2017年1月25日 星期三

自動車出事,罰誰

自動車出事,罰誰
20170125
吳景欽

美國國家交通安全管理局(NHTSA),對去年發生於佛羅里達州的特斯拉自動駕駛事故,做出初步調查認為,原因並非在系統出問題,而是駕駛者未能專注所致。惟若所謂自動車或無人駕駛車,仍須駕駛者時時注意,果真能叫自動駕駛嗎?更值關注的是,面對此等科技產物與趨勢,法制又該如何調整?

關於自動車最須被面對的課題,即是發生事故時法律責任的歸屬。傳統上,對交通安全的規範乃是以駕駛人的過失為歸責核心。故於車禍時,就刑事責任來說,基於罪止一身原則,就只能以肇事者為處罰對象。而就民事責任來說,除駕駛人外,雖可能擴及於法律先推定有過失的雇用人,甚至是商品製造人,但若其能舉證自己無過失,仍可免除賠償責任。至於汽車所有人,除非其同時為雇主或製造者,否則,依現行法制,車輛提供者並無須擔負賠償責任。

而目前汽車現況,雖已有定速、導航、自動煞車與停車等等設計,但因仍以人力操控為主導,尚不能納入自動駕駛的領域。故所謂自動車,應是指加速、減速、操控全由電腦運作,至於駕駛人僅能在緊急狀態下介入,甚或根本不讓人力有操縱可能之無人車。

所以於自動車仍有駕駛人在其內的情況,由於人力可隨時操控,故當事故發生時,依然可適用現有法制為處理。只是於此時,人為操控已變成是輔助,就刑事究責,固然仍應以駕駛人為歸責重心,但就民事責任來說,是否該同時提高商品製造者的注意義務與責任,即是法規範必須重新思考之處。

比較麻煩者,還是無人駕駛的自動車。由於此種型態,已經脫離自然人的掌控,則當事故發生時,刑罰對象到底要歸屬於車輛製造者、所有人,抑或是程式設計者,就會產生疑問,也勢必會對所謂行為責任原則,造成極大的衝擊。至於民事部分,目前以駕駛人為歸責中心的體系,亦將無用武之地,致須重行建立以製造者為究責核心且採無過失責任的規範結構。至於車輛提供者、程式設計者,是否亦該納入連帶賠償的範圍,更得有不同以往的思維與立法模式。

雖然對自動車的規範,欲設計一套完整的法制,實非一蹴可幾,這對自動車發展與研究相對落後的我國而言,更是如此。不過至少於現階段,主管部門得趕緊制訂實驗車上路的規範與準則,此可能包括自動車上路前的檢測標準、是否要有駕駛人在其內及需要何種資格、可行走的路段與時間,甚至是責任險的最低額等等。畢竟,法律的修改與規範完整性,雖肯定無法趕得上社會的進步與進化,卻也不能成為科技發展的絆腳石。(作者為真理大學法律系副教授兼系所主任)


(中國時報)

2017年1月14日 星期六

A Big Test for Big Batteries

A Big Test for Big Batteries
By DIANE CARDWELL and CLIFFORD KRAUSS
JAN. 14, 2017 ESCONDIDO, Calif. —

In Southern California in the fall of 2015, a giant natural gas leak not only caused one of the worst environmental disasters in the nation’s history, it also knocked out a critical fuel source for regional power plants. Energy regulators needed a quick fix. But rather than sticking with gas, they turned to a technology more closely associated with flashlights: batteries. They freed up the utilities to start installing batteries — and lots of them.

It is a solution that’s audacious and risky. The idea is that the batteries can store electricity during daylight hours (when the state’s many solar panels are flooding the grid with power), then release it as demand peaks (early evening, when people get home). In effect, the rechargeable batteries are like an on-demand power plant, and, in theory, able to replace an actual plant. Utilities have been studying batteries nationwide. But none have moved ahead with the gusto of those in Southern California.

