丄344718 九、發明說明： 【發明所屬之技術領域】 本發明係有關一種燃料電池燃料供應控制方法，尤其是 才曰-種於燃料電池反應過程中’藉由量測燃料電池於反應過 程中之特徵值變化以觸負載之變化，進而控㈣料供料 機增進燃料電池運作效較—_料電池之簡供應方法。。 【先前技術】 燃料電池是利用電化學反應將化學能觀為電能輸出的 種發電裝置。其工作原理是利用含有氫之燃料與氧化劑（如 空氣或者是氧氣）分別輸送到電池的陽極與陰極，陽極發生氧 ，反應將燃料解離成氫離子與電子，氫離子從陽極透過質子 父換膜而至陰極，結合經外負載電路傳導至陰極之電子，而 與氧氣發线狀社成水。只錢續輯地供舰料，辦 料電池就可以稍地發電。藉由其高效率與低污染的兩大特 點，使得此技術開發以來一直廣受矚目。 在燃料電池中，直接甲醇燃料電池（Direc1： Methan〇1 heJCell’DMFC，以下簡稱DMFC)，可以更方便地應用於各種 型電器用品（筆記型電腦、pDA、Gps)的電源供應上。 疋各國近年來積極投入的燃料電池系統。DMFC與其他技術如 PEMFC之不同點在於，其燃料改由液態甲醇取代氫氣做為燃 料，大幅提昇燃料電池燃料儲存、運送之方便性，安全性。 曰然而在DMFC的燃料供應中，熟悉此項技術之人都了解， 過置的燃料供應（如曱醇）會導致嚴重的燃料（甲醇）穿越 1344718 (methane^ cr〇ss〇ver)在陰極ψ醇直接與氧氣作用產生混成 電位降Omxed PGtentlal)的縣，進而導致_電池之 效率不佳的情形，嚴重時會料負電壓現象進而損壞電池。 另外’為了配合燃料電池所供應之負載需要，所以如何適量 ^控制燃料之供給量變成相#重要。例如在卵技術中有所 Μ用燃料滚度感測器，如甲醇濃度感測器，來感測燃料（甲 酵）濃度狀態，以判斷供應燃料（甲醇)之時機。此外，在 ，術中有揭露以t化學量測的方式來作為曱醇感測器。然而 4方法具錢項賴，-、增力统之複雜度以 ，成本。—、湘電化學方法所作的燃料濃度制器由於電 =有老化現象，因此該燃料濃錢測㈣要經常校正，以 f夺量測之準顧1相物理方式做成龍料濃度减㈣ =溫度的函數’ #測數據通常需做溫度的修正或補償，因此 影響控制精度更增加了控制之複雜度。 料漢度感測器的方式，會額外增加成本以及李 方複因此如何錢制❹⑶來作為㈣ 方法，也逐漸形成趨勢。在美國專利us = 所揭露的技術中，係利用偵測燃料電池之電流、甲 J之溫度：來判斷甲醇燃料之濃度，是-種間接量測的t出 種方法叉限於膜電極組的老化影響。此外，又、圭 ㈣錢6,99⑽所揭露的一種 =專利 路電流綱路電壓，作為狀燃料供應之方法池之短 要週期性的將_電池進行開路或短路，對於此技術需 易造成損傷，㈣響電池之穩定性與壽命。·.，"1池本身容 综合上述，因此虽需-種避免使用感測器而且不受限於膜 習用技術之缺點 電極組老化之轉魏_供應方法來解決 【發明内容】 法，读之主要目的係為提供一種燃料電池之燃料供應方 得燃料ΐ二，進而改變燃料濃度與供應量’使 枓電池燃料使収率之目的。 … 法，月之人要目的係為提供一種燃料電池之燃料供應方 斯！哉里測燃料電池於反應時之特徵值變化’達到判 、載疋否改變之目的。 法，的另一目的是提供一種燃料電池之燃料供應方 電流或者ΐ咖:料電池於反應時之特徵值變化，例如電壓、 泠旦2疋功率，進行數值運算與比較，進而控制燃料之供 ^ π# a去燃料感測器之設置，達到降低成本與控制燃料電 池敢佳功率輸出之目的。 电 為了達到上述之目的，本發明提供一種燃料電池之燃 應方法、，其係包括有下列步驟：（a)注人-特定量之燃料於一 燃料電池内’（b)釺對負載量測—燃料電池特徵值，並根據該 燃料電池特徵值決定-監控⑹於該監控時間内，隨時 憤測負載並判斷是否有顯著變化；以及⑷如果有變化的話， 則根據負載變化情形，選擇回到該步驟（a)以及（b)其中之一。 車乂佳的疋，该燃料特徵值係可選擇為燃料電池之電流、電 壓、功率、及其組合其中之一者。 較佳的是，判斷該負載是否有變化之方法係為於該監控時 間内隨時偵測該燃料電池之特徵值變化量是否超過預定門檻 值來判定。該特徵值係可選擇為燃料電池之電流、電壓以及 功率及其組合其中之—。 較佳的是，該步驟（d)更包括有下列步驟：（dl)如果該負 載增加的話，則回到步驟（a);以及（d2)如果該負載減少的 居，則回到步驟（b)。其中判斷該負載增加或減少之方式係為 判斷該燃料電池特徵值之變化是否由低變高或由高變低:其 方式係為根據該燃料電池特徵值變化曲線之斜率為正值或負 值來進行。而該簡電池雖值係可選擇_:料電池之 、電壓、功率及絲合其中之—。此外，亦可根據前後 時間點特徵值的差異為正值或負值來進行判斷。為了達到上 述之目的，本發明更提供一種燃料電池之燃料供應方法，其 係包括有下列步驟：（a)注入一特定量之燃料於一燃料電池 内；（b)針對負載量測一燃料電池特徵值，並根據該燃料電池 特徵值決定一監控時間；（c)於該監控時間内，判斷負載是否 有顯著變化；（d)如果有顯著變化的話，則根據負載變化情形 選擇回到步驟以及（b)其中之一；以及（e)如果沒有變化的 栝，則於S亥監控時間内判斷該特定量燃料是否足夠。 士較佳的是，該步驟（e)更包括有下列步驟：（el)於該監控 時間區間内擷取該燃料電池之一第一特徵值；（e2)到達該監 控時間時則擷取該燃料電池之一第二特徵值；（e幻將該第二 特徵值與該第一特徵值進行比較，如果該第二特徵值小於該 第一特徵值則回到步驟（a)注入燃料於該燃料電池内。（e4)如 果該第二特徵值大於該第一特徵值，則於監控時間區間後一 時間點前偵測一第三特徵值；（e5)於該時間點時擷取該燃料 1344718 電池之-第四特徵值；(e6)判斷該燃料電池 ，過門難，如果超過該Η健刺步驟⑷；（e7= = ，過該fm值’則將該第三特徵值與該第四特徵 如果該第四雜值小於_三特徵__步驟⑷進行 該第三特徵值則_該步驟（=e8)如果料四特徵值大於 其中該第-⑽㈣Μ於該監 壓最小值、觸小值、功％、㈣ 2 =之: 此外，該第-特徵值亦可^該監控期間内之^^值 測得該燃料電池之特徵值的移動平均值外=斤 之-時間區段所測得該燃料電池之特徵:控值月曰内 Mean Square, RMS)或其他數學 〜万根值（Root 用以與第三賴值職_，心來的崎值，設立 趨勢。 』Μ判疋燃料電池特徵值曲線 該第三特徵值係為於該時間 燃料電池之-特徵值的移動平均信=《4間區段所測得該 數學計算出來職計值。者是姑根值或者是由 較佳的是，該步驟（e)更包括右 控時間點賴擷取計算該燃料雷列步驟：（el)到達該監 如果該第-斜率為負值則4電步：徵㈣^ 料電池内；（e3)如果該第一斜率&進仃注入燃料於該燃 判斷該燃料電池特徵值之第二斜率.值則於一時間點量測並 徵值之變化是否超過該門檻值^如（e4)判斷該燃料電池特 步驟（d) ; (e5)如果未超過該門松超過邊門檻值則回到該 果第二斜率為負值則回到步驟(:\叫斷該第二斜率’如 退订注入燃料於該燃料電 1344718 ’池内;以及(⑹如果第二斜率為正，則回到該步驟欧 【實施方式】 •’ 為使貴審查委貝能對本發明之牡辨 * ^ 货乃之将徵、目的及功能有更 • 進一步的認知與晴解，下文特將太β # ^ 谓林明之裝置的相關細部結 構以及設計_念原由進行說明，以使得 本發明之特點，詳細說明陳述如下： 一委貝了 • tf參閱圖一所示，該圖係為本發明之第-較佳實施例示意 圖。該燃料電池之_供應方法，其係包括有下列步驟：首 先進行步驟10 ’注入-特定量之燃料於—燃料電池内。該燃 料係為-供應於燃料電池之富氫燃料，如質子交換膜辦料電 池（Polpier Ekctrolyte Fuel Cell，pEFC)。該富氫燃料 係為甲醇或乙醇或蝴氫化合物。除此之外，該富氫燃料也不 限於液體，例如應用於質子交換膜（Pr〇tcm Exchange Membrane Fuel Cell，PEMFC)的氫氣也可以，亦即凡供應於 φ 各種不同的燃料電池的燃料皆適用，本文舉直接甲醇燃料電 池（Direct Methanol Fuel Cell, DMFC)之曱醇供應為例。然 ' 後進行步驟11 ’針對負載量測一燃料電池特徵值，並根據該 燃料電池特徵值決定一監控時間。該燃料電池特徵值為燃料 電池所輸出之電壓 '電流、功率及其組合其中之一。 接下來說明本發明之監控時間決定之方式，請參閱圖二所 示’該圖係為燃料電池接收特定量燃料時之極化曲線示意 圖。該極化曲線圖係為最佳條件下電壓與電流之關係圖以及 功率與電流之關係曲線圖。在圖示中，最佳條件下該功率係 有一最大值Pn>ax ’而燃料電池之最短監控時間，可以用在該特 定量燃料注人到燃料電池後，維持負« W於最大功率P_ 之時間長度。 ' 因此’在圖-步驟11 t針對負載量測—辦料電池特徵值 之後（該燃料電池特徵值以燃料電池所輸出之電流為例），即 可根據該燃料電池特徵值與燃料電池可輪出之最大電流相 前述紅監叫㈣行調整。例 出簡電池之最短監控時間為1分鐘其輸 值：】二*培，而在步驟11所偵測到的燃料電池特徵 ΐ:=:由比例換算，則可得知步驟11所決定之 Ια控時間為5分鐘。 二U之後，判斷負載是否有變化。 ^ 疋叆化之方法係為於該監控時間内根據該 燃枓電池特徵值變化量是否和 ^ 疋否超過—門檻值來判定。該特徵值 糸可選擇為燃料電池之電流、電壓、功率及其組合之一者。 果有又化的4貝]再進仃步驟13判斷 加還是減少。 反之如果沒有變化的話，則推、 則進仃步驟14於該監控時間内判斷 邊特定量之燃料是否足夠。 接下來說明圖-之步驟u至步驟13如何判斷。為了說明 如何判斷步驟η至步㈣，首先說明，燃料電池裝置之結構。 凊茶閱圖三所示，該圖係為太麻An 土… b〜為本發明之燃料電池示意圖。該燃 料電池4基本上包括有供庫甲赌& 灿 ，。n 應甲蛘燃料以及氧氣（空氣）之管 路女以及排出水以及二氧化碳之管路。她料電池4的令間 則有一陽極板41、一陰極板4〇 極板41與_板40 及1子交無42。沾 ㈣以及該負載5形成-迴路=二使該陽極板41、陰f °亥負栽5連接有一量測儀6, 該量測儀6可為電壓計或 測儀係為-電流計，因此C計，在本實施射，該量 如果該量測儀6為電壓^心計係與該負載串聯。此外， 之電壓。而控制單元7 ^ #，則採並聯接法，以量測負載端 6 ^ ,χ 元8，進行供給簡之動=4㈣斷之結果控韻料供應單 料進仃步驟12酬負載是否有變化時，係透過偵測該辦 枓電池之特徵值的轡务θ τ , 打边貝列邊a 四A以否超過—服值。請參關三、圖 載增加之電流變化示μ 係為燃料電池反應過程負 备哉 :圖，而圖四β係為燃料電池反應過程 =減ν之電流變化示意圖。在圖四Α中，燃料於τ。的時間 2注人’此時燃料電池電化學反應*產生能量供應負载5 電k。雜㈣單元7會根__電池所輸出之特徵值， 在本實施例中為燃料電池輸出之電流，來決定—監控時間 T_。接著，控制單元7會定期根據電流計所量測之電二，二 斷在兩時間點的變化’例如：在圖四Α中時間.點1時之量測 點301所量測的電流丨，與時間點Tz之量測點3〇2所量測的電 流h的差異變化百分比ΔΙ/L如果超過門檻值的話，則判定 負載有變化。在本實施例中，該門檻值係為2〇%，7|與丁2通常 間隔少於1秒内，甚至於到1微秒，但不以此為限，使用者 可以根據需求而自行設定該門檻值與時間間隔之大小，門檀 值的設立通常是用以偵測並適應動態負載變化，同時區別過 濾量測電路本身的雜訊與誤差。 如果超過該門檻值的話，則會進行步驟13。在步驟13中， 判斷負載增加還是減少的方式，其中判斷該負載增加或減少 1344718 之方式係為判斷該燃料電池特徵值之變化是否由低變高。其 方式係為根據該特徵值變化之斜率為正值或貞值來進行判斷 或者根據轉徵值之前後時間點的差異為正值或負值來進行 判斷。