{"id":13514,"date":"2025-07-17T15:13:41","date_gmt":"2025-07-17T13:13:41","guid":{"rendered":"https:\/\/onload.reinhausen.com\/de\/?p=13514"},"modified":"2025-07-17T15:13:43","modified_gmt":"2025-07-17T13:13:43","slug":"artificial-intelligence-ai-transformers-maschinenfabrik-reinhausen","status":"publish","type":"post","link":"https:\/\/onload.reinhausen.com\/en\/07-2025\/artificial-intelligence-ai-transformers-maschinenfabrik-reinhausen\/","title":{"rendered":"\u201eWithout Data, AI is worthless.\u201c"},"content":{"rendered":"\n
\n\n\n\n

Artificial intelligence is already helping doctors with diagnoses, cars can drive autonomously thanks to its support, and AI is even writing software autonomously. Will AI soon be controlling our power grids fully automatically?<\/strong><\/h4>\n\n\n\n

Prob\u00ada\u00adbly not, since we are still in the begin\u00adning. And in crit\u00adi\u00adcal infra\u00adstruc\u00adture in par\u00adtic\u00adu\u00adlar, it is still usu\u00adal\u00adly peo\u00adple who make the final deci\u00adsion. This is also the case in med\u00adi\u00adcine. But Arti\u00adfi\u00adcial intel\u00adli\u00adgence will grad\u00adu\u00adal\u00adly play an ever greater role in mas\u00adter\u00ading the increas\u00ading com\u00adplex\u00adi\u00adty of our net\u00adworks. And it is already being used in many areas. At MR, we have been work\u00ading with machine learn\u00ading for 20 years. It all start\u00aded with our dehy\u00addrat\u00ading breathers. That was the first time we imple\u00adment\u00aded self-learn\u00ading algo\u00adrithms for con\u00adtrol. We have grad\u00adu\u00adal\u00adly com\u00adplet\u00aded our sen\u00adsor port\u00adfo\u00adlio and equipped our prod\u00aducts with intel\u00adli\u00adgent algo\u00adrithms. In par\u00adtic\u00adu\u00adlar our solu\u00adtions for eval\u00adu\u00adat\u00ading oper\u00adat\u00ading resources are becom\u00ading increas\u00ading\u00adly com\u00adpre\u00adhen\u00adsive and pre\u00adcise thanks to AI. How\u00adev\u00ader, the poten\u00adtial is far from exhaust\u00aded.<\/p>\n\n\n\n

What is the greatest potential of AI solutions?<\/strong><\/h4>\n\n\n\n

In view of the tech\u00adno\u00adlog\u00adi\u00adcal advances in com\u00adput\u00ader and data pro\u00adcess\u00ading, cur\u00adrent\u00adly par\u00adtic\u00adu\u00adlar\u00adly in large lan\u00adguage mod\u00adels, there seem to be no lim\u00adits. In the near future, I see the great\u00adest advan\u00adtages in the sta\u00adtus assess\u00adment of the var\u00adi\u00adous assets in the net\u00adwork and in net\u00adwork man\u00adage\u00adment. How\u00adev\u00ader, for AI to be able to exploit its advan\u00adtages, it first needs a lot of data \u2013 with\u00adout data, AI is worth\u00adless. Com\u00adpre\u00adhen\u00adsive data col\u00adlec\u00adtion for trans\u00adform\u00aders, cir\u00adcuit break\u00aders and cables is there\u00adfore essen\u00adtial. To make this pos\u00adsi\u00adble, we are work\u00ading on reduc\u00ading the costs of dig\u00adi\u00adtal\u00adiza\u00adtion.<\/p>\n\n\n

\n
<\/figure>\n<\/div>\n\n\n

How would you like to reduce the costs of digitalization transformers?<\/strong><\/h4>\n\n\n\n

