TWI782771B - Artificial intelligence system applied to the decommissioning of nuclear power plants and the analysis method thereof - Google Patents
Artificial intelligence system applied to the decommissioning of nuclear power plants and the analysis method thereof Download PDFInfo
- Publication number
- TWI782771B TWI782771B TW110140158A TW110140158A TWI782771B TW I782771 B TWI782771 B TW I782771B TW 110140158 A TW110140158 A TW 110140158A TW 110140158 A TW110140158 A TW 110140158A TW I782771 B TWI782771 B TW I782771B
- Authority
- TW
- Taiwan
- Prior art keywords
- radiation
- artificial intelligence
- nuclear power
- computing host
- packaging
- Prior art date
Links
Images
Abstract
Description
本發明係關於一種系統及其方法,特別是一種應用於核能發電廠除役之人工智慧系統及其分析方法。The present invention relates to a system and its method, especially an artificial intelligence system and its analysis method applied to the decommissioning of nuclear power plants.
核能發電廠,又稱核電廠或是核電站,是一種以核反應為熱力源之熱電廠,核能發電廠屬於高效率的能源建設,對於溫室氣體、二氧化碳排放幾乎是零,但核能發電廠建設成本高昂,技術需求高,養護成本亦高,在控制良好且周邊緊急應對系統完善的情況下,核能發電廠其實是相當安全的設施。A nuclear power plant, also known as a nuclear power plant or a nuclear power plant, is a thermal power plant that uses nuclear reactions as its heat source. Nuclear power plants are high-efficiency energy constructions that emit almost zero greenhouse gas and carbon dioxide emissions. However, the construction cost of nuclear power plants is high. The technical requirements are high, and the maintenance cost is also high. In the case of good control and the surrounding emergency response system, nuclear power plants are actually quite safe facilities.
核能發電是藉由原子核結構由於核分裂(Nuclear fission)或核融合(Nuclear fusion)發生變化時,所產生的能量致使發電,上述之核融合之概念是由兩個比較小的原子核成一個較大的原子核,於兩個小原子核在融合過程中會釋放能量,而理論上來說,核融合的產生的能量應大於核分裂之能量,但目前以現今研究來說,核融合的消耗電力,原本其產生的電力還多。Nuclear power generation is generated by the energy generated when the nuclear structure changes due to nuclear fission or nuclear fusion. The above-mentioned concept of nuclear fusion is to form a larger nucleus from two smaller nuclei. The atomic nucleus releases energy during the fusion process of two small nuclei. Theoretically speaking, the energy produced by nuclear fusion should be greater than the energy produced by nuclear fission. More electricity.
正因為如此,目前核能發電廠所使用的都是核分裂之技術,核分裂是將較大的原子核分裂為兩個較小的原子核,相對於核融合,核分裂在分裂較大原子的過程中會產生巨大的能量,透過此分裂過程所產生的熱能來轉換為電力。Because of this, nuclear power plants are currently using the technology of nuclear fission. Nuclear fission is the splitting of a larger nucleus into two smaller nuclei. Compared with nuclear fusion, nuclear fission will produce huge The energy is converted into electricity through the heat generated by this splitting process.
目前世界上的核能發電所使用的主要為鈾-235這種輻射物質來進行核分裂之反應,採來的鈾礦經過提煉及濃縮後製成可用於核能發電之燃料棒(鈾濃度約為3%),將燃料棒放入反應爐堆中,進行核分裂反映後產生熱,使熱所產生之蒸氣推動發電機進行發電。At present, nuclear power generation in the world mainly uses uranium-235, a radioactive substance, for nuclear fission reactions. The mined uranium ore is refined and enriched to make fuel rods that can be used for nuclear power generation (the concentration of uranium is about 3%. ), put the fuel rods into the reactor, and generate heat after the nuclear fission reaction, so that the steam generated by the heat drives the generator to generate electricity.
然而,核能發電的過程中,從鈾礦開採、濃縮鈾礦,到反應爐運轉,在處理乃至到最後反應爐除役,核燃料循環的整個過程中,皆會產生對人體及環境具有危害的核廢料,此種核廢料中存在著輻射放射性,又稱為放射性廢料,長時間處於這種輻射放射性環境內,會使人體細胞產生畸變,進而造成健康影響。However, in the process of nuclear power generation, from uranium mining, uranium enrichment, to the operation of the reactor, in the process of processing and finally to the decommissioning of the reactor, and the entire process of the nuclear fuel cycle, nuclear energy that is harmful to the human body and the environment will be produced. Waste, this kind of nuclear waste contains radioactivity, also known as radioactive waste. Long-term exposure to this radioactive environment will cause distortion of human cells and cause health effects.
核廢料又可分為低階核廢料及中階核廢料,低階核廢料包含了在核電廠內使用的紙張、布料、工具、服裝或過濾器等等,目前處理方式掩埋於掩埋場,而中階核廢料則包含了核電廠內之合成樹脂、化學汙泥、廢燃料棒之金屬護套以及除役後之反應爐部件。Nuclear waste can be divided into low-level nuclear waste and intermediate-level nuclear waste. Low-level nuclear waste includes paper, cloth, tools, clothing or filters used in nuclear power plants. The current treatment method is buried in landfills, while Intermediate nuclear waste includes synthetic resins, chemical sludge, metal sheaths of spent fuel rods, and decommissioned reactor components in nuclear power plants.