以圖四A為例，在圖中，h係大於I,，因此在時間點 T2時會進行步驟1〇重新注入特定量之燃料於該燃料電池 内’然後繼續後續之步驟，並產生新的監控時間T—。反之， 如圖四β所示’雖然量測點402之電流12與量測點401之電 流L·之間的差異變化百分比超過門檻值，但由於電流^小於 電流1| ’負裁減少既有燃料夠用，因此在時間點丁2的時候， 並不會注入燃料，而是進行步驟11，根據量測之電流大小， 重新建立另一個監控時間，其長度為丄344718 IX. Description of the Invention: [Technical Field] The present invention relates to a fuel cell fuel supply control method, in particular, in the fuel cell reaction process, by measuring the fuel cell in the reaction process The characteristic value changes to change the load, and then the (four) material feeder improves the fuel cell operation efficiency--the material supply method. . [Prior Art] A fuel cell is a power generating device that uses an electrochemical reaction to output a chemical energy as an electric energy. The working principle is that the fuel containing hydrogen and the oxidant (such as air or oxygen) are respectively delivered to the anode and the cathode of the battery, the anode generates oxygen, the reaction dissociates the fuel into hydrogen ions and electrons, and the hydrogen ions are exchanged from the anode through the proton parent. The cathode is combined with the electrons that are conducted to the cathode through the external load circuit, and is made into water with the oxygen. Only the money will continue to be used for ship materials, and the battery can be used to generate electricity. With its two characteristics of high efficiency and low pollution, this technology has been widely recognized since its development. Among fuel cells, direct methanol fuel cells (Direc1: Methan〇1 heJCell'DMFC, hereinafter referred to as DMFC) can be more conveniently used in the power supply of various types of electrical appliances (notebooks, pDA, Gps).燃料 The fuel cell system that countries have actively invested in in recent years. The difference between DMFC and other technologies such as PEMFC is that its fuel is replaced by liquid methanol instead of hydrogen as fuel, which greatly improves the convenience and safety of fuel cell fuel storage and transportation. However, in the fuel supply of DMFC, anyone familiar with the technology knows that a super-fuel supply (such as sterol) can cause serious fuel (methanol) to pass through 1344718 (methane^cr〇ss〇ver) at the cathode. The alcohol directly reacts with oxygen to produce a mixed potential drop of Omxed PGtentlal), which leads to a poor efficiency of the battery. In severe cases, it will cause a negative voltage phenomenon and damage the battery. In addition, in order to meet the load demand of the fuel cell, it is important to control the amount of fuel supplied to the phase. For example, in the egg technology, a fuel rolling sensor, such as a methanol concentration sensor, is used to sense the state of the fuel (methylation) concentration to determine the timing of supplying fuel (methanol). In addition, in the surgery, it has been exposed as a sterol sensor by means of t chemical measurement. However, the 4 methods have a lot of money, and the complexity is increased by the cost. -, the fuel concentration system made by Xiang Electrochemical Method because of electricity = aging phenomenon, so the fuel thick money measurement (four) should be corrected frequently, to take the measurement of the 1 phase of the physical method to reduce the concentration of the dragon material (four) = temperature The function '#measurement data usually needs to be corrected or compensated for temperature, so the influence of control precision increases the complexity of control. The way the Handu sensor is used will increase the cost and how Li Fangfu will make the money (3) as a method. In the technology disclosed in the US patent us =, the current of the fuel cell is detected, and the temperature of the fuel cell J is used to judge the concentration of the methanol fuel, which is an indirect measurement method for the aging of the membrane electrode group. influences. In addition, another type of patent road current profile voltage disclosed by Gui (4) Qian 6,99 (10), as a short-term method of the fuel supply method, the battery should be opened or short-circuited, which is easy to cause damage to this technology. (4) The stability and life of the battery. ·., "1 pool itself can be integrated into the above, so it is necessary to avoid the use of sensors and is not limited by the shortcomings of the film-use technology. The main purpose is to provide a fuel cell fuel supply to the fuel, and then to change the fuel concentration and supply amount to make the battery fuel for the purpose. ... Law, the person of the moon wants to provide a fuel supply for fuel cells! In the case of measuring the change in the characteristic value of the fuel cell during the reaction, the purpose of the determination is to change or not. Another purpose of the method is to provide a fuel cell fuel supply current or a change in the characteristic value of the battery during the