Arti\u00adfi\u00adcial intel\u00adli\u00adgence helps us here too. At the moment, I can\u2019t tell you too many details, but we are cur\u00adrent\u00adly test\u00ading algo\u00adrithms that draw con\u00adclu\u00adsions from the mea\u00adsure\u00adment data of a few sen\u00adsors and cal\u00adcu\u00adlate val\u00adues for oth\u00ader fic\u00adti\u00adtious or dig\u00adi\u00adtal sen\u00adsors. The soft\u00adware com\u00adpen\u00adsates for the func\u00adtion of oth\u00ader sen\u00adsors, so to speak, by intel\u00adli\u00adgent\u00adly com\u00adbin\u00ading the mea\u00adsure\u00adment data with his\u00adtor\u00adi\u00adcal data and eval\u00adu\u00adat\u00aded mod\u00adels. This enables us to assess the con\u00addi\u00adtion of trans\u00adform\u00aders cost-effec\u00adtive\u00adly and effi\u00adcient\u00adly.<\/p>\n\n\n\n

Regarding more efficient condition assessment: Can this already be realized today?<\/strong><\/h4>\n\n\n\n

ETOS\u00ae<\/sup> is already estab\u00adlished on the mar\u00adket at field lev\u00adel for mon\u00adi\u00adtor\u00ading and col\u00adlect\u00ading data from trans\u00adform\u00aders. With its Asset Intel\u00adli\u00adgence func\u00adtion and our DGA inter\u00adpre\u00adta\u00adtion, for exam\u00adple, we already offer data and mod\u00adel-based solu\u00adtions for more effi\u00adcient con\u00addi\u00adtion assess\u00adment includ\u00ading rec\u00adom\u00admend\u00aded actions for oper\u00ada\u00adtors. That trans\u00adform\u00aders them\u00adselves will soon be active\u00adly con\u00adtact\u00ading asset man\u00adagers for con\u00addi\u00adtion-based main\u00adte\u00adnance or nec\u00ades\u00adsary repairs is obvi\u00adous. Mes\u00adsages will then come either from the oper\u00adat\u00ading sys\u00adtem on the trans\u00adformer \u2013 such as our ETOS\u00ae<\/sup> \u2013 or from a dig\u00adi\u00adtal twin in an asset per\u00adfor\u00admance man\u00adage\u00adment sys\u00adtem such as TESSA\u00ae<\/sup> APM.<\/p>\n\n\n\n

Can AI also help to optimize grid management?<\/strong><\/h4>\n\n\n\n

Def\u00adi\u00adnite\u00adly, that\u2019s the future! We are already work\u00ading on such a solu\u00adtion togeth\u00ader with a part\u00adner. Real-time con\u00addi\u00adtion assess\u00adment of var\u00adi\u00adous trans\u00adformer assets, cir\u00adcuit break\u00aders and cables in com\u00adbi\u00adna\u00adtion with his\u00adtor\u00adi\u00adcal data and, for exam\u00adple, weath\u00ader fore\u00adcasts, offers immense poten\u00adtial. One exam\u00adple here is proac\u00adtive trans\u00adformer cool\u00ading in order to with\u00adstand upcom\u00ading pow\u00ader peaks or over\u00adloads with a min\u00adi\u00admized reduc\u00adtion in ser\u00advice life; and this is just the begin\u00adning. To some extent, this is already a real\u00adi\u00adty, but in the future AI will play a much greater role in net\u00adwork man\u00adage\u00adment.<\/p>\n\n\n\n

\n
<\/div>\n\n\n\n
\n

 \u201eWe have been work\u00ading with trans\u00adform\u00aders and their com\u00adpo\u00adnents for sev\u00ader\u00adal decades and know exact\u00adly what we are doing. This exper\u00adtise is embed\u00added in our algo\u00adrithms.\u201c<\/p>\nRoland G\u00f6tz, Head of Inno\u00adva\u00adtion and Tech\u00adnol\u00ado\u00adgy at MR<\/cite><\/blockquote>\n\n\n\n

<\/div>\n<\/div><\/div>\n\n\n\n

How can you ensure that AI is used safely in critical infrastructure?<\/strong><\/h4>\n\n\n\n