This idea has far­-reaching potential. But the challenge of storing electricity has vexed engineers, researchers, policy makers and entrepreneurs for centuries. Even as countless technologies have raced ahead, batteries haven’t yet fulfilled their promise. And the most powerful new designs come with their own risks, such as fire or explosion if poorly made or maintained. It’s the same problem that forced Samsung to recall 2.5 million Galaxy Note 7 smartphones in September because of fire risk.

After racing for months, engineers here in California have brought three energy-storage sites close to completion to begin serving the Southern California electric grid within the next month. They are made up of thousands of oversize versions of the lithium­-ion batteries now widely used in smartphones, laptop computers and other digital devices. One of the installations, at a San Diego Gas & Electric operations center surrounded by industrial parks in Escondido, Calif., 30 miles north of San Diego, will be the largest of its kind in the world, developers say. It represents the most crucial test yet of an energy-­storage technology that many experts see as fundamental to a clean­-energy future.

Here, about 130 miles southeast of Aliso Canyon, the site of the immense gas leak in 2015 — the global­-warming equivalent of operating about 1.7 million cars over the course of a year — 19,000 battery modules the size of a kitchen drawer are being wired together in racks. They will operate out of two dozen beige, 640­ square­-foot trailers. Made by Samsung, the batteries are meant to store enough energy to serve as a backup in cases of fuel shortages. They are also designed to absorb low­-cost energy, particularly solar power, during the day and feed it back to the grid after dusk. They in effect can fill in for the decades-­old gas­-fired plants that might lack the fuel to fully operate because of the disastrous leak.

“California is giving batteries the opportunity to show what they can do,” said Andrés Gluski, chief executive of AES, which is installing the storage systems. AES is installing a smaller array for the electric utility in El Cajon, a suburb east of San Diego. And separately Tesla, the company perhaps better known for its electric cars, is building an array for a different utility on the grid, Southern California Edison, near Chino, Calif.

The stakes are high for both energy storage companies. If their projects struggle or fail, it could jeopardize not only the stability of Southern California’s grid but also interest in the technology over all. After a smaller, but pioneering battery project at a wind farm on Oahu in Hawaii went up in flames in 2012, investment in battery storage all but dried up for a few years. That installation, which used 12,000 lead­-acid batteries to help even out fluctuations in the power flow, caught fire three times in its first 18 months of operation. The storage developer, Xtreme, eventually went bankrupt.

The wind farm turned to a different technology to smooth its output. Keeping a close eye on the Southern California battery efforts is Susan Kennedy, who helped shepherd California’s energy policy for more than a decade as a state utility regulator and high-­level operative for two governors — Gray Davis, a Democrat, and Arnold Schwarzenegger, a Republican. She now runs an energy storage start­up, one not involved in the battery­-building response to the Aliso Canyon gas leak.

“The moment one fails,” Ms. Kennedy said of the big bet on batteries, “they won’t build any more.” As soon as AES’s chief executive, Mr. Gluski, learned last June that San Diego Gas & Electric had awarded AES the big battery contract, he leapt out of his chair and interrupted a meeting in his board room at the company’s headquarters in Alexandria, Va. As employees watched in astonishment, he barreled down two flights of stairs, grabbed a mallet and, with a ceremonial flourish, banged a gong that one of his executives kept on hand for big news.

Mr. Gluski had not had much occasion to celebrate since he had taken the AES reins five years earlier. The company was struggling with debt and trying to coax profits from far-­flung fossil-­fuel projects around the developing world that are buffeted by instability in politics, currency and commodity prices.

His first steps included an austerity program in which he gave up many of his own executive perks: No more country­-club membership. No more corporate Audi A8, with driver. But the more far-­reaching part of his plan would be AES’s battery division, which was then fledgling. The unit had roots in two midlevel executives who had been speculating about a Jetsons-­like future over beers.

Those two men, John Zahurancik, a science fiction fan, and Chris Shelton, a former physics teacher, had started talking about batteries a decade ago, before electric cars became fashionable or even feasible.