reaction, such as voltage, power, and numerical calculation and comparison, thereby controlling the supply of fuel. ^ π# a go to the fuel sensor settings to achieve the goal of reducing costs and controlling the fuel cell's power output. In order to achieve the above object, the present invention provides a fuel cell burning method, which comprises the following steps: (a) injecting a certain amount of fuel into a fuel cell '(b) 釺 loading measurement - a fuel cell characteristic value, and determined according to the fuel cell characteristic value - monitoring (6) during the monitoring time, instigating the load at any time and judging whether there is a significant change; and (4) if there is a change, according to the load change situation, selecting back One of the steps (a) and (b). In the case of Che Yujia, the fuel characteristic value can be selected as one of the current, voltage, power, and a combination of the fuel cell. Preferably, the method for determining whether the load changes is determined by detecting whether the amount of change in the characteristic value of the fuel cell exceeds a predetermined threshold value during the monitoring time. The characteristic value can be selected as the current, voltage and power of the fuel cell and a combination thereof. Preferably, the step (d) further comprises the steps of: (dl) returning to step (a) if the load is increased; and (d2) returning to step (b) if the load is reduced. ). The manner of determining whether the load is increased or decreased is to determine whether the change in the characteristic value of the fuel cell is changed from low to high or from high to low: the mode is based on whether the slope of the fuel cell characteristic value curve is positive or negative. Come on. The value of the battery is _: battery, voltage, power and silk. In addition, the judgment may be made based on whether the difference in the feature values of the front and rear time points is a positive value or a negative value. In order to achieve the above object, the present invention further provides a fuel supply method for a fuel cell, comprising the steps of: (a) injecting a specific amount of fuel into a fuel cell; and (b) measuring a fuel cell for a load amount. The characteristic value, and determining a monitoring time according to the fuel cell characteristic value; (c) determining whether the load has a significant change during the monitoring time; (d) if there is a significant change, selecting the returning step according to the load change situation and (b) one of them; and (e) if there is no change, it is judged whether the specific amount of fuel is sufficient during the monitoring time of Shai. Preferably, the step (e) further comprises the steps of: (el) capturing a first characteristic value of the fuel cell during the monitoring time interval; and (e2) obtaining the current time when the monitoring time is reached. a second characteristic value of the fuel cell; (e magically comparing the second characteristic value with the first characteristic value, and returning to the step (a) injecting fuel if the second characteristic value is less than the first characteristic value (e4) if the second characteristic value is greater than the first characteristic value, detecting a third characteristic value before a time point after the monitoring time interval; (e5) capturing the fuel at the time point 1344718 The fourth characteristic value of the battery; (e6) determining that the fuel cell is difficult to pass, if the step (4) is exceeded, (e7==, the fm value is passed), then the third characteristic value is the fourth If the fourth impurity value is less than the _three feature __ step (4) to perform the third eigenvalue _ the step (= e8) if the fourth eigenvalue is greater than the first - (10) (four) Μ at the minimum value of the monitoring voltage, the small value , work%, (4) 2 =: In addition, the first feature value can also be ^^ value during the monitoring period The moving average of the characteristic value of the fuel cell is the characteristic of the fuel cell measured by the kilogram-time segment: Mean Square, RMS in the month of the control value or other mathematical value of 10,000 roots (Root is used with The third affiliation _, the heart of the rugged value, set the trend. 