Any\u00adone who uses Chat\u00adG\u00adPT knows that answers can also be wrong because AI some\u00adtimes hal\u00adlu\u00adci\u00adnates a lit\u00adtle. Although the results seem plau\u00adsi\u00adble, they are fac\u00adtu\u00adal\u00adly wrong. Of course, this must not hap\u00adpen with our AI solu\u00adtions and mod\u00adels, which have to be one hun\u00addred per\u00adcent reli\u00adable. To train our algo\u00adrithms, we there\u00adfore only use data that we have col\u00adlect\u00aded and processed our\u00adselves in field tests or that has been made avail\u00adable to us by net\u00adwork oper\u00ada\u00adtors for these pur\u00adpos\u00ades. <\/p>\n\n\n\n

Every step, from label\u00ading the data to check\u00ading its plau\u00adsi\u00adbil\u00adi\u00adty and test\u00ading, is car\u00adried out by our own experts. We have been work\u00ading with trans\u00adform\u00aders and their com\u00adpo\u00adnents for sev\u00ader\u00adal decades and know exact\u00adly what we are doing. This exper\u00adtise is embed\u00added in our algo\u00adrithms. Because our prod\u00aducts are used in crit\u00adi\u00adcal infra\u00adstruc\u00adture, they are also sub\u00adject to the new EU AI Act which impos\u00ades par\u00adtic\u00adu\u00adlar\u00adly strict rules on the han\u00addling of data and the use of arti\u00adfi\u00adcial intel\u00adli\u00adgence in gen\u00ader\u00adal. Of course, we com\u00adplete\u00adly com\u00adply with all known require\u00adments of the EU AI Act.<\/p>\n\n\n\n

What happens to customer data?<\/strong><\/h4>\n\n\n\n

Quite sim\u00adply: it stays with the cus\u00adtomer. Nei\u00adther MR nor third par\u00adties have access to cus\u00adtomer data, either at the field lev\u00adel with ETOS\u00ae<\/sup> or in the Asset Per\u00adfor\u00admance Man\u00adage\u00adment Sys\u00adtem with TESSA\u00ae<\/sup>. To this end, we have imple\u00adment\u00aded an ISMS infor\u00adma\u00adtion secu\u00adri\u00adty man\u00adage\u00adment sys\u00adtem based on ISO 27001 to ensure that data pro\u00adtec\u00adtion and IT secu\u00adri\u00adty are guar\u00adan\u00adteed for our sys\u00adtem solu\u00adtions dur\u00ading devel\u00adop\u00adment and, of course, lat\u00ader dur\u00ading oper\u00ada\u00adtion. But of course we are always hap\u00adpy to receive data from real net\u00adwork oper\u00ada\u00adtions because this makes our AI solu\u00adtions even bet\u00adter which also pro\u00advides ben\u00ade\u00adfits to our cus\u00adtomers.<\/p>\n\n\n\n

<\/div>\n\n\n\n

4 examples of how artificial intelligence can help energy suppliers<\/h2>\n\n\n\n
\n\n\n\n
\n
\n
<\/span>
\n

<\/p>\n<\/div><\/div>\n\n\n\n

<\/span>
\n

<\/p>\n<\/div><\/div>\n\n\n\n

<\/span>
\n

<\/p>\n<\/div><\/div>\n\n\n\n

<\/span>
\n

<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n
\n
<\/div>\n\n\n\n
\n
\n

1. Optimized network management<\/h3>\n\n\n\n

The increas\u00ading com\u00adplex\u00adi\u00adty of grids is mak\u00ading con\u00adtrol more and more dif\u00adfi\u00adcult. AI can help to opti\u00admize and, in future, auto\u00admate grid man\u00adage\u00adment by incor\u00adpo\u00adrat\u00ading both fore\u00adcast data and the sta\u00adtus data of oper\u00adat\u00ading resources into the deci\u00adsion-mak\u00ading process.<\/p>\n<\/div><\/div>\n<\/div>\n\n\n\n