上述之核能電廠除役後之反應爐部件,涉及放射性物質及輻射劑量,所以核能電廠的除役工作相較其他產業的工廠或設施拆除,多了人員受到游離輻射所引起的輻射曝露危險,一般需耗費大量人力來進行,因此會導致現場除役工作人員吸收到輻射劑量。The above-mentioned reactor components after the decommissioning of the nuclear power plant involve radioactive substances and radiation doses, so the decommissioning of the nuclear power plant is compared with the dismantling of factories or facilities in other industries, and more people are exposed to radiation exposure risks caused by ionizing radiation. It is labor-intensive to carry out and therefore results in radiation doses absorbed by the decommissioning workers on site.
依據目前現有之技術,利用輻射偵檢設備進行刮除前之輻射偵檢,之後由混凝土除污刮除設備進行表面污染的混凝土表面刮除,並同時使刮除下之污染混凝土碎屑由吸取機構予以收集,且確認污染混凝土碎屑之外釋限值,如此,可一次完成刮除前輻射偵檢、表面污染的混凝土表面刮除及刮除後輻射偵檢等三個工作,而達到降低污染混凝土廢棄物數量、降低處理成本、減少大量除污後偵檢作業、減少人員暴露劑量以及符合輻射防護合理抑低原則之功效。According to the current existing technology, the radiation detection equipment is used to detect the radiation before scraping, and then the concrete surface contamination is scraped by the concrete decontamination scraping equipment, and at the same time, the scraped contaminated concrete debris is sucked In this way, the three tasks of radiation detection before scraping, surface contamination of concrete surface scraping and radiation detection after scraping can be completed at one time, so as to reduce The amount of contaminated concrete waste, reduction of treatment costs, reduction of detection operations after a large number of decontamination, reduction of personnel exposure dose and the effect of meeting the principle of reasonable suppression of radiation protection.
然而,依據目前現有之技術,雖可透過輻射偵檢設備進行輻射偵檢,但其並未對輻射量進行進一步的最佳化計算設置,且並即時透過檢測出來的輻射量做進一步的分割、儲存、容量以及容器之分析。However, according to the current existing technology, although the radiation detection equipment can be used for radiation detection, it does not further optimize the calculation and setting of the radiation amount, and further divides and divides the radiation amount through the detected radiation amount in real time. Analysis of storage, capacity and containers.
但是,且先前之技術,並無利用影像擷取裝置以及結構光掃描裝置等設備直接對場地進行建置,因此需使用者進入現場,容易使現場使用人員受到輻射汙染。However, the previous technology did not use image capture devices, structured light scanning devices and other equipment to directly build the site, so users need to enter the site, which may easily cause site users to be exposed to radiation pollution.
為此,如何在處理除役之核能發電廠內之輻射汙染源,並減少處理人員暴露於危險環境中,同時提升核能發電廠除役效率,縮短除役時程,為本領域技術人員所欲解決的問題。For this reason, how to deal with radiation pollution sources in decommissioned nuclear power plants, reduce the exposure of processing personnel to dangerous environments, and at the same time improve the decommissioning efficiency of nuclear power plants and shorten the decommissioning time schedule is what those skilled in the art want to solve The problem.
本發明之一目的,在於提供一種應用於核能發電廠除役之人工智慧系統及其分析方法,其係透過影像擷取裝置、3D結構光物體掃描以及輻射劑量量測裝置建構除役之核能發電廠廠區影像,並將核能發電廠廠區影像傳輸至運算主機,藉由運算主機內之人工智慧分析單元計算最佳化之切割含輻射汙染之區域之能力,以及封裝條件。An object of the present invention is to provide an artificial intelligence system and its analysis method applied to the decommissioning of nuclear power plants, which is to construct decommissioned nuclear power generation through image capture devices, 3D structured light object scanning and radiation dose measurement devices The image of the factory area, and the image of the nuclear power plant area is transmitted to the computing host, and the artificial intelligence analysis unit in the computing host calculates the optimal ability to cut the area containing radiation pollution, as well as the packaging conditions.
針對上述之目的,本發明提供一種應用於核能發電廠除役之人工智慧分析方法,其步驟包含透過一掃描模組掃描一核能發電廠產生一區域影像後,將該區域影像傳輸至一運算主機;使用一輻射量測裝置分別量測該核能發電廠之複數個區域產生複數個輻射參數,並將該些個輻射參數傳輸至該運算主機;該運算主機將該區域影像及該些個輻射參數疊合後,產生一輻射值域;該運算主機利用一人工智慧分析單元依據該輻射值域進行分析,獲得一輻射強度分佈;該人工智慧分析單元依據該輻射強度分佈之一強度值計算取得至少一污染區域;透過該人工智慧分析單元判斷該至少一汙染區域之一物件參數,並依據該物件參數計算一封裝參數。For the above purpose, the present invention provides an artificial intelligence analysis method applied to the decommissioning of nuclear power plants. The steps include scanning a nuclear power plant through a scanning module to generate an image of an area, and then transmitting the image of the area to a computing host ; Use a radiation measurement device to measure a plurality of radiation parameters produced in multiple areas of the nuclear power plant, and transmit these radiation parameters to the computing host; the computing host will image the area and the radiation parameters After overlapping, a radiation value range is generated; the computing host uses an artificial intelligence analysis unit to analyze the radiation value range to obtain a radiation intensity distribution; the artificial intelligence analysis unit calculates and obtains at least A polluted area; an object parameter of the at least one polluted area is judged by the artificial intelligence analysis unit, and a packaging parameter is calculated according to the object parameter.