』 Μ 疋 fuel cell characteristic value curve The third characteristic value is the moving average letter of the fuel cell - characteristic value at that time = "4 sections The mathematically calculated job value is measured as the value of the root or is preferably, the step (e) further includes the right control time point to calculate the fuel mine column step: (el) arrive at the If the first slope is negative, then 4 steps: (4) in the battery; (e3) if the first slope & injecting fuel into the fuel to determine the second slope value of the fuel cell characteristic value. Then, at a time point, measure whether the change of the levy exceeds the threshold value. (e4) determine the fuel cell special step (d); (e5) return to the fruit if the door sling exceeds the threshold value. If the second slope is negative, return to the step (:\call the second slope), such as unsubscribing to inject fuel into the fuel. Electricity 1344718 'in the pool; and ((6) if the second slope is positive, then return to this step Europe [Embodiment] • 'In order to make the review of the invention of the invention, the purpose, function and function There are more • further cognition and clearing. The following is a description of the relevant details of the structure of the device and the design of the device, so that the characteristics of the present invention are described in detail as follows: 1f is a schematic view of a first preferred embodiment of the present invention. The fuel cell supply method includes the following steps: first, step 10 'injecting - a specific amount of fuel to - Inside the fuel cell. The fuel is a hydrogen-rich fuel supplied to a fuel cell, such as a Polpier Ekctrolyte Fuel Cell (pEFC). The hydrogen rich fuel is methanol or ethanol or a hydrogen compound. In addition, the hydrogen-rich fuel is not limited to liquids, for example, hydrogen applied to a proton exchange membrane (Pr〇tcm Exchange Membrane Fuel Cell, PEMFC), that is, a fuel supplied to various fuel cells of φ. For example, the sterol supply of Direct Methanol Fuel Cell (DMFC) is taken as an example. Then, 'step 11' is followed to measure a fuel cell characteristic value for the load, and a monitoring time is determined according to the fuel cell characteristic value. The fuel cell characteristic value is one of the voltage 'current, power, and a combination thereof outputted by the fuel cell. Next, the manner in which the monitoring time of the present invention is determined will be described. Referring to Figure 2, the figure is a schematic diagram of the polarization curve of the fuel cell when it receives a specific amount of fuel. The polarization plot is a plot of voltage vs. current versus power versus current for the best conditions. In the figure, under the optimal conditions, the power system has a maximum value Pn>ax' and the shortest monitoring time of the fuel cell can be used to maintain the negative power of the maximum power P_ after the specific amount of fuel is injected into the fuel cell. length of time. Therefore, after the graph-step 11 t is for the load measurement-loading battery characteristic value (the fuel cell characteristic value is taken as the output current of the fuel cell), the fuel cell characteristic value and the fuel cell can be rounded according to the fuel cell characteristic value. The maximum current phase is adjusted by the aforementioned red squad (four) line. For example, the shortest monitoring time of the battery is 1 minute, and the value of the battery is: 2 amps, and the fuel cell characteristics detected in step 11 =:=: by scaling, the Ια determined in step 11 can be known. The control time is 5 minutes. After the second U, it is judged whether the load has changed. ^ The method of deuteration is determined based on whether the amount of change in the characteristic value of the burning battery and the value of the threshold value exceeds the threshold value during the monitoring time. The characteristic value 糸 can be selected to be one of a current, a voltage, a power of the fuel cell, and a combination thereof. If there is a re-introduction of 4 b], then step 13 to determine whether to add or reduce. Otherwise, if there is no change, then push, then step 14 determines whether a certain amount of fuel is sufficient during the monitoring time. Next, how to judge from step u to step 13 of the figure will be explained. In order to explain how to judge step η to step (4), first, the structure of the fuel cell device will be explained. See the picture shown in Figure 3, the figure is Taima An soil... b~ is a schematic diagram of the fuel cell of the present invention. The fuel cell 4 basically includes a gambling & n Pipes for beryllium fuel and oxygen (air) and pipes for draining water and carbon dioxide. In the case of the battery 4, there is an anode plate 41, a cathode plate 4, a plate 41 and a plate 40 and a branch. Dip (4) and the load 5 forming - loop = two, the anode plate 41, the negative f ° He negative plant 5 is connected to a measuring instrument 6, the measuring instrument 6 can be a voltmeter or a measuring instrument is an ammeter, therefore In the case of C, this amount is measured if the measuring instrument 6 is a voltage system in series with the load. In addition, the voltage. The control unit 7 ^ #, then adopts the parallel connection method to measure the load end 6 ^ , the unit 8 , and the supply of the simple motion = 4 (four) breaks the result of the control of the rhyme supply material billing step 12 whether the