In 2006, Mr. Shelton had come across a professor’s paper that predicted a future dominated by electric cars that, when parked, could be connected to the power grid so their batteries could act as storage devices to help balance electricity demand.

He and Mr. Zahurancik bounced the idea off some AES colleagues, who said it was at least theoretically feasible. So the two continued their bull sessions but decided that stationary battery arrays may make more sense than relying on electric cars. At the time, lithium-­based batteries, the standard in consumer products, were widely in use in the transportation and power tool industries, but no one had paired them with the technology necessary to serve the power grid.

Earlier grid­-scale experiments with lead­-acid and other types of batteries worked only for a year or two before conking out. A different technology, “flow batteries,” which use chemicals dissolved in liquids in tanks, were considered even more experimental. But lithium packs more energy per weight than other metals, offering the promise of greater energy density and longevity. The trick would be to figure out how to harness all that power, which creates heat, while avoiding the fires such batteries have caused in any number of vehicles and gadgets, including Teslas, HP computers, hover-boards and, most recently, Samsung Galaxy Note 7s.

The two hit upon a design and persuaded executives to begin a pilot project in 2008. That eventually led to the first commercial lithium-­ion battery on a grid. Mr. Zahurancik, who owns the gong, is now president of AES Energy Storage. Mr. Shelton is now the company’s vice president and chief technology officer. AES does not actually make its batteries but buys them, along with other equipment, from manufacturers like Samsung, LG Chem and Panasonic. It designs and assembles the arrays, stacking the boxy batteries into racks inside locker-­like containers.

In Escondido, where local radio stations still carry public service announcements about the natural­-gas shortage, the AES battery packs are being installed at a critical spot on the regional electrical grid: the place where the giant wires from power plants and wind and solar arrays connect to the network of local wires. The batteries are intended to relieve the pressure on the system. Mainly, they will serve as a kind of sponge, soaking up excess or low-­cost solar energy during the day and then squeezing it back into the grid in the evening, when demand surges as the sun sets.

There is enough capacity in the containers full of batteries to power about 20,000 homes for four hours. The idea is that they help the utility lessen its dependence on the type of natural gas plants known as “peakers,” which can turn on and off quickly to meet sudden peaks of demand but are generally used only for short periods and at great expense. And peakers, by burning fossil fuel, are also at odds with California’s green-­energy goals. The project is also being watched closely by advocates for renewable energy. The reason: If utility-­scale battery installations work as designed, they would help wind or solar generators to act more like conventional power plants by working steadily even when the sun isn’t shining or the wind isn’t blowing.

“Energy storage is really the tool to do renewables integration for a utility infrastructure company like us,” said Josh Gerber, advanced technology integration manager of SDG&E, as workers smoothed the thigh­-high concrete pads that support the containers at the Escondido site. “Without it, you have more risk that the variability of renewables is going to cause reliability problems.”

Under the contract, AES is responsible for making sure the batteries perform for 10 years, after which SDG&E will take over. One potential downside is that if the batteries are fully charged and discharged each day, they could degrade more quickly. The executives involved expressed confidence in the design and reliability, despite Samsung’s recent smartphone problems.

Not only are these batteries a different configuration than the smartphone units, executives said, but the larger footprint allows for the inclusion of sophisticated monitoring as well as industrial safety and cooling and ventilation equipment. The project, along with the smaller array AES has installed in El Cajon, could provide the proof­-of­-concept leap Mr. Gluski has been striving for.

AES has a deal for an even bigger, $1 billion project in Long Beach with Southern California Edison that is not part of the Aliso Canyon remediation effort; it is projected to go online by the end of 2020. The electric company plans to use batteries to replace part of its aging gas plant along the San Gabriel River.

Long term, Mr. Gluski plans to shift the company’s power­-generation portfolio — still heavily based on coal and natural gas — toward more renewable energy. He sees the storage systems as vital components in turning solar and wind energy into a dominant power source in the parts of Latin America, Asia and Africa where AES is active.

