因應國際能源轉型政策，未來石油產品市場需求將逐漸趨緩，而石化產品市場需求則持續成長，因此，煉油工業也要積極面對未來市場的巨變。2014 年埃克森美孚公司在新加坡正式啟用了世界上第一套年產能100 萬噸乙烯的工廠，採用原油直接製造烯烴新製程的煉廠，該廠越過原油裂解為石油腦過程，將原油直接轉化為輕質烯烴。根據IHS 諮詢報告，原油直接製造烯烴技術相較傳統具有製程與成本優勢。亞洲現有四個COTC（Crude Oilto Chemicals）專案正推動中，包括中國的三個和汶萊的一個煉製結構被配置為生產最大量的對二甲苯。沙烏地Aramco/SABIC 公司也開發了自有的原油直接製造烯烴化學品煉化廠，但尚未興建完成商業化營運。
　　中國和沙烏地阿拉伯計畫啟動的幾個原油-化學品專案計畫，有可能在未來幾年重塑全球石油化工產業。煉油廠配置COTC 新技術架構，將以生產化學品為最大化而不是傳統的運輸燃料。台灣石油工業面對國際情勢及石化市場改變，但同時也受限於國內土地利用、環保法規等因素限制，除洞悉國際石化市場變化趨勢外，並考量國內現實環境，政府應研議提出具體可行的轉型長期策略建議。

　　In response to international energy transition policies, market demand for petroleum products is expected to ease, while demand for petrochemical products will continue to grow. Therefore, the refining industry must be prepared to respond to rapidly changing market conditions . In 2014, ExxonMobil commissioned a world-scale facility in Singapore that produces 1 million tons per year of ethylene directly from crude oil, bypassing the traditional naphtha cracking process. According to the IHS Chemical Process Economics Report, this new crude to olefins process provides cost advantages over traditional production processes. There are currently four COTC projects in Asia, including three in China and one in Brunei, with refinery structures configured to produce maximum volumes of paraxylene. Saudi Aramco/SABIC has also developed its own crude to olefins process, but facilities for commercial operation are currently still under construction.
　　Several crude oil to chemical projects initiated by China and Saudi Arabia have the potential to reshape the global petrochemical industry in the coming years. COTCs configure a refinery to produce maximum chemicals instead of traditional transportation fuels. 
　　The Taiwan oil industry is not only faced with changes in the international environment and the petrochemical market, but is also restricted by domestic land use and environmental protection regulations. In addition to understanding the dynamics of both the international petrochemical market and the actual domestic environment, the government should also propose long-term strategic recommendations that can ensure the success of energy transition.

　　人工智慧在近幾年快速發展並廣泛應用於各領域，石油工業也正處這波浪潮上，特別是高風險的油氣探勘，開始利用自動化、資訊化和整合化的系統性評估，來降低鑽井與探勘風險。本文主要彙集人工智慧在石油產業上、中與下游的應用，並透過近年來國際大型石油公司的合作案例，從智慧機器人的海底漏油探測、虛擬AI 客戶銷售助手、智慧製造平台和自動機器人等，甚至是近期中石油發表的探勘開發夢想雲，以及台灣中油公司在尼日合資A 礦區的數位化油田先導試驗等案例，探討未來可能發展方向，從近程的管線自動化監控與分析和油氣田生產數位化遠程監控與分析，並延伸至夢想雲自動探勘、鑽井與大數據分析等，相信都能讓油氣能源產業邁入新世紀。

　　In recent years, artificial intelligence has developed rapidly and extensively in various fields and oil/ gas industry is also on this. Especially for high-risk oil and gas exploration, it has already introduced the systematic evaluation of automation, information technology, and integration in order to reduce drilling and exploration risks.
　　This paper is mainly to collect the artificial intelligence application of the upstream, midstream and downstream in the oil/gas industry, and through the actual cases of international oil companies in recent years, such as submarine oil spill detection of smart robots, virtual AI customer assistants, smart manufacturing platforms, and automatic robots, etc. Especially for the dream cloud of exploration and development which announced by CNPC in November 2018 and the pilot test of the digital oil field of CPC Corporation, Taiwan in joint venture A field, Niger.
　　Based on recent cases and experiences to discuss possible future development trends, such as the process of the pipeline automation monitoring and analysis, remote monitoring and analysis of oil and gas field production, and remote dream cloud automatic exploration and drilling. It is believed that the oil and gas energy industry can enter the new century in the near future.