load is changed At the same time, by detecting the characteristic value θ τ of the battery of the battery, the edge of the edge of the column A 4 is exceeded or exceeded. Please refer to Figure 3. The increased current change shows that the μ system is the fuel cell reaction process 哉: Fig. 4, and the Fig. 4 β is the fuel cell reaction process = minus ν current change diagram. In Figure IV, the fuel is at τ. The time of 2 injections' at this time the fuel cell electrochemical reaction* produces an energy supply load of 5 k. The (4) unit 7 will determine the characteristic value outputted by the battery, in this embodiment, the current outputted by the fuel cell to determine the monitoring time T_. Then, the control unit 7 periodically measures the change of the two points at the two time points according to the estimometer, for example, the current measured by the measuring point 301 at time 1 in FIG. If the percentage change ΔΙ/L of the current h measured at the measuring point 3〇2 with the time point Tz exceeds the threshold value, it is determined that the load changes. In this embodiment, the threshold value is 2〇%, and the interval between 7| and 2 is usually less than 1 second, or even 1 microsecond, but not limited thereto, the user can set it according to the requirement. The threshold value and the time interval are set, and the threshold value is usually set to detect and adapt to the dynamic load change, and at the same time distinguish the noise and error of the filter measurement circuit itself. If the threshold is exceeded, then step 13 is performed. In step 13, a manner of determining whether the load is increased or decreased is determined, wherein the manner of determining whether the load is increased or decreased by 1344718 is to determine whether the change in the characteristic value of the fuel cell is changed from low to high. The method is based on whether the slope of the change in the characteristic value is a positive value or a 贞 value, or whether the difference between the time points before and after the conversion value is a positive value or a negative value. Taking Figure 4A as an example, in the figure, h is greater than I, so at time T2, step 1 will be performed to re-inject a certain amount of fuel into the fuel cell' and then continue the subsequent steps and generate new Monitoring time T-. On the contrary, as shown in FIG. 4β, although the percentage change of the difference between the current 12 of the measuring point 402 and the current L· of the measuring point 401 exceeds the threshold value, since the current ^ is smaller than the current 1| 'the negative cutting is reduced The fuel is sufficient, so when the time is 2, the fuel is not injected, but step 11 is performed, and according to the magnitude of the measured current, another monitoring time is established, and the length is
Tmc»n2，。 請參閱圖五所示，該圖係為本發明之燃料電池之燃料供應 方法之第二較佳實施例流程示意圖。燃料電池之燃料供應方 法其係包括有下列步驟：首先進行步驟201，注入一特定量之 燃料於一燃料電池内。該燃料係為一富氫燃料。該富氫燃料 係為一甲醇或乙醇或硼氫化合物。除此之外，該富氫燃料也 不限於液體，例如氮氣也可以。該燃料電池之結構如圖三所 示，其細部說明如前所述，在此不做贅述。 請參閱圖六Α所示，該圖係為燃料電池反應過程中之電流 變化示意圖。在圖六Α十，特徵值係以電流來代表。然後進 行步驟202 ’針對負載量測一燃料電池特徵值’並根據該燃料 電池特徵值決定一監控時間Τ_ι。接著於該監控時間T—内判 斷燃料電池之特徵值變化量百分比是否超過門檻值。該特徵 值可為電流、電麼、功率之一或二·者間之任思組合之一。忒 監控時間Τηκ>η|之決定方式如前所述，在此不做贅述。 1344718 在步驟203中之門植值為20%，亦即如果前後特徵值、，里 量超過前值的20%(變化量百分比），則視為滿足步驟^⑽是異 . 門檻值之大小可以根據情況與經驗而定，並不以本實施例: • 之20%為限。以圖六A為例，在時間點η之量測點5〇^二=二 丨2係超過時間點Τι時之量測點501的電流I,，且罢sw 左兵里超過 - [的20% ’因此會進行步驟204，判斷特徵值的變化是否由低 變高。而步驟204之判斷的方式可以根據量測點5〇1之電^ Ιι與量測點502之電流I2之差異是正值或者是負值，或者$ • 根據量測點501與量測點502之間的斜率為正或負來進行判 斷。量測點501與量測點502之時間間隔（τντ,)則可以根據 實際負載隨時間變動的情況而定。 • 在圖六A中’因為電流I2大於電流h，因此會回到步驟 201，於時間點T2重新注入特定量之燃料，然後以步驟2〇2決 定新的監控時間Town2。在監控時間Tnwn2之過程中，並未滿足步 驟203,因此會進行判斷還沒消耗完畢之燃料是否足夠的程 序。 ® 而判斷燃料是否足夠的程序’首先利用步驟205，於該監 • 控時間T_内擷取燃料電池之一第一特徵值。該第一特徵值， 可為於該監控期間内所量測到之電流最小值、電壓最小值、 功率最小值及其組合其中之一的最小值。在本實施例中係利 用電流來進行判斷’亦可用功率來進行判斷，功率特徵值為 電壓特徵值、電流特徵值之組合。請參閱圖六β所示，該圖 係為燃料電池反應過程中之功率變化示意圖。一般而言，燃 料電池之應用在某特定時間内可視為固定阻抗，固定電麈， 或者是固定電流等三種固定負載情況之一，例如：燃料電池 1^44/1» if記型電腦的電源供應，就可以視為固定阻抗，因 圖」βΠ曲線基本上是與圖六A之電流曲線-致，如 精ϊ的效值的編有增力補解析度與 是固定電壓之情況::二上•並不限制在固定阻抗或者 ^ +凡在圖六Α之實施例中第一特徵值係為電 =最 亦即為量測點503之電流l3。除此之外，該第一 j值亦可為H監控期間内之_時間區Tmc»n2,. Referring to FIG. 5, the figure is a schematic flow chart of a second preferred embodiment of a fuel supply method for a fuel cell of the present invention. The fuel cell fuel supply method includes the following steps: First, step 201 is performed to inject a specific amount of fuel into a fuel cell. The fuel is a hydrogen rich fuel. The hydrogen rich fuel is a methanol or ethanol or a boron hydride compound. In addition to this, the hydrogen-rich fuel is not limited to a liquid, and for example, nitrogen may be used. The structure of the fuel cell is shown in Fig. 3. The detailed description thereof is as described above, and will not be described herein. Please refer to Figure 6Α, which is a schematic diagram of current changes during fuel cell reaction. In Figure XX, the characteristic values are represented by current. Then, step 202 is performed to measure a fuel cell characteristic value for the load and a monitoring time Τ_ι is determined based on the fuel cell characteristic value. Then, within the monitoring time T-, it is judged whether the percentage change of the characteristic value of the fuel cell exceeds the threshold value. The characteristic value can be one of current, electricity, power, or a combination of two.