本發明提供一實施例,其中透過一掃描模組掃描一核能發電廠產生一區域影像後,將該區域影像傳輸至一運算主機之步驟中,該掃描模組包含一3D掃描裝置及一影像擷取裝置。The present invention provides an embodiment, wherein after a region image is generated by scanning a nuclear power plant through a scanning module, the step of transmitting the region image to a computing host, the scanning module includes a 3D scanning device and an image capture Take the device.
本發明提供一實施例,其中該3D掃描裝置係為3D雷射掃描裝置或3D結構光掃描裝置。The present invention provides an embodiment, wherein the 3D scanning device is a 3D laser scanning device or a 3D structured light scanning device.
本發明提供一實施例,其中透過一掃描模組掃描一核能發電廠產生一區域影像後,將該區域影像傳輸至一運算主機之步驟中,該運算主機係為一個人電腦或一伺服器。The present invention provides an embodiment, wherein after a nuclear power plant is scanned by a scanning module to generate an area image, the area image is transmitted to a computing host, and the computing host is a personal computer or a server.
本發明提供一實施例,其中使用一輻射量測裝置分別量測該核能發電廠之複數個區域產生複數個輻射參數之步驟中,該些個輻射參數分別包含一輻射量及一輻射座標值The present invention provides an embodiment, wherein in the step of using a radiation measuring device to respectively measure a plurality of radiation parameters generated in a plurality of areas of the nuclear power plant, the radiation parameters respectively include a radiation amount and a radiation coordinate value
本發明提供一實施例,其中透過該人工智慧分析單元判斷該至少一汙染區域之一物件參數,並依據該物件參數計算一封裝參數之步驟中,該物件參數係包含一物件比活度以及一物件容積。The present invention provides an embodiment, wherein an object parameter of the at least one polluted area is judged by the artificial intelligence analysis unit, and in the step of calculating a packaging parameter according to the object parameter, the object parameter includes an object specific activity and a Object volume.
本發明提供一實施例,其中透過該人工智慧分析單元判斷該至少一汙染區域之一物件參數,並依據該物件參數計算一封裝參數之步驟後,包含步驟:該人工智慧分析單元依據該封裝參數之一封裝容器容積或一裝填物比活度計算一封裝結果,其中該封裝結果係為一封裝容器量值。The present invention provides an embodiment, wherein after the step of determining an object parameter of the at least one polluted area through the artificial intelligence analysis unit, and calculating a packaging parameter according to the object parameter, the steps include: the artificial intelligence analysis unit according to the packaging parameter A packing result is calculated according to the volume of a packing container or the specific activity of a filling, wherein the packing result is a value of a packing container.
本發明提供一實施例,其中該人工智慧分析單元將該封裝參數傳輸至一外部裝置之步驟中,該外部裝置係為一機械手臂或一擬真機器人。The present invention provides an embodiment, wherein the artificial intelligence analysis unit transmits the packaging parameters to an external device, and the external device is a mechanical arm or a virtual robot.
本發明提供一實施例,其中於當該封裝參數之一封裝容器容積或一裝填物比活度之其中之一到達一門檻值,則對該封裝容器進行一封裝之步驟後,包含步驟:當該外部裝置切割之該汙染物質之該封裝容器容積、該封裝容器表面輻射量或該裝填物比活度之大於一門檻值;該外部裝置產生一訊息,並將該訊息回傳至該人工智慧分析單元;該人工智慧分析單元進行分析並計算出一最佳化分割封裝參數;將該最佳化分割封裝參數傳輸至該外部裝置;以及該外部裝置依據該最佳化分割封裝參數分割該汙染區域,並將該汙染區域填充至該封裝容器。The present invention provides an embodiment, wherein when one of the packaging parameters, one of the volume of the packaging container or the specific activity of a filling, reaches a threshold value, after the step of packaging the packaging container, the steps include: when The volume of the packaging container of the pollutant cut by the external device, the radiation amount of the surface of the packaging container, or the specific activity of the filling is greater than a threshold value; the external device generates a message, and sends the message back to the artificial intelligence an analysis unit; the artificial intelligence analysis unit analyzes and calculates an optimal segmentation and packaging parameter; transmits the optimal segmentation and packaging parameter to the external device; and the external device divides the pollution according to the optimal segmentation and packaging parameter area, and fill the contaminated area into the packaging container.