Whatever progress it has made, AES still has its share of problems, with $20 billion in debt and a stock price less than one-­fifth the value that it had at the start of the century. It faces wary, if not outright skeptical, treatment by Wall Street utility analysts and energy experts, who say the technology AES is peddling on such a large scale in California remains untested and financially risky.

“The problem comes if there is a hiccup with the battery storage business in California,” said Charles Fishman, a utilities analyst at Morningstar. “You don’t have the deep-­pocket parent that can push money to it and keep it out of trouble.” Despite all the battery activity in California, executives around the utility industry remain cautious.

“The reason we don’t have widespread batteries on our system is because it is not cost-­effective for us,” said Alice Jackson, vice president for strategic revenue initiatives at Xcel, a giant electricity and gas utility serving eight Western and Midwestern states. Xcel has been testing batteries about as long as AES, but almost exclusively in small pilots. “It’s fair to say we don’t have long­-range experience with this technology to say that it is perfect, or a nirvana,” Ms. Jackson said. “It’s something we’ll observe as California goes through its experience.”

California’s latest experiment with batteries is but the latest bout in the state’s long struggle to match its energy needs with its environmental sensibilities. In the early 2000s, after market deregulation and Enron’s notorious manipulation of gas supplies led to blackouts and financial instability among the power companies, state officials decided to lessen reliance on natural gas by encouraging the development of wind and solar. Under Mr. Schwarzenegger, who was governor until 2011, officials pushed through a raft of overlapping regulations that created a boom in renewables, especially solar. But that upended the traditional patterns of supply and demand, making the overall energy system technically and economically difficult to manage.

Batteries were the logical solution. But the technology wasn’t fully developed and was still too expensive. In order for companies to make the necessary investments, they needed a signal that there would be a big enough market for their products. So in 2010, the state approved one of the first energy-­storage mandates, ultimately requiring utilities to install some form of storage equipment in their territories. That set off a flurry of new investment and innovation and, after the sudden closure of the San Onofre nuclear plant on the coast in northwest San Diego County in 2012 when a steam generator tube sprung a leak, new contracts for battery installations.

But the Aliso Canyon accident, which began on Oct. 23, 2015, when the Southern California Gas Company first detected the leak, put that process on fast-forward. The noxious­-smelling gas and intermittent oily mist that spewed forth over almost four months traveled into the surrounding neighborhoods on the strong winds that sweep down from the Santa Susana Mountains. At the same time, it forced the battery strategy into its most high-­profile test yet. Now it’s show-time, and the pressure to succeed is high all around.

For AES, it could signify an important step for a long­-troubled conventional­-energy relic that is seeking to revitalize itself as a powerhouse in battery storage and other advanced technologies. For clean­-energy advocates — including residents of the Porter Ranch section of Los Angeles, so picture­-perfect that Steven Spielberg chose it as the setting for the 1982 movie “E.T.,” but where many still complain of the rashes, headaches and other debilitating symptoms that drove thousands from their homes during the leak — it could be a powerful weapon in the fight to keep the gas depot closed.

But the pressure may be highest for the Southern California utilities, their reputations still blackened by waves of forced electricity cuts that followed the Enron debacle. No one wants to contemplate a repeat of that chapter, when blackouts affected factories and even some hospitals. “When the power goes out, people die,” said Ms. Kennedy, the former state official. “Failure is not an option here on any level.”


2017年1月9日 星期一

三十年後,線上經濟體當道

三十年後,線上經濟體當道
20170110
林建甫

17日我應邀到北京參加第二屆新經濟智庫大會。該會選在798藝術區舉行,以「遠見2046」為主題,聚焦「線上經濟、未來科技」,討論新技術、新經濟和新治理三大問題。這三大問題彼此互相連結,因為新技術帶來新經濟,新經濟呼喚新治理模式。參加者是來自中外的120多家智庫的800多位負責人。大家齊聚一堂,是一場知識界未來學的饗宴。