　　台灣石化產業面對環評及空污法規範衝擊，2017 年台塑集團擴建案與台灣中油公司高值化投資案因此受阻，空污法對排放總量的管制與削減，也讓國內產能擴充不易；台灣石化業外銷以中國為主，但中國正積極發展石化產業，對台灣形成進口替代威脅。
　　此外，各國政府將於2040 年停產汽、柴油車政策，電動車將進入快速成長期，對石油市場帶來新的變數。國際海事組織（International Maritime Organization,IMO）為大幅抑制全球船舶產生的污染，自2020 年1 月1 日起實施新排放標準，將船舶燃油的硫含量從目前3.5％調降至0.5％，此新規定是能源與航運市場有史以來最大改變。
　　過去十年來，隨著頁岩油開採技術不斷進化，美國石油產出翻了兩倍有餘，依據EIA數據顯示，美國2018 年6 至8 月石油產能持續攀升，每日原油產出逼近1,100 萬桶，已經超越俄羅斯，對油品市場影響甚大。
　　油品市場結構會因電動車興起、頁岩油氣的增產及IMO 2020 船舶燃油硫含量降低等因素而調整，煉廠需因應市場變化進行轉型。

　　Taiwan petrochemical Industry facing the impact of EIA and air pollution law, the expansion of Formosa Petrochemical Corporation in 2017 and the high-value investment project of CPC were restricted and limited. The constraint and reduction of total emissions by air pollution law also made it more difficult to expand domestic production capacity. The product from Taiwan petrochemical industry is mainly exported to China, but China is actively developing petrochemical industry to meet its domestic requirement instead of being imported from Taiwan and also threaten Taiwan petrochemical Industry. In addition, many countries announce their policy to shut down the factories to produce the gasoline and diesel vehicles in 2040, and electric vehicles will be developing rapidly and bringing a profound impact on the oil market. The International Maritime Organization (Maritime Organization, IMO) has been implementing new emission standards since January 1, 2020 to significantly curb pollution from ships worldwide, reducing the sulfur content of ship fuel from its current 3.5% to 0.5%. The new rule is the biggest change in the energy and shipping markets ever.
　　Over the past decade, as shale oil mining technology has evolved, U.S. oil output has doubled by more than twice times. According to EIA data, U.S. oil production capacity continued to climb up from June to August in 2018, and daily crude oil output approached 11 million barrels surpassing Russia to impact greatly on the oil market.
　　To face the oil market structure adjusted by the rise of electric vehicles, the significant production increase in shale oil and gas and the sharp reduction of sulfur content of bunker by IMO 2020, the refinery production configuration needs to be upgraded and transformed to meet the tremendous changes for the future market.

　　石化廠中為了存放製程上所需的各種原物料乃藉由建立球槽區將不同的原物料存放於各個球槽內部，經由管線來傳輸原物料至製程上各個單位，而球槽區管線大多採用保溫包覆於管線外部，因此管線包覆層下腐蝕(Corrosion under Insulation, CUI)為主要造成球槽區管線腐蝕、破管之原因；球槽區為了防止內容物洩漏造成重大損失，在每座球槽四周都以混凝土牆做為防溢堤，管線在球槽區間傳輸時便須穿越1.3 公尺長的防溢堤，而防溢堤混凝土若有劣化時便會造成管線腐蝕的另外一項因素。本文乃以導波檢測法與數位輪廓射線照相進行球槽區包覆管線之CUI 檢測，導波可檢測出主管上腐蝕包括局部腐蝕、均勻腐蝕與管支撐腐蝕，輪廓射線照相則負責檢出小插管與主管連接點的腐蝕程度；再以導波檢測法搭配定量電磁超音波針對防溢堤管線進行腐蝕檢測，導波可為防溢堤管線進行腐蝕快篩，評估出那一條管線具有較高之腐蝕風險，搭配電磁超音波可進行管線穿越防溢堤界面腐蝕之深度定量評估，綜合此三項非破壞性檢測技術，在拆除最少的包覆材料與不用開挖防溢堤的條件下，可為球槽區管線安全的管理提供一項有效性檢測方案。

　　In the petrochemical plant, the various spherical tank area are established to store the various raw materials required for the process. Pipelines with insulations are responsible for transfering raw materials to the process of the various units. Therefore, the corrosion under insulation (CUI) is the main cause of corrosion and pipe leakage in the dike. In order to prevent significant loss caused by leakage of contents, a concrete wall is used as an overflow prevention embankment around the spherical tank. When the pipeline go through the 1.3-meter-long concrete wall of the dike, the interface of the concrete wall is another area with high risk of corrosion. In this article, the guided wave testing and radiography testing profile are used to detect the corrosion under insulation. Guided wave testing is a technique used to inspect the local corrosion, general corrosion and corrosion under pipe clamp support. Radiography testing profile is used to evaluate the corrosion profile on the small vent or drain. Then, guided wave testing and QEUT are used to inspect and evaluate the corrosion at the interface of the dike. Combining the three NDT techniques, it is possible to provide an effective corrosion inspection program in the spherical tank area with removing the least dismantling of the insulation material.