針對上述之目的,本發明提供其係設置於一核能發電廠內,其包含一運算主機,其係包含一人工智慧分析單元;一掃描模組,其係包含一3D掃描裝置及一影像擷取裝置,該3D掃描模組及該影像擷取裝置係電性連接該運算主機,該掃描模組係用以掃描該核能發電廠並產生一區域影像; 一輻射量測裝置,其係電性連接於該運算主機,該輻射量測裝置係用以量測該核能發電廠之複數個區域產生複數個輻射參數;以及一外部裝置,其係電性連接該運算主機;其中,該運算主機將該區域影像及該些個輻射參數疊合後,產生一輻射值域,該運算主機利用一人工智慧分析單元依據該輻射值域進行分析,其係用以獲得一輻射強度分佈,依據該輻射強度分佈取得至少一污染區域,透過該人工智慧分析單元判斷該至少一汙染區域之一物件參數,並依據該物件參數計算一封裝參數。For the above-mentioned purpose, the present invention provides that it is installed in a nuclear power plant, and it includes a computing host, which includes an artificial intelligence analysis unit; a scanning module, which includes a 3D scanning device and an image capture device, the 3D scanning module and the image capture device are electrically connected to the computing host, and the scanning module is used to scan the nuclear power plant and generate an area image; a radiation measurement device, which is electrically connected On the computing host, the radiation measuring device is used to measure a plurality of radiation parameters produced in a plurality of areas of the nuclear power plant; and an external device, which is electrically connected to the computing host; wherein, the computing host will After the regional image and the radiation parameters are superimposed, a radiation value range is generated. The computing host uses an artificial intelligence analysis unit to analyze according to the radiation value range, which is used to obtain a radiation intensity distribution. According to the radiation intensity distribution At least one polluted area is obtained, an object parameter of the at least one polluted area is judged by the artificial intelligence analysis unit, and a packaging parameter is calculated according to the object parameter.
本發明提供一實施例,其中該運算主機係為一個人電腦或一伺服器。The present invention provides an embodiment, wherein the computing host is a personal computer or a server.
本發明提供一實施例,其中該些個輻射參數分別包含一輻射量及一輻射座標值。The present invention provides an embodiment, wherein the radiation parameters respectively include a radiation amount and a radiation coordinate value.
為使 貴審查委員對本發明之特徵及所達成之功效有更進一步之瞭解與認識,謹佐以較佳之實施例及配合詳細之說明,說明如後:In order to enable your review committee members to have a further understanding and understanding of the characteristics of the present invention and the achieved effects, I would like to provide a better embodiment and a detailed description, as follows:
習知在處理除役之核能發電廠內之輻射汙染源之方法中,依據目前現有之技術,並無利用影像擷取裝置以及結構光掃描裝置等設備直接對場地進行建置,因此需使用者進入現場,容易使現場使用人員受到輻射汙染,再者,雖可以透過輻射偵檢設備進行輻射偵檢,但其並未對輻射量進行進一步的最佳化計算設置,且並即時透過檢測出來的輻射量做進一步的分割、儲存、容量以及容器之分析。It is known that in the method of dealing with radiation pollution sources in decommissioned nuclear power plants, according to the current existing technology, there is no equipment such as image capture devices and structured light scanning devices to directly construct the site, so users need to enter In the field, it is easy to cause the site users to be exposed to radiation pollution. Moreover, although the radiation detection equipment can be used for radiation detection, it does not further optimize the calculation and setting of the radiation amount, and the detected radiation can be transmitted immediately. Quantity for further segmentation, storage, capacity and container analysis.
本發明透過影像擷取裝置以及3D掃描裝置,將拆除物件外觀尺寸建模,再藉由輻射量測裝置對該物件進行量測,依據影像擷取裝置將所量測得到的數值與3D掃描裝置之資料相結合,接著透過人工智慧分析單元進行多重條件最佳化分類與切割分析,並提供分類與切割建議,縮短核能發電廠除役工作時程,且降低工作人員曝露於輻射環境之時間。The present invention uses the image capture device and the 3D scanning device to model the appearance and size of the demolished object, and then uses the radiation measurement device to measure the object, and compares the measured value with the 3D scanning device according to the image capture device. Combined with the data, the artificial intelligence analysis unit is used to optimize the classification and cutting analysis of multiple conditions, and provide classification and cutting suggestions to shorten the decommissioning work schedule of nuclear power plants and reduce the exposure time of workers to the radiation environment.
在下文中,將藉由圖式來說明本發明之各種實施例來詳細描述本發明。然而本發明之概念可能以許多不同型式來體現,且不應解釋為限於本文中所闡述之例示性實施例。Hereinafter, the present invention will be described in detail by illustrating various embodiments of the present invention by means of the accompanying drawings. Inventive concepts may, however, be embodied in many different forms and should not be construed as limited to the illustrative embodiments set forth herein.
首先,請參閱第1A圖,其為本發明之一實施例之步驟流程示意圖,以及第2圖,其為本發明之一實施例之系統示意圖,如圖所示,本實施例之步驟:First of all, please refer to Fig. 1A, which is a schematic flow chart of the steps of one embodiment of the present invention, and Fig. 2, which is a schematic diagram of the system of one embodiment of the present invention, as shown in the figure, the steps of this embodiment:
步驟S10:透過掃描模組掃描核能發電廠產生區域影像後,將區域影像傳輸至運算主機;Step S10: After the area image is generated by scanning the nuclear power plant through the scanning module, the area image is transmitted to the computing host;
步驟S20:使用輻射量測裝置分別量測核能發電廠之區域產生輻射參數,並將輻射參數傳輸至運算主機;Step S20: Use the radiation measuring device to measure the radiation parameters generated in the area of the nuclear power plant, and transmit the radiation parameters to the computing host;
步驟S30:運算主機將區域影像及輻射參數疊合後,產生輻射值域;Step S30: The computing host superimposes the area image and the radiation parameters to generate a radiation value range;
步驟S40:運算主機利用人工智慧分析單元依據輻射值域進行分析,其係用以獲得輻射強度分佈;Step S40: The computing host uses the artificial intelligence analysis unit to analyze according to the radiation value range, which is used to obtain the radiation intensity distribution;
步驟S50:人工智慧分析單元依據輻射強度分佈之強度值計算取得污染區域;以及Step S50: the artificial intelligence analysis unit calculates and obtains the contaminated area according to the intensity value of the radiation intensity distribution; and
步驟S60:透過人工智慧分析單元判斷汙染區域之物件參數,並依據物件參數計算封裝參數。Step S60: Judging the object parameters of the polluted area through the artificial intelligence analysis unit, and calculating the packaging parameters according to the object parameters.