<\/div>\n<\/div>\n\n\n\n
\n
<\/div>\n\n\n\n
\n
\n

2. Self-learing transformers<\/h3>\n\n\n\n

A trans\u00adformer that opti\u00admizes itself? This may also be pos\u00adsi\u00adble with AI-based sys\u00adtems. A trans\u00adformer would then con\u00adtin\u00adu\u00adous\u00adly ana\u00adlyze grid para\u00adme\u00adters and sen\u00adsor data and com\u00adbine them with his\u00adtor\u00adi\u00adcal data in order to proac\u00adtive\u00adly adjust its tap-chang\u00aders to com\u00adpen\u00adsate for grid fluc\u00adtu\u00ada\u00adtions in real time.<\/p>\n<\/div><\/div>\n<\/div>\n\n\n\n

<\/div>\n<\/div>\n\n\n\n
\n
<\/div>\n\n\n\n
\n
\n

3. Predictive maintenance<\/h3>\n\n\n\n

Of course, it is best if no faults occur in the first place. To this end, AI-sup\u00adport\u00aded sys\u00adtems could inde\u00adpen\u00addent\u00adly ana\u00adlyze the sen\u00adsor data for trans\u00adform\u00aders and oth\u00ader oper\u00adat\u00ading equip\u00adment and auto\u00admat\u00adi\u00adcal\u00adly report when main\u00adte\u00adnance is required, even before a fault occurs. AI could also plan ser\u00advice calls inde\u00adpen\u00addent\u00adly and thus make bet\u00adter use of scarce human resources.<\/p>\n<\/div><\/div>\n<\/div>\n\n\n\n

<\/div>\n<\/div>\n\n\n\n
\n
<\/div>\n\n\n\n
\n
\n

4. Fast error detection<\/h3>\n\n\n\n

Local\u00adiz\u00ading net\u00adwork faults is often com\u00adplex and time con\u00adsum\u00ading, although rapid inter\u00adven\u00adtion is required in such cas\u00ades to min\u00adi\u00admize sup\u00adply inter\u00adrup\u00adtions. AI can help to ana\u00adlyze net\u00adwork dis\u00adrup\u00adtions in real time and auto\u00admat\u00adi\u00adcal\u00adly rec\u00adom\u00admend actions to rec\u00adti\u00adfy them.<\/p>\n<\/div><\/div>\n<\/div>\n\n\n\n

<\/div>\n<\/div>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n
<\/div>\n\n\n\n

AI made by Reinhausen<\/h2>\n\n\n\n
\n\n\n\n

The fol\u00adlow\u00ading prod\u00aducts and solu\u00adtions with AI tech\u00adnol\u00ado\u00adgy are already avail\u00adable.<\/p>\n\n\n\n

\n
\n

Intelligent breathers<\/h3>\n\n\n\n
\n
<\/figure>\n<\/div>\n\n\n

MESSKO\u00ae<\/sup> MTRAB\u00ae<\/sup> 2.5 dehy\u00addrat\u00ading breathers<\/a> con\u00adtin\u00adu\u00adous\u00adly mon\u00adi\u00adtor mois\u00adture con\u00adtent, and a self-learn\u00ading algo\u00adrithm rec\u00adog\u00adnizes the breath\u00ading behav\u00adior of a trans\u00adformer to ensure that regen\u00ader\u00ada\u00adtion only takes place dur\u00ading the exha\u00adla\u00adtion phase, there\u00adby ensur\u00ading that mois\u00adture is removed to the out\u00adside and no mois\u00adture gets into the insu\u00adlat\u00ading oil.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n

Smart stethoscope<\/h3>\n\n\n\n
\n
<\/figure>\n<\/div>\n\n\n

MSENSE\u00ae<\/sup> VAM is a robust mea\u00adsure\u00adment sys\u00adtem<\/a> that ana\u00adlyzes the vibra\u00adtions that occur dur\u00ading the switch\u00ading process of an on-load tap-chang\u00ader. These vibra\u00adtions are then eval\u00adu\u00adat\u00aded using a dynam\u00adic and self-learn\u00ading lim\u00adit val\u00adue curve. MSENSE\u00ae<\/sup> VAM thus enables mechan\u00adi\u00adcal irreg\u00adu\u00adlar\u00adi\u00adties, time dif\u00adfer\u00adences in the switch\u00ading process or anom\u00adalies in the on-load tap-chang\u00ader to be detect\u00aded and report\u00aded.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n