798藝術區位於北京朝陽區酒仙橋街道大山子地區。原為大陸國營事業798廠等電子工業的老廠區所在地。它們是50年代初由蘇聯援建、東德負責設計建造的重點工業項目,幾十年來經歷了無數的風雨滄桑。後來藝術家和文化機構進駐,經過改造逐漸發展成為畫廊、藝術中心、餐飲酒吧等各種空間的聚合,成為了北京都市文化的新地標。大會選在大煤氣儲槽圓球中開會,有古老的房子,新式的音響投影設備。外面零度以下小雪紛飛,裡面大夥熱烈討論,別有風味。

其實從2008年金融危機以後,整個全球經濟都進入到低迷、遲疑當中,IMF總裁拉加德因此認為這是「新平庸經濟」,大家對如何走出、突破困境都看不清楚。然而阿里巴巴集團認為未來30年是人類最關鍵的30年。大會主題的設計就是跳脫對當今的糾結,直接討論30年後的2046,再來反思當下。其董事局主席馬雲曾多次表示:「新技術將融合到傳統行業的方方面面,引領人類社會天翻地覆的變化。」「未來30年一定不會只是互聯網公司的天下,未來30年是『用好互聯網技術』的公司或年輕人的天下。」

2046將是什麼樣子?估計那就是「奇點」(singularity)的降臨。「奇點」本是天體物理學術語,是指「時空中的一個普通物理規則不適用的點」。在美國未來學家雷·庫魯韋爾(Ray Kuraweil)的理論中,「奇點」是指人工智慧超過人類智慧,機器也可敏捷地思考、行動和溝通,人類就會與機器相互融合。屆時互聯網、物聯網、區塊鏈、基因技術、認知模式識別、雲計算、大數據等技術都成熟,改變無處不在。藉此遠見,洞察未來,才能判斷現在。

卡內基美隆大學賈斯汀·卡塞爾(Justine Cassel)教授,她也是世界經濟論壇人工智慧委員會主席,在大會報告「人機互動的未來」。她認為下一代人機溝通介面具有兩個屬性:通過對話和肢體語言進行自然互動;採用人類的社交模式而非驅動機器的任務模式。未來最大的突破會是,語言的使用讓介面變得更加直觀、自然。這主要依靠深度學習(deep learning)演算法和自然語言來實現。卡塞爾也認為,未來的人機互動會以人為主,如此一來就可以將語言、行為、表情各種表現納入設計。機器從真實的人群中得到諮詢建議並不斷反覆運算,直到互動設計的實用化、理想化來「讓人信服」。

對於技術如何發展?商業如何演化?治理需如何創新?新的需求環環相扣,就需要智庫思想者來領航,來解決問題。因此大會隆重舉行了中國大陸「遠見智庫論『潭』」成立儀式,由清華大學公共管理學院、中國社會科學院經濟所、北京大學光華管理學院、國研中心資環所、華為戰略部、京東方產業研究中心、資訊社會50人論壇、華夏基石諮詢公司、競爭力智庫、阿里研究院等10家共同發起。將這一「論潭」打造成匯聚原創性技術思想、開創性商業模式,以及將創意設想轉化為創新價值的正能量中心與國際化平台。

對於新的技術引發新的商業革命,也將孕育新的經濟形態。檢討第一次工業革命,以蒸汽機為代表,使用機器,產生了「工場」;第二次工業革命,以電力為代表,使用生產線,產生了「工廠」;而第三次工業革命,以電腦、互聯網為標誌,產生了「公司」。未來新的經濟生態是什麼?阿里研究院在大會發布《線上經濟體發展報告》,首次提出「線上經濟體」概念,線上的所有數據變成最具共享性的生產資料,計算變成隨時隨地的公共服務,互聯網變成最具滲透力的基礎設施。數據改變商業本質,計算重塑經濟活動。「線上經濟體」是經濟生態組織演化的高級形態,是未來全球經濟的主要生態體系。(作者為台灣經濟研究院院長、台灣大學經濟系教授)


(中國時報)