如步驟S10至步驟S60所述之步驟,本實施例透過一掃描模組20掃描一核能發電廠產生一區域影像後,將該區域影像傳輸至一運算主機10,接著使用一輻射量測裝置30分別量測該核能發電廠之複數個區域產生複數個輻射參數,並將該些個輻射參數傳輸至一運算主機10,該運算主機10將該區域影像及該些個輻射參數疊合後,產生一輻射值域,該運算主機10利用一人工智慧分析單元12依據該輻射值域進行分析,獲得一輻射強度分佈,該人工智慧分析單元12依據該輻射強度分佈之一強度值計算取得至少一污染區域,透過該人工智慧分析單元12判斷該至少一汙染區域之一物件參數,並依據該物件參數計算一封裝參數。As described in steps S10 to S60, in this embodiment, a nuclear power plant is scanned by a
其中,於本實施例中,該掃描模組20包含一3D掃描裝置22及一影像擷取裝置24,進一步,該3D掃描裝置22係使用3D結構光掃描裝置或3D雷射掃描裝置。Wherein, in this embodiment, the
目前3D掃描裝置之技術成熟,該3D掃描裝置22係用來偵測並分析物體或環境之形狀,透過該3D掃描裝置22蒐集到的資料,經常被用來作為三維重建之計算,在於虛擬世界中建立數位模型,而目前舉凡工業設計、瑕疵檢測、逆向工程、機器人導引、地貌測量、醫學資訊、生物資訊、刑事鑑定、數位文物典藏、電影製片、遊戲創作素材等等都可見其應用。At present, the technology of 3D scanning device is mature. The
因此,本實施例透過該3D掃描裝置22來掃描該核能發電廠之環境結構,再結合該影像擷取裝置24取得該核能發電廠之環境影像,將掃描所得之環境結構與環境影項疊合後,形成該區域影像,其中,該區域影像包含一區域影像座標值。Therefore, this embodiment uses the
接著,於本實施例中,該運算主機10係為一個人電腦或一伺服器,該些個輻射參數分別包含一輻射量及一輻射座標值,上述步驟中,透過該運算主機10將該區域影像及該些個輻射參數疊合後,也就是利用該輻射座標值對應該區域影像座標值疊合產生該輻射值域,再透過該人工智慧分析單元12依據該輻射值域進行分析,獲得該輻射強度分佈。Next, in this embodiment, the
其中,該人工智慧分析單元12係藉由現有演算法進行分析計算,其中,用以處理影像計算部分之演算法係使用YOLO (You only look once,簡稱YOLO)演算法,結合捲積神經網路 (Convolutional Neural Networks,簡稱CNN),YOLO演算法(V1至V4)係為一種物件偵測(object detection)之類神經網路演算法,可針對影像中每個物件進行物件追蹤、偵測以及判斷,CNN則是一種前饋神經網路,它的人工神經元可以回應一部分覆蓋範圍內的周圍單元,對於大型圖像處理有出色表現。Wherein, the artificial
於本實施例中,可以透過YOLO演算法以及捲積神經網路進行該核能發電廠之該輻射強度分布計算,然而,上述之YOLO演算法及捲積神經網路係為本發明之一較佳實施例,進一步亦可使用現有之演算法,如:蒙地卡羅、GNN、SSD、VGG16、RESNET50、Logistic Regression或LSDM,不以上述之YOLO或CNN為限。In this embodiment, the calculation of the radiation intensity distribution of the nuclear power plant can be performed through the YOLO algorithm and the convolutional neural network. However, the above-mentioned YOLO algorithm and the convolutional neural network are one of the preferred embodiments of the present invention. In an embodiment, an existing algorithm such as Monte Carlo, GNN, SSD, VGG16, RESNET50, Logistic Regression or LSDM may be used, not limited to the aforementioned YOLO or CNN.
其中,該人工智慧分析單元12根據該輻射強度分布之該強度值,取得該至少一汙染區域,於此定義之該輻射強度分布之該強度值大於1x10
2(Bq/Kg) ,其中,前述之比活度也稱為比放射性,指放射源的放射性活度與其品質之比,即單位品質產品中所含某種核素的放射性活度。
Wherein, the artificial
其中,上述之該物件參數係包含一物件容積以及一物件比活度,也就是說,該人工智慧分析單元12取得該至少一汙染區域時,該人工智慧分析單元12依據該物件參數之該物件容積以及該物件比活度,計算出該至少一汙染區域之一元件欲封裝時之該封裝參數。Wherein, the above-mentioned object parameters include an object volume and an object specific activity, that is to say, when the artificial
進一步包含步驟:Further include steps:
步驟S65:人工智慧分析單元依據封裝參數之封裝容器容積或裝填物比活度計算封裝結果。Step S65: The artificial intelligence analysis unit calculates the encapsulation result according to the encapsulation parameters such as the volume of the encapsulation container or the specific activity of the filling.
