不需電池　能量採集技術魅力足

能量採集(energy harvesting)技術的發展至少已有20年的歷史，適合應用在建築物監控、量測工具，以及安裝在不容易拆卸之處、專門監測建築物結構狀態等方面的所謂「永久性裝置(perpetual devices)」。由於應用廣泛，該技術也逐漸受到產業界的重視。

那些永久性裝置的關鍵特性，就是他們能無限期執行任務，而且使用者不用擔心換電池的問題。在美國加州嵌入式系統會議(Embedded Systems Conference)期間的一場能量採集技術座談會上，與會者針對二十年來該技術的發展歷程分享了不少意見；多數人都認為，無線與低功耗電子元件的發展，以及感測器、MEMS等元件的演進，是對今日能量採集技術造成基礎性變革的因素。

德州儀器(TI)的工程經理Dave Freeman表示，低功耗電子元件與無線技術的發展，讓能量採集技術打開在許多應用領域的「能見度」。伊利諾大學(University of Illinois)副教授Patrick Chapman也同意，由於無線通訊元件的功耗越來越低，是讓小型能量採集裝置終於找到「用武之地」的主要因素。

自動化解決方案供應商EnOcean的應用工程師Eugene You補充指出，無線元件的成本越來越低、讓永久性裝置的應用越來越廣泛，這也是量採集技術日益蓬勃發展的原因。電池製造商Cymbet的行銷副總裁Steve Grady還表示，使用能量採集裝置也會需要電力儲存設備，例如該公司正在開發的一種長壽命薄膜電池。

但無論是哪一種新技術，都需要有大量需求來推波助欄，才能降低技術的成本，並使該技術成為廣泛應用的主流；而能量採集技術的「殺手級應用」為何？

MEMS技術專家Perpetuum的業務開發總監Keith Abate認為，答案是無線應用；TI的Freeman則指出，許多不同應用的「無線感測網路」將會推動能量採集技術的普及。而Cymbet的Grady將智慧建築(smart buildings)與智慧監測(intelligent monitoring)方案視為能量採集技術的殺手級應用：「我認為各種經濟振興方案將有助於增加能量採集技術的應用。」

伊利諾大學的Chapman透露，最近的技術新進展讓MEMS感測器未來可能變成具備能量採集技術的元件；而他認為，可能需要植入人體的醫療感測器應用會是上述這類能量採集元件的殺手級應用。而能量採集技術應用的「聖杯」，也就是該技術與無線元件或感測器的結合。

透過能量採集裝置的協助，一套系統能收集/處理資訊、或是傳送並分析其內容；而這類資訊的共通性也是一個需要關注的議題。對此Cymbet的Grady則認為，在該領域使用網際網路通訊協定(IP)是合適的，如此一來，收集自不同無線感測網路的資料就能整合在一起。

不過使用IP的問題在於，這對無線元件的耗電量太大，此外如果使用IP，就需要為成千上萬個感測器都指定一個IP位址。而針對各種無線感測網路開發標準通訊協定的工作，目前還僅在起步階段。

Grady表示，無線感測網路可使用的技術，包括Wi-Fi、藍牙(Bluetooth)、Zigbee等等；但若缺乏共通性，恐怕難以形成真正的能量採集殺手級應用。

(參考原文：Advancements in wireless, sensors and MEMs are set to change energy harvesting ，by Junko Yoshida)

SAN JOSE, Calif. — Technologies that harvest or scavenge energy have been in development for the last 20 years. They enable so-called "perpetual devices" to monitor buildings, to gauge tools and machines installed in inaccessible places, or to sense structural integrity or movement.
The key to such perpetual devices is that they perform these tasks indefinitely, thereby eliminating the need to replace batteries. Energy harvesting is increasingly becoming a hot topic. But "what has really changed energy harvesting in the last two decades?" asked Patrick Mannion, editorial director of TechOnline, who moderated the energy harvesting panel at Embedded Systems Conference on Tuesday (March 31).

The answer, in sum, is a confluence of the development of wireless and lower power electronics, along with advancements in sensors and MEMS. The result is a combination of factors that is fundamentally changing today's energy-harvesting market.

The development of low-power electronics and wireless technologies made it "viable" to use harvested energy for a number of applications, said Dave Freeman, engineering manager at Texas Instruments.

Patrick Chapman, associate professor at University of Illinois, agreed. "The power of wireless communication devices has become so low that we can finally start doing something about small, harvested energy," he said.

Eugene You, application engineering manager at EnOcean, added, "Don't forget the price of wireless devices." With a steady decrease in the cost of wireless devices, he said, "We can now put [perpetual devices] everywhere for sending data wirelessly, and forget about it."

Nonetheless, you still need a power storage device, said Steve Grady, vice president of marketing at Cymbet Corp. Some systems are required to occasionally handle a large burst of energy. If you can store energy, such as in a thin-film battery Cymbet is developing, "you can respond to it faster," Grady added. And that energy storage device "needs to hold charge long-term."

Every new technology often needs a hot mass market that can help lower the cost of the technology and push it to a broadly penetrated mainstream technology. What's the killer app for energy harvesting technology?
"Wireless is the killer app for energy harvesting," said Keith Abate, director of business development, at Perpetuum.

TI's Freeman agreed. "Wireless sensor networks " used in so many different applications " will drive the acceptance of energy harvesting," he explained.

Cymbet's Grady chooses smart buildings and intelligent monitoring as killer apps. "I see the stimulus package helping to create a lot of access to energy harvesting," he added.

A new wrinkle is that MEMS sensors themselves may become energy harvesting devices, said the University of Illinois' Chapman. Further, "Things like medical sensor devices that could be flown into blood could be also killer apps, for example," he added.

The Holy Grail for energy harvesting applications lies in the use of the technology when combined with wireless and sensors.

With the help of energy harvesting devices, one can collect data, process it, transport it and analyze it. What's lacking is interoperability among such data.

The use of Internet Protocols is ideal for that purpose, said Cymbet's Grady, so that data collected from different wireless sensor networks can be pulled together. The problem of using IP is that it's power-hungry when used in wireless devices. Further, if using IPs, one needs a whole new address space for billions of sensors.

Developing the standard protocol for a variety of wireless sensor networks is still "a few years off," said Grady.

One can use a variety of wireless technologies, including WiFi, Bluetooth, Zigbee or others. But without interoperability, a killer wireless/sensor application for energy is hardly a sure thing, Grady acknowledged.