监控 The monitoring time Τηκ> η| is determined as described above and will not be described here. 1344718 In step 203, the phyton value is 20%, that is, if the eigenvalues before and after the eigenvalue exceeds 20% of the previous value (% change), it is considered that the step ^(10) is different. The threshold value can be Depending on the situation and experience, it is not limited to 20% of this embodiment: •. Taking Figure 6A as an example, at the time point η, the measuring point is 5〇^2=2丨2 is the current I of the measuring point 501 exceeding the time point Τι, and the stroke is more than - [20 % ' Therefore, step 204 is performed to determine whether the change in the feature value is changed from low to high. The method of determining in step 204 may be based on whether the difference between the voltage of the measuring point 5〇1 and the current I2 of the measuring point 502 is a positive value or a negative value, or $ • according to the measuring point 501 and the measuring point 502. The slope between the positive or negative is judged. The time interval (τντ,) between the measuring point 501 and the measuring point 502 can be determined according to the actual load variation with time. • In Figure 6A, because current I2 is greater than current h, it returns to step 201, refilling a specific amount of fuel at time point T2, and then determining a new monitoring time, Town2, in step 2〇2. In the course of monitoring the time Tnwn2, the step 203 is not satisfied, so that a procedure for judging whether or not the fuel that has not been consumed is sufficient is performed. The procedure for determining whether the fuel is sufficient is first used in step 205 to draw a first characteristic value of the fuel cell during the monitoring time T_. The first characteristic value may be a minimum value of one of a current minimum value, a voltage minimum value, a power minimum value, and a combination thereof measured during the monitoring period. In the present embodiment, the current is used for the judgment. The power can be used for the judgment. The power characteristic value is a combination of the voltage characteristic value and the current characteristic value. Please refer to Figure 6 for a diagram showing the power variation during the fuel cell reaction. In general, the application of a fuel cell can be regarded as one of three fixed load conditions, such as fixed impedance, fixed power, or fixed current, for a certain period of time, for example: fuel cell 1^44/1» Supply, it can be regarded as a fixed impedance, because the graph "β Π curve is basically the current curve with Figure 6A - such as the efficiency of the fine 编 编 增 增 增 增 增 增 增 增 增 增 增 增 增 增 增 增 增 增 增The upper limit is not limited to a fixed impedance or the first characteristic value in the embodiment of Fig. 6 is electric = most is the current l3 of the measuring point 503. In addition, the first j value may also be the _ time zone during the H monitoring period.
池4!徵值的移動平均值或者是該監控期_之-時間區= 測付料電池特徵值的均方根值（Root Mean Square，RMS) ^或者是由相關數學方法計算出來之統計值，可有很多種定 義。 ，接著’進行步驟206 ’到達該監控時間時，擷取該燃料電 池之一第二特徵值。該第二特徵值係可為電壓、f流或者是 功，及其組合之_，在圖六A之實施例中，係為功率，亦即 為量測點5G4之電流卜織進行步驟2G7，判斷該第二特徵 值是否小於該第一特徵值。如果該第二特徵值小於該第一特 徵值的話，則代表燃料已經瀕臨消耗殆盡之邊緣或者是消耗 殆盡，因此會回到步驟201重新注入燃料。反之，如果該第 二特徵值大於該苐一特徵值，則代表燃料還沒有用完。以圖 六A說明，量測點503之電流h係小於量測點5〇4之電流[， 因此代表著還有多餘的燃料尚未被用完。The moving average of the pool 4! eigenvalue is the root mean square value (Root Mean Square, RMS) of the monitoring period _-time zone = the measured battery eigenvalue ^ or the statistical value calculated by the relevant mathematical method There are many definitions. Then, when the step 206 is reached to the monitoring time, a second characteristic value of the fuel cell is retrieved. The second characteristic value may be voltage, f-flow or work, and a combination thereof. In the embodiment of FIG. 6A, the power is, that is, the current of the measuring point 5G4 is performed in step 2G7. Determining whether the second feature value is less than the first feature value. If the second characteristic value is less than the first characteristic value, it means that the fuel is already on the verge of exhaustion or is exhausted, so that it returns to step 201 to refill the fuel. On the other hand, if the second characteristic value is larger than the first characteristic value, it means that the fuel has not been used up. As illustrated in Fig. 6A, the current h of the measuring point 503 is smaller than the current of the measuring point 5〇4, so that there is still excess fuel that has not been used up.