並於步驟S65中,本實施例透過該人工智慧分析單元12依據該封裝參數之一封裝容器容積或一裝填物比活度計算一封裝結果,其中,該封裝結果係為一封裝容器量值,該封裝容器量值係為該封裝容器之數量,藉由該至少一汙染區域之該元件之該物件容積以及該物件比活度來計算需要幾個該封裝容器60。And in step S65, the present embodiment uses the artificial
接著,於本實施例中,請參閱第1B圖,其為本發明之一實施例之封裝程序之流程示意圖,以及一併參閱第2圖,於步驟S60後,包含步驟S70至步驟S90,步驟如下:Next, in this embodiment, please refer to FIG. 1B, which is a schematic flow chart of the encapsulation procedure of an embodiment of the present invention, and refer to FIG. 2 together, after step S60, including step S70 to step S90, the step as follows:
步驟S70:人工智慧分析單元將封裝參數傳輸至外部裝置;Step S70: the artificial intelligence analysis unit transmits the packaging parameters to the external device;
步驟S80:外部裝置依據封裝參數分割汙染區域,並將汙染區域填充至封裝容器;以及Step S80: The external device divides the contaminated area according to the packaging parameters, and fills the contaminated area into the packaging container; and
步驟S90:當物件比活度或物件容積符合封裝參數之封裝容器容積或裝填物比活度之其中之一之門檻值時,外部裝置對封裝容器進行封裝程序。Step S90: When the specific activity of the object or the volume of the object meets the threshold value of one of the volume of the packaging container or the specific activity of the filling of the packaging parameter, the external device performs a packaging process on the packaging container.
至步驟S70至步驟S90所示之步驟,本實施例之該人工智慧分析單元12將該封裝參數傳輸至一外部裝置50,該外部裝置50依據該封裝參數分割該至少一汙染區域,並將該至少一汙染區域之該元件填充至一封裝容器60。Going to the steps shown in step S70 to step S90, the artificial
接著,將該封裝參數傳輸至該核能發電廠之該外部裝置50,使該外部裝置50以此作為依據分割該至少一汙染區域之該元件後,將分割後之部分該至少一汙染區域之該元件填充放入該封裝容器60,其中,該外部裝置50係為一機械手臂或一擬真機器人,該封裝容器60係為鋼、混凝土或鉛之其中之一混製而成的桶裝容器。Then, transmit the packaging parameters to the
其中,於步驟S90所示之步驟,當上述之該物件比活度或該物件容積符合該封裝參數之一封裝容器容積或一裝填物比活度之其中之一之一門檻值時,該外部裝置50對該封裝容器60進行一封裝程序。Wherein, in the step shown in step S90, when the above-mentioned specific activity of the object or the volume of the object meets one of the threshold values of the packaging parameter, the volume of the packaging container or the specific activity of a filling, the external The
更進一步,於步驟S90後包含步驟S95,步驟如下:Furthermore, step S95 is included after step S90, and the steps are as follows:
步驟S95:外部裝置針對完成封裝程序之封裝容器量測封裝容器表面輻射量,其中,封裝容器表面輻射值小於預設值。Step S95: The external device measures the surface radiation amount of the packaging container for the packaging container that has completed the packaging process, wherein the surface radiation value of the packaging container is smaller than a preset value.
因此,當該封裝容器容積或該裝填物比活度之其中之一條件超過最大值時,該外部裝置50就會將該封裝容器60進行該封裝程序,同時該封裝程序後,該外部裝置50亦會量測該封裝容器表面輻射量,測量是否在封裝完成後,該封裝容器60之該封裝容器表面輻射量小於一預設值,當該封裝容器表面輻射值小於該預設值時,才能進行後續處置。Therefore, when one of the conditions of the volume of the packaging container or the specific activity of the filling exceeds the maximum value, the
而上述之該封裝容器容積、該封裝容器表面輻射量或該裝填物比活度請參考表一,表一為該封裝容器60之該封裝參數。
其中,由表一可得知,該封裝容器容積之該門檻值係介於2.1至6.5 (m 3)之間,而該裝填物比活度之該門檻值係介於1x10 4至2x10 11(Bq/kg)之間,該封裝容器表面輻射量之該預設值介於100至300 (uSv/h)之間。 Wherein, it can be seen from Table 1 that the threshold value of the volume of the packaging container is between 2.1 and 6.5 (m 3 ), and the threshold value of the specific activity of the filling is between 1x10 4 and 2x10 11 ( Bq/kg), the preset value of the surface radiation of the packaging container is between 100 and 300 (uSv/h).
另,請參考第1C圖,其為本發明之一實施例之步驟流程示意圖,如圖所示,於步驟S90後,包含步驟:In addition, please refer to FIG. 1C, which is a schematic flow chart of the steps of an embodiment of the present invention. As shown in the figure, after step S90, the steps include:
步驟S100:當外部裝置切割之汙染物質之封裝容器容積、封裝容器表面輻射量或裝填物比活度之大於門檻值;Step S100: When the volume of the packaging container of the pollutants cut by the external device, the radiation amount on the surface of the packaging container, or the specific activity of the filling is greater than the threshold value;
步驟S110:外部裝置產生訊息,並將訊息回傳至人工智慧分析單元;Step S110: the external device generates a message, and sends the message back to the artificial intelligence analysis unit;
步驟S120:人工智慧分析單元進行分析並計算出最佳化分割封裝參數;Step S120: the artificial intelligence analysis unit analyzes and calculates the optimal segmentation and packaging parameters;
步驟S130:將最佳化分割封裝參數傳輸至外部裝置;以及Step S130: Transmitting the optimal segmentation and packaging parameters to an external device; and
步驟S140:外部裝置依據最佳化分割封裝參數分割汙染區域,並將汙染區域填充至封裝容器。Step S140: The external device divides the contaminated area according to the optimized segmentation and packaging parameters, and fills the contaminated area into the packaging container.