Blood test for tap-changers<\/h3>\n\n\n\n
\n
<\/figure>\n<\/div>\n\n\n

Unlike con\u00adven\u00adtion\u00adal, often inac\u00adcu\u00adrate approach\u00ades to oil analy\u00adsis based on gas ratios and lim\u00adit val\u00adues, the MR solu\u00adtion for on-load tap-chang\u00aders uses math\u00ade\u00admat\u00adi\u00adcal-sta\u00adtis\u00adti\u00adcal algo\u00adrithms. In addi\u00adtion to gas con\u00adcen\u00adtra\u00adtions, meta\u00adda\u00adta such as the type, age and oper\u00adat\u00ading fre\u00adquen\u00adcy of the tap-chang\u00ader are also tak\u00aden into account. Com\u00adbined with MR\u2019s expert knowl\u00adedge, the result is a reli\u00adable analy\u00adsis of the insu\u00adlat\u00ading medi\u00adum.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n

Blood test for transformers<\/h3>\n\n\n\n
\n
<\/figure>\n<\/div>\n\n\n

Just as blood pro\u00advides infor\u00adma\u00adtion about a person\u2019s state of health, oil does the same for trans\u00adform\u00aders. The online oil analy\u00adsis devices in the MSENSE\u00ae<\/sup> DGA series<\/a> con\u00adtin\u00adu\u00adous\u00adly check for fault gas\u00ades and mois\u00adture in the insu\u00adla\u00adtion medi\u00adum. Using AI, a train\u00ading data set is used to deter\u00admine the cor\u00adre\u00adla\u00adtion between the actu\u00adal tar\u00adget val\u00adue, the sen\u00adsor sig\u00adnal and the fault. The advan\u00adtages: The design of the mea\u00adsur\u00ading sys\u00adtem is more cost-effec\u00adtive and less com\u00adplex.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n

Warning system for flashovers<\/h3>\n\n\n\n
\n
<\/figure>\n<\/div>\n\n\n

At what volt\u00adage does a flashover poten\u00adtial\u00adly occur in a trans\u00adformer? The BDV indi\u00adca\u00adtor in ETOS\u00ae<\/sup> mon\u00adi\u00adtors the dielec\u00adtric strength of the insu\u00adlat\u00ading oil<\/a> and warns when inter\u00adven\u00adtion is required. The break\u00addown volt\u00adage is main\u00adly influ\u00adenced by the rel\u00ada\u00adtive humid\u00adi\u00adty in the oil and is cal\u00adcu\u00adlat\u00aded using an AI-based mod\u00adel approach. With the help of oil-spe\u00adcif\u00adic para\u00adme\u00adters embed\u00added in the soft\u00adware, the absolute humid\u00adi\u00adty in ppm (mg\/kg) can be cal\u00adcu\u00adlat\u00aded.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n

Virtual transformer doctor<\/h3>\n\n\n\n
\n
<\/figure>\n<\/div>\n\n\n

ETOS\u00ae<\/sup> Asset Intel\u00adli\u00adgence ana\u00adlyzes all sen\u00adsor data of a pow\u00ader trans\u00adformer holis\u00adti\u00adcal\u00adly. A Bayesian net\u00adwork then checks which error pat\u00adterns best match the error pat\u00adterns that have occurred and also not occurred. In addi\u00adtion, the a pri\u00adori prob\u00ada\u00adbil\u00adi\u00adties for typ\u00adi\u00adcal trans\u00adformer faults and the accu\u00adra\u00adcy of the sen\u00adsors are tak\u00aden into account. The sys\u00adtem then pro\u00advides a prob\u00ada\u00adbil\u00adi\u00adty esti\u00admate for all known trans\u00adformer prob\u00adlems and sup\u00adports spe\u00adcial\u00adists in mak\u00ading quick and clear diag\u00adnoses.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n