在燃料還足夠的情況下，會繼續進行步驟208，於一時間 點506前，偵測一第三特徵值。在圖六A中該第三特徵值係 為量測點505之電流I5。之後，會進行步驟209，於該時間點 時戴取該燃料電池之一第四特徵值。該第四特徵值在圖六A 16 1344718 中係為量測點506之電流Ie。接著進行步驟210，判斷該燃料 電池特徵值變化量是否超過門檻值，門檻值設定的法則如前 所述’本步驟之門檻值係為20%。在此步驟中，需回到圖六Α 判斷在此時間點之電流變化是否超過門檻值。在圖六Α中並 無超過門檻值之情形。此時再回到步驟211，判斷第四特徵值 是否小於該第三特徵值。在圖六Α中，電流I6係小於電流h， 滿足步驟211之條件，所以會回到步驟201，注入特定量燃料 於該燃料電池中，然後迴圈持續進行後續之步驟。反之‘，’、如 果第四特徵值大於第三特徵值的話，則代表著燃料還有剩 餘，因此會回到步驟208，持續擷取第三特徵值以及第四特徵 值以監控燃料是否耗盡。至於量測點505與量測點5〇6之日"夺 間間隔可以視需求而定，本實施例係為丨秒，但不以此為限。 4繼續參閱圖五與圖六A所示，當在時間點Te注入燃料 之後，會決定監控時間T_3。在該監控時間τ_3内，同樣;持 續進行步驟203判斷燃料電池之特徵值變化量是否超過&檻 值。在圖六Α中之監控時間Τ,3内之量測點5〇7與量測點5& 之間發生特徵值變化量超過門檻值之情況，因此會進行步驟 2〇4，判斷特徵值之變化是否由低變高。由於在監控時間 之電流18小於1?，其代表著負載變小，所以會進行步驟2〇2， ，據負載之變化’決定·監控時間τ_，然後持續進行後續 乂驟。如此反覆運行，以隨時調整燃料之供應維持燃料電池 之運作。 請參閱圖七所示’該圖係為本發明之燃料電池之燃料供應 2之第三較佳實施例流程示意®。在本實關巾係包射 々2〇至步，驟28,其大部分之步驟係與前述第二較佳實施例 1344718 相同’所不同是本實施例之步驟24與2B係利用斜率正負， 來判斷是否要補充燃料。在步驟24中，於該監控時間到達時 操取計算該燃料電池特徵值之第一斜率，然後進行步驟25判 断該第一斜率是否為正。如果該第一斜率為負值則回到步驟 20進行注入燃料於該燃料電池内。反之如果該第—斜率為正 值’則進行步驟26於一時間點量測並判斷該燃料電池特徵值 之第二斜率。然後進行步驟27判斷燃料電池特徵值變化是否 超過預設門檻值’步驟27係類似步驟22，如果斜率超過門播 值則回到該步驟23。反之如果未超過該門檻值，則進行步驟 28判斷步驟26所量測之第二斜率是否為正，如果第二斜率為 負值則回到步驟20進行注入燃料於該燃料電池内；如果第二 斜率為正’則回到該步驟26。 請參閱圖八所示’該圖係為利用本發明第二較佳實施例之 方法，由燃料電池所得到之功率輸出與負載變化關係曲線 圖，其為以25W為額定輸出之直接曱醇燃料電池系統供應電 能給筆記型電腦的實際量測實驗結果圖。其中圖八之曲線^^0 代表著通過負載之電流的變化，亦即代表著負載的變化，在 高負載3A時為筆記型電腦開啟時耗能情況，在低負載〇 8a 時為筆Z型電關p㈣耗能情況，直接曱賴料電池必須維 持自身BOP系統運轉。而區域9〇1内的曲線則代表著辦料電 池所輸出之功率的隨著負載變化之曲線。由圖中可知，利用 本叙明之方法，燃料電池之功率輸出的確可以隨著負載之變 動大小而自_整甲_料的配給以達到穩定的功率輸出， 而且從筆記腦的開關魏’ g載切換下，本方法不但可 滿足系統之取小f求’還可以自動找尋到最佳最大的功率輸 出點。 准以上所述者，僅為本發明之較者去 範圍。即大凡依本發明申請“範圍;做 =範圍，故都應視為本發明的進一步實施狀^ 圖，中之步驟m注入特定量之燃料，該特定量在前二 t例係為固定的燃料量，因此在步驟2Q2中的監控之貫 :因為負載之情況而有不i之時間長度。然則 201 科置，然後以固定的監控時間長度來實施。總之 = 供電之總能量係為燃料噴量與監控時間長度的函數，也 論是固定燃料注射量，紐㈣監控時間長度或者 = 料/主射里’固定監控時間長度都可以作為本發明之實施方气然 本發明之實施例雖以動態負載作為實施例說明，但= 之應用並料動態負載為限。例如：在一㈣電池:上 當燃料電池在開啟與關閉時，系統之負載可視為—變動’ 的情況’另外’如果燃料電池系統在開啟後之正常運作下栽 如果是充做充電器則可視為固定負載之情形。因此，透過本 發明之方法’可以應用於上述之動態負載與固定負載級合 情況。 α < …綜合上述，本發明提供之燃料電池之燃料供應方法，具有 動恕偵測負載變化以控制燃料供應，使得燃料電池輸出可以 順應負載之變化而調整，目此可以滿足業界之各種需求例如 汽車業與3G產轉’進而提高該產#之競爭力以及帶動週遭 產業之發展，誠已符合發明專利法所規定申請發明所需具備 19 1344718 之要件，故爰依法呈提發明專利之申請，謹請 允撥時間惠予審視，並賜准專利為禱。 貴審查委員In the event that the fuel is sufficient, step 208 is continued to detect a third characteristic value prior to a time point 506. The third characteristic value in Fig. 6A is the current I5 of the measuring point 505. Thereafter, step 209 is performed at which a fourth characteristic value of the fuel cell is taken. The fourth characteristic value is the current Ie of the measuring point 506 in FIG. 6A 16 1344718. Next, in step 210, it is judged whether the fuel cell characteristic value change amount exceeds the threshold value, and the threshold value setting rule is as described above, and the threshold value of the step is 20%. In this step, you need to go back to Figure 6 to determine if the current change at this point in time exceeds the threshold. There is no threshold value in Figure VI. At this time, returning to step 211, it is determined whether the fourth feature value is smaller than the third feature value. In Figure Α, current I6 is less than current h, satisfying the condition of step 211, so returning to step 201, a specific amount of fuel is injected into the fuel cell, and then looping continues for the subsequent steps. Conversely, ',', if the fourth eigenvalue is greater than the third eigenvalue, it means that there is still fuel remaining, so it will return to step 208 to continuously capture the third eigenvalue and the fourth eigenvalue to monitor whether the fuel is exhausted. . As for the date of the measurement point 505 and the measurement point 5〇6, the interval may be determined according to requirements, and this embodiment is a leap second, but is not limited thereto. 4 Referring to Figure 5 and Figure 6A, when the fuel is injected at the time point Te, the monitoring time T_3 is determined. In the monitoring time τ_3, similarly; step 203 is continuously performed to determine whether the amount of change in the characteristic value of the fuel cell exceeds the & 值 value. In the monitoring time in Figure Α, the eigenvalue change exceeds the threshold between the measuring point 5〇7 and the measuring point 5& and therefore step 2〇4 is performed to determine the characteristic value. Whether the change changes from low to high. Since the current 18 at the monitoring time is less than 1?, which means that the load becomes small, step 2〇2 is performed, and the time τ_ is determined according to the change of the load, and then the subsequent steps are continued. This is repeated to maintain the fuel supply at any time to maintain the fuel cell operation. Referring to Figure 7, the figure is a schematic flow diagram of a third preferred embodiment of the fuel supply 2 for a fuel cell of the present invention. In the present embodiment, the majority of the steps are the same as the second preferred embodiment 