　　隨著物聯網科技蓬勃發展，帶動企業研發大數據分析，利用人工智慧演算法將多元數據轉為知識，以提升決策效能。中油公司面對內外環境的變遷，兼顧經濟、環保及社會多面向需求，亟需透過「多元數據收集與資料視覺化」協助跨領域溝通，並藉由「數據整合與人工智慧」研發更精準預測模式，以發展智能決策。本文首先說明在巨量化的浪潮下，大數據分析的起因與可能衍生問題，其次介紹企業建置大數據平台的系統架構與資料視覺化功能，最後以石化廠膨脹機設備異常偵測為例，說明資料視覺化設計與數據分析建模的過程。本研究使用五種分類演算法進行分析，研究發現，預測模型平均準確率達99.9%，其中以隨機森林演算法最優，提供後續數據分析應用之參考。

　　Internet of Things (IOT) revolution drives the Big Data Analysis, enabling companies to use artificial intelligence in turning data into actionable knowledge for improving decision-making. Taking into account the needs of the economy, the environment and the society, CPC Corporation Taiwan proposes "data visualization" to help cross-disciplinary communication and develops more precise "artificial intelligence" models to provide intelligent forecasting. This paper presents an application of Big Data Analysis in the CPC petrochemical factory. Firstly, the evolution and features in big data application are described. Secondly, a Big Data architecture for Big Data Analysis is depicted. Next, this paper illustrates the process of visualization design for data analysis. Then, the preliminary problem of the expander anomaly in petrochemical plants is discussed. Finally, this study tackles different classification algorithms for expander anomaly detection. It is found that the average accuracy of five prediction models is 99.9%. Excellent predictive algorithm is Random Forest, which provides for future research.

　　本文利用台灣中油公司自營站 2010 年1 月1 日至2018 年9 月30 日每日的汽車與機車來客數，共計約190 萬筆資料，先以視覺化相關易懂圖形，如樹狀圖與文字雲等，幫助快速了解全台9 個營業處自營站汽機車來客數的分佈狀況及每週來客數的多寡與相關特性，並得到一些有趣且有意義的結論。另以時間數列介入函數探討9 個營業處在10 天的農曆假期期間，汽機車來客數的變化情形。依據9 個營業處汽機車來客數樣態除可作為公司銷售與輸儲規劃的依據，也可做為挑選促銷活動日期的參考。農曆年假期效果的探討，則可作為加油站在農曆年期間調度的參考。

　　This research used approximately 1.9 million data of daily customer numbers of cars and motorcycles of CPC Self-Operated gasoline stations from 2010.01.01 to 2018.09.30. First of all, we applied data visualization such as tree maps and word clouds to help quickly understand the distribution of customer numbers of cars and motorcycles between weekdays and 9 business branches. Then we implemented time series intervention analysis on the change of customer numbers of cars and motorcycles during 10-day Lunar New Year holidays. Understanding the distribution of customer numbers of cars and motorcycles of 9 business branches can help the strategy of marketing, transporting and storing. Research about Lunar New Year Effects can be a reference of the policy of storage during the holidays.

　　中油綠能所配合綠色能源研發政策，進行鈦酸鋰（Li4Ti5O12, LTO）負極材料生產計畫，建立鈦酸鋰材料製備實驗室與鋰電池電性量測實驗室，以及鋰鈦氧材料試量產工廠，從混漿／噴霧造粒產粉製程，燒結處理及過篩除水包裝，可日產100 公斤，年試量產量18 公噸，將成為本公司儲能材料生產重要里程碑。本計畫包括試量產材料製備程序改善、鋰鈦氧化物材料改質、放大生產試驗開發與全電池應用。試量產放大生產試驗開發部分，目前已完成年產能約18 噸，材料全產線建置，本年度除完成整線製程最適化之外，改善減少二氧化鈦等原物料殘留，將可助於後續電池性能提升及改善電池製程加工性，並送多家電池廠進行後續試驗，進行鋰鈦氧全電池製作最適化。鋰鈦氧儲能模組載具應用部分，完成風光式(風能/太陽能)LED 路燈實地應用，電動48V 堆高機實機應用，電容量為200Ah，與原鉛酸電池組相較，不但充電時間縮短，充電時亦無瀰漫酸氣情況，整體性能優於鉛酸電池組。

　　In order to coordinate the policy for development of green energy, Green Technology Research Institute (GTRI),CPC corporation, built the pilot plant to produce lithium titanium oxide (LTO, Li4Ti5O12) and related material research/characterization labs. The daily and year production capacities of the pilot plant can achieve 100 kg/day and 18 t/year, respectively, including the preparation of precursor slurry, spray-drying process, thermal annealing treatment, mechanical screening and water removal/package process. In this study, the Pilot plant production line in CPC was optimized and improved the reactivity to minimize the amount of unreacted TiO2 , leading to better performance. Moreover, the as-prepared LTO powders were also evaluated the electrochemical performance and the full lithium ion battery (LIB). The fabricated LIBs have been successfully integrated into the wind/solar powered LED Street light and applied for electric stackers. Compared with the traditional Lead-acid battery, the LTO based LIBs exhibit faster charge time and superior energy storage performance.