也就是說,當該外部裝置50切割下來之該至少一汙染區域之該元件過大,無法容納於表一內之該封裝容器60時,此時,該外部裝置50會產生一訊息,並將該訊息回傳至該運算主機10之該人工智慧分析單元12,該人工智慧分析單元12則會再次進行分析運算,計算出一最佳化分割封裝參數,再將該最佳化分割封裝參數傳輸給予該外部裝置50,使該外部裝置50再度進行分割封存之運行。That is to say, when the component of the at least one contaminated area cut by the
另外,本實施例中更包含一環境感測裝置40,該環境感測裝置40係為有毒氣體感測器、PM2.5粉塵感測器、溫度感測器、濕度感測器、磁場感測器、電磁波感測器、紫外線感測器或生命跡象感測器之其中之一或其上述任意選擇之一,該環境感測裝置40透過將測得之一環境資訊傳輸至該運算主機10,使外部監控人員可得知切割時之該核能發電廠內之環境參數,以此判斷何時可進入該核能發電廠內。In addition, this embodiment further includes an
本發明係關於一種應用於核能發電廠除役之人工智慧系統及其分析方法,其包含該人工智慧分析單元12之該運算主機10、該影像擷取裝置24、該3D掃描裝置22、該輻射量測裝置30、透過應用於核能發電廠除役之人工智慧系統使該人工智慧分析單元12依據該影像擷取裝置24、及該3D掃描裝置22與該輻射量測裝置30,進行該核能發電廠內部之除役物件(該汙染區域之該元件)之辨識、標記與量測。The present invention relates to an artificial intelligence system and its analysis method applied to the decommissioning of nuclear power plants, which includes the
透過該影像擷取裝置24、及該3D掃描裝置22將欲拆除之該核能發電廠之除役物件(該汙染區域之該元件)之外觀尺寸進行建模,再藉由該輻射量測裝置30對該核能發電廠進行量測,依據該影像擷取裝置24以及該3D掃描裝置22獲得之結果相結合後,透過該人工智慧分析單元12進行分析,進行多重條件最佳化分類與切割分析,並提供分類與切割建議,進而可縮短核能發電廠除役工作時程,且降低工作人員曝露於輻射環境之時間。Through the
以上所述之實施例,本發明之方法係為一種應用於核能發電廠除役之人工智慧系統及其方法,其係透過影像擷取裝置、3D結構光物體掃描以及輻射劑量量測裝置建構除役之核能發電廠廠區影像,並將核能發電廠廠區影像傳輸至運算主機,藉由運算主機內之人工智慧分析單元計算最佳化之切割含輻射汙染之區域之能力以及封裝條件。In the above-mentioned embodiment, the method of the present invention is an artificial intelligence system and its method applied to the decommissioning of nuclear power plants. The image of the nuclear power plant site in service, and the image of the nuclear power plant site is transmitted to the computing host, and the artificial intelligence analysis unit in the computing host calculates the optimal ability to cut the area containing radiation pollution and packaging conditions.
故本發明實為一具有新穎性、進步性及可供產業上利用者,應符合我國專利法專利申請要件無疑,爰依法提出發明專利申請,祈 鈞局早日賜准專利,至感為禱。Therefore, the present invention is novel, progressive and can be used in industry. It should meet the patent application requirements of my country's patent law. I file an invention patent application in accordance with the law. I pray that the bureau will grant the patent as soon as possible. I sincerely pray.
惟以上所述者,僅為本發明之較佳實施例而已,並非用來限定本發明實施之範圍,舉凡依本發明申請專利範圍所述之形狀、構造、特徵及精神所為之均等變化與修飾,均應包括於本發明之申請專利範圍內。However, the above-mentioned ones are only preferred embodiments of the present invention, and are not used to limit the scope of the present invention. For example, all equal changes and modifications are made according to the shape, structure, characteristics and spirit described in the scope of the patent application of the present invention. , should be included in the patent application scope of the present invention.
10:運算主機10: Computing host
12:人工智慧分析單元12: Artificial intelligence analysis unit
20:掃描模組20: Scanning module
22:3D掃描裝置22:3D scanning device
24:影像擷取裝置24: Image capture device
30:輻射量測裝置30:Radiation measuring device
40:環境感測裝置40: Environmental sensing device
50:外部裝置50: External device
60:封裝容器60: Packaging container
S10、S20、S30、S40、S50、S60、S65、S70、S80、S90、S95、S100、S110、S120、S130、S140:步驟S10, S20, S30, S40, S50, S60, S65, S70, S80, S90, S95, S100, S110, S120, S130, S140: steps
第1A圖:其為本發明之一實施例之步驟流程示意圖; 第1B圖:其為本發明之一實施例之封裝程序之流程示意圖; 第1C圖:其為本發明之一實施例之步驟流程示意圖;以及 第2圖:其為本發明之一實施例之系統示意圖。 Fig. 1A: It is a schematic flow chart of the steps of an embodiment of the present invention; Fig. 1B: It is a schematic flow chart of the encapsulation procedure of one embodiment of the present invention; Figure 1C: It is a schematic flow chart of the steps of an embodiment of the present invention; and Figure 2: It is a system diagram of an embodiment of the present invention.
