon ƙarfafawa mai zurfi waɗanda za su iya ci gaba da koyo da inganta ayyukan gini a ainihin lokaci, kamar yadda AlphaGo ya ƙware Go.
- Gine-gine Masu Aiki da Grid (GEB): Zurfafa ginshiƙin "Ayyuka na Hankali" don sanya gine-gine su zama masu aiki, masu sauri a cikin grid mai wayo, suna ba da ayyukan grid kamar daidaita mitar.
- Bincike Mai Kiyaye Sirri: Haɓaka koyo na tarayya ko dabarun ɓoyayyen suna don ba da damar fahimta da haɗin kai (I2, I3) ba tare da lalata sirrin mazauna ba, babban abin damuwa tare da bayanai masu kyau.
- Tagwayen Digital: Ƙirƙirar ƙirar gine-gine masu inganci waɗanda ake ci gaba da sabunta su da bayanan na'urar auna (I1, I2), yana ba da damar kwaikwayo, tsara yanayi, da kulawa mai ƙarfi (I3, I5).
- AI a cikin Madauki na Mutum: Haɓaka "Haɗa Mazauna" (I4) tare da ƙarin ƙirar halayya da musanyawar fuska waɗanda ke keɓance ra'ayi da ƙarfafawa, ƙara tasiri.
8. Nassoshi
- Batra, N., Singh, A., Singh, P., Dutta, H., Sarangan, V., & Srivastava, M. (2014). Data Driven Energy Efficiency in Buildings. arXiv preprint arXiv:1404.7227.
- Ma'aikatar Makamashi ta Amurka. (2023). Grid-Interactive Efficient Buildings. An samo daga https://www.energy.gov/eere/buildings/grid-interactive-efficient-buildings
- Zhu, J., Park, T., Isola, P., & Efros, A. A. (2017). Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks. Proceedings of the IEEE International Conference on Computer Vision (ICCV). (Nassoshi na CycleGAN don kwatancen AI mai haifarwa).
- Hukumar Makamashi ta Duniya (IEA). (2022). Gine-gine. An samo daga https://www.iea.org/topics/buildings
- Kolter, J. Z., & Johnson, M. J. (2011). REDD: A Public Data Set for Energy Disaggregation Research. Proceedings of the SustKDD Workshop.
9. Nazari na Asali & Bita Mai Ma'ana
Babban Fahimta: Tsarin "I 5" na Batra da sauransu ba kawai rarrabuwa ba ne; yana da ƙirar balaga don canjin dijital na fannin makamashi na gine-gine. Takardar ta gano daidai cewa ainihin matsalar ba ta ƙara zama tattara bayanai (I1), amma haɗin tsarin da amfani da hankali na waɗannan bayanan (I2-I5). Wannan yayi daidai da juyin halitta da aka gani a wasu fannonin da ke da yawan bayanai, kamar hangen nesa na kwamfuta, inda ƙirƙirar manyan, daidaitattun bayanai kamar ImageNet ko MNIST (wanda aka ambata a cikin PDF) suka kasance abin haifar da juyin juya halin AI. Fannin gine-gine yanzu yana tattara "ImageNet" ɗinsa ta hanyar na'urori masu wayo da BMS, amma algorithms da tsarin kasuwanci don amfani da shi har yanzu suna cikin ƙuruciyarsu.
Kwararar Ma'ana & Ƙarfafawa: Ƙarfin tsarin yana cikin cikakkiyar ma'ana, madauki na rufaffiyar. Yana daidaita jerin tafiya daga kayan aiki na zahiri zuwa aiki ta atomatik, yana jaddada cewa bayanai ba tare da fahimta ba hayaniya ne, kuma fahimta ba tare da aiki ba ilimi ne. Haɗa "Haɗa Mazauna" (I4) yana da hankali musamman, yana yarda cewa mafi kyawun tsarin sarrafa AI na iya lalacewa ta hanyar halayyar ɗan adam—darasi da aka saba koya da wahala a cikin aikawa na ainihin duniya. Yin amfani da NILM a matsayin nazarin hali mai tsallakewa zaɓi ne na koyarwa mai kyau, yana sanya tsarin da ba a taɓa gani ba ya zama abin taɓo.
Kurakurai & Gaps: Babban aibi, gama gari ga takardun tsarin na zamaninsa (2014), shine rashin bayyana manyan ƙalubale a cikin "Haɗin Kai" (I2). Takardar ta ambaci tsarin da ba a haɗa su ba amma ba ta cika fuskantar gine-ginen gado ba, ka'idoji na mallaka (BACnet, Modbus), da haɗarin tsaro na dijital waɗanda ke sanya haɗin bayanai ya zama tsada da rikitaccen injiniyanci, ba kawai mataki na ra'ayi ba. Bugu da ƙari, yayin da yake nuna sirri, babban tasirin sirri na bayanan halayya masu kyau da ake buƙata don I3 da I4 ba a bincika su sosai ba. Tsarin kuma ya dogara sosai akan hangen nesa na fasaha mai kyau; yana rage shingen ƙungiya da na kuɗi don karɓa, musamman a cikin kasuwar gidaje da aka raba.
Fahimta Mai Aiki & Hangen Nesa na Gaba: Ga masu aiki, fahimtar mai aiki ita ce daina bin "I's" keɓaɓɓu. Sanya na'urori masu auna 10,000 (I1) ba tare da shirin I3 da I5 ba ɓarna ne. Fara da bayyanannen amfani (misali, rage kololuwa, gano kuskure) kuma a yi aiki a baya ta hanyar tsarin. Gaba yana cikin motsawa daga bayanin / shawarwarin bincike zuwa tsarin mai haifarwa da mai cin gashin kansa. Tunanin "CycleGAN don gine-gine": tsarin wanda ba kawai ya gano rashin inganci ba amma yana haifarwa da kuma kwaikwayo da yawa ingantattun dabarun sarrafawa (kamar GANs suna haifar da hotuna) kafin a tura mafi kyawun ɗaya. Haɗin kai tare da manyan makamashi mai sabuntawa da tsarin mota zuwa grid (V2G) zai zama iyaka na gaba, canza gine-gine daga masu amfani marasa aiki zuwa manyan nodes a cikin cikakken makamashi mai rahusa, mai juriya. I 5 sun ba da zane na asali, amma shekaru goma masu zuwa za su kasance game da gina "kwakwalwar" AI da "tsarin jijiya" mai aiki don tabbatar da gaskiyarsa.