Intelligentes Flottenmanagement<\/h3>\n\n\n\n
\n
<\/figure>\n<\/div>\n\n\n

The TESSA\u00ae<\/sup> APM asset man\u00adage\u00adment soft\u00adware<\/a> sup\u00adports the plan\u00adning of main\u00adte\u00adnance mea\u00adsures and invest\u00adments. It is based on online data from ETOS\u00ae<\/sup> as well as offline data, such as that col\u00adlect\u00aded dur\u00ading on-site inspec\u00adtions. The intel\u00adli\u00adgent soft\u00adware eval\u00adu\u00adates the data, rec\u00adog\u00adnizes trends and makes rec\u00adom\u00admen\u00adda\u00adtions for action. AI also helps in this regard.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n

Innovative cable monitoring<\/h3>\n\n\n\n
\n
<\/figure>\n<\/div>\n\n\n

Intel\u00adli\u00adgent algo\u00adrithms are also used in HiMON\u00ae<\/sup> \u2014 a mod\u00adu\u00adlar mea\u00adsure\u00adment and con\u00addi\u00adtion assess\u00adment sys\u00adtem<\/a> that the Rein\u00adhausen sub\u00adsidiary HIGHVOLT has devel\u00adoped for rapid local\u00adiza\u00adtion of cable break\u00addowns and detec\u00adtion of par\u00adtial dis\u00adcharges.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n

Smart production<\/h3>\n\n\n\n
\n

Rein\u00adhausen is increas\u00ading\u00adly rely\u00ading on AI not only in the prod\u00aducts them\u00adselves, but also in its pro\u00adduc\u00adtion process\u00ades where it is used as a tool to opti\u00admize pro\u00adduc\u00adtion process\u00ades and short\u00aden deliv\u00adery times.<\/p>\n<\/div><\/div>\n<\/div><\/div>\n<\/div><\/div>\n\n\n\n

\n
\n\n\n\n

YOUR CONTACT PERSON<\/h2>\n\n\n\n

Do you habe any ques\u00adtions about arti\u00adfi\u00adcial intel\u00adli\u00adgence at Rein\u00adhausen?<\/strong>
Roland G\u00f6tz is here for you:
<\/i>\u2009 R.Goetz@reinhausen.com<\/a><\/p>\n<\/div><\/div>\n","protected":false},"excerpt":{"rendered":"

Arti\u00adfi\u00adcial intel\u00adli\u00adgence is a key tech\u00adnol\u00ado\u00adgy that is also increas\u00ading\u00adly find\u00ading its way into the ener\u00adgy sec\u00adtor. Roland G\u00f6tz, Head of Inno\u00adva\u00adtion and Tech\u00adnol\u00ado\u00adgy at Maschi\u00adnen\u00adfab\u00adrik Rein\u00adhausen (MR), explains what con\u00adtri\u00adbu\u00adtion the Rein\u00adhausen group is already mak\u00ading and what oppor\u00adtu\u00adni\u00adties arise for grid oper\u00ada\u00adtors from com\u00adpre\u00adhen\u00adsive data col\u00adlec\u00adtion and the use of AI.<\/p>\n","protected":false},"author":4,"featured_media":13706,"comment_status":"closed","ping_status":"open","sticky":false,"template":"template-cover.php","format":"standard","meta":{"_acf_changed":false,"wp_typography_post_enhancements_disabled":false,"footnotes":""},"categories":[44,1],"tags":[],"class_list":["post-13514","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-07-2025","category-onload"],"acf":[],"_links":{"self":[{"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/posts\/13514","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/comments?post=13514"}],"version-history":[{"count":1,"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/posts\/13514\/revisions"}],"predecessor-version":[{"id":14624,"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/posts\/13514\/revisions\/14624"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/media\/13706"}],"wp:attachment":[{"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/media?parent=13514"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/categories?post=13514"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/onload.reinhausen.com\/en\/wp-json\/wp\/v2\/tags?post=13514"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}