S10、S20、S30、S40、S50、S60、S65:步驟 S10, S20, S30, S40, S50, S60, S65: steps
Claims (7)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW110140158A TWI782771B (en) | 2021-10-28 | 2021-10-28 | Artificial intelligence system applied to the decommissioning of nuclear power plants and the analysis method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW110140158A TWI782771B (en) | 2021-10-28 | 2021-10-28 | Artificial intelligence system applied to the decommissioning of nuclear power plants and the analysis method thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI782771B true TWI782771B (en) | 2022-11-01 |
TW202318326A TW202318326A (en) | 2023-05-01 |
Family
ID=85794321
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW110140158A TWI782771B (en) | 2021-10-28 | 2021-10-28 | Artificial intelligence system applied to the decommissioning of nuclear power plants and the analysis method thereof |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI782771B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW436813B (en) * | 1997-07-24 | 2001-05-28 | Transports De L Ind Nucleai So | Device for permanent inspection of the tightness of container closing covers for radioactive materials |
TW201413667A (en) * | 2012-09-17 | 2014-04-01 | Oriental Inst Technology | Ambient detection system and portable ambient measurement device thereof |
US20180246236A1 (en) * | 2016-06-24 | 2018-08-30 | Hitachi, Ltd. | Sensor apparatus, planning processing system, and planning method |
-
2021
- 2021-10-28 TW TW110140158A patent/TWI782771B/en active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW436813B (en) * | 1997-07-24 | 2001-05-28 | Transports De L Ind Nucleai So | Device for permanent inspection of the tightness of container closing covers for radioactive materials |
TW201413667A (en) * | 2012-09-17 | 2014-04-01 | Oriental Inst Technology | Ambient detection system and portable ambient measurement device thereof |
US20180246236A1 (en) * | 2016-06-24 | 2018-08-30 | Hitachi, Ltd. | Sensor apparatus, planning processing system, and planning method |
Also Published As
Publication number | Publication date |
---|---|
TW202318326A (en) | 2023-05-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR102013749B1 (en) | Method and apparatus for decommissioning planning of nuclear facility | |
López et al. | Neutronic modeling of debris beds for a criticality evaluation | |
TWI782771B (en) | Artificial intelligence system applied to the decommissioning of nuclear power plants and the analysis method thereof | |
Jharko et al. | Aspects of nuclear power plant digital decommissioning | |
Bylkin et al. | Composition and structure of simulation models for evaluating decommissioning costs for nuclear power plant units. | |
Heo et al. | Framework for the development of guidelines for nuclear power plant decommissioning workers based on risk information | |
Testoni et al. | Tracer use for the protection of water resources in nuclear sites | |
Rahman | PRELIMINARY ASSESSMENT OF CONTINUOUS ATMOS-PHERIC DISCHARGE FROM THE LOW RADIOACTIVE WASTE INCINERATOR | |
Yaar et al. | Evaluation of Hydrogel Technologies for the Decontamination of 137Cs From Building Material Surfaces | |
Chabal et al. | iDROP: An Innovative Software Program to Design Nuclear Decommissioning Scenarios | |
Galson et al. | Performance assessment and the safety case: lessons from the European Commission PAMINA project | |
Kim et al. | Development of Integrated Decontamination Equipment to Remove Concrete Surface Contamination | |
Depauw et al. | 3D Digital Simulation Tools Used to Support Nuclear Facilities Decommissioning-Benefits for the Projects Progress and their Safety | |
Tanaka | DEVELOPING DATA INTERFACE AND DATE-RECORD FORMAT FOR DIGITAL DECOMMISSIONING | |
Haolei et al. | Study on Near Surface Radionuclide Activity Under Vehicle Heat Pipe Small Modular Reactor Accident Condition | |
Kristofova et al. | ADVANCED DECOMMISSIONING COSTING CALCULATION TOOLS FOR PILOT VIENNA TRIGA MARK-II RESEARCH REACTOR | |
Song et al. | A Study on the Dose Assessment of Workers in the Transition Period | |
Song et al. | An analysis on waste stream classification of radioactive waste generated in domestic institute and international radioactive waste tracking system | |
Ding et al. | A Novel Approach for Predicting Mid-Long Term Radiation Dose in Case of Hypothetical Stsbo Nuclear Accident for an Operating Nuclear Power Plant | |
Bailey et al. | System Integration for Remote Systems at DOE Hanford Nuclear Site-18712 | |
Parvin | Research reactor decommissioning and radioactive waste management | |
Strons et al. | Development of CFD Flow Analysis for Nuclear Facility Ventilation Systems | |
Shuman et al. | Documentation of a simple environmental pathways model of the Radioactive Waste Management Complex at the Idaho National Engineering Laboratory.[DOSTOMAN code] | |
Little Jr et al. | Confinement of airborne activity from melted antimony slugs | |
Kim et al. | Effect of Detector Pose Uncertainty in Localization of Radiation Sources |