1344718. The difference between step 24 and 2B of the present embodiment is that the slope is positive or negative. To determine if you want to refuel. In step 24, a first slope of the fuel cell characteristic value is calculated when the monitoring time arrives, and then step 25 is performed to determine whether the first slope is positive. If the first slope is negative, return to step 20 to inject fuel into the fuel cell. Conversely, if the first slope is positive, step 26 is performed at a time point to determine and determine a second slope of the fuel cell characteristic value. Then, step 27 is made to determine whether the fuel cell characteristic value change exceeds the preset threshold value. Step 27 is similar to step 22, and if the slope exceeds the gate broadcast value, the process returns to step 23. Otherwise, if the threshold value is not exceeded, step 28 is performed to determine whether the second slope measured in step 26 is positive, and if the second slope is negative, return to step 20 to inject fuel into the fuel cell; if the second If the slope is positive, then return to step 26. Please refer to FIG. 8 ' is a graph showing the relationship between the power output and the load obtained by the fuel cell by using the method of the second preferred embodiment of the present invention, which is a direct sterol fuel with a rated output of 25 W. The actual measurement results of the battery system to supply power to the notebook. The curve of Fig. 8 ^^0 represents the change of the current through the load, which means the change of the load. When the notebook is turned on at a high load of 3A, the energy consumption is the case when the notebook is turned on. At the low load 〇8a, it is the pen type Z. The power supply p (four) energy consumption situation, the battery directly depends on the battery must maintain its own BOP system operation. The curve in the area 〇1 represents the curve of the power output from the storage battery as a function of load. As can be seen from the figure, with the method described in this way, the power output of the fuel cell can indeed be distributed with the load of the whole material to achieve a stable power output, and the switch from the note brain Under the switch, this method can not only satisfy the system's small f-seeking, but also automatically find the best and maximum power output point. The above is only the scope of the present invention. That is to say, according to the invention, the scope of the invention, the scope of the invention, should be regarded as a further embodiment of the invention, in which step m injects a specific amount of fuel, which is a fixed fuel in the first two cases. Quantity, therefore the monitoring in step 2Q2: because of the load situation, there is no length of time. However, 201 is set, and then implemented with a fixed monitoring time length. In short, the total energy of the power supply is the fuel injection amount. The function of the length of the monitoring time, which is also the fixed fuel injection amount, the length of the monitoring time of the neon (4) or the length of the fixed monitoring time of the material/main shot can be regarded as the implementation of the present invention. The load is described as an example, but the application is limited to the dynamic load. For example: on a (four) battery: when the fuel cell is turned on and off, the load of the system can be regarded as a "variable" situation. In addition, if the fuel cell system In the normal operation after being turned on, if it is charged as a charger, it can be regarded as a fixed load. Therefore, the method of the present invention can be applied to the above dynamic negative In combination with the fixed load, α < ...Integrally, the fuel supply method of the fuel cell provided by the present invention has the function of detecting the load change to control the fuel supply, so that the fuel cell output can be adjusted according to the change of the load. This can meet the various needs of the industry, such as the automotive industry and 3G production and turn to further enhance the competitiveness of the production and promote the development of the surrounding industry, Cheng has met the requirements of 19 1344718 required to apply for inventions under the invention patent law, so 爰To file an application for an invention patent in accordance with the law, please allow time to review and grant a patent as a prayer.
20 1344718 【圖式簡單說明】 圖ίτ、為本發明之燃料電池之燃料供應方法之第一較佳實施 . 例流程示意圖。 .·. 料電池在最佳條件時之極化曲線示意圖。 •圖三係為燃料電池系統示意圖。 圖四A係為燃料電池反應過程負朗加之電流變化示意圖。 圖四B係為燃料電池反應過程負載減少之電流變化示意圖。 • 圖五係為本發明之燃料電池之燃料供應方法之第二較佳實施 例流程示意圖。 圖六A係為燃料電池反應過程中之電流隨時間變化示意圖。 • 圖六B係為燃料電池反應過程中之功率隨時間變化示意圖。 圖七係為本發明之燃料電池之燃料供應方法之第三較佳實施 例流程示意圖。 圖八係為利用本發明第二較佳實施例之方法之燃料電池所得 到之功率輸出與負載變化關係曲線圖。 【主要元件符號說明】 1 -燃料電池之燃料供應方法 UM4-步驟 2-燃料電池之燃料供應方法 201〜211-步驟 20〜28-步驟 301、302-量測點 1344718 401、402-量測點 501〜508-量測點 4- 燃料電池 4 0 -陰極 41 -陽極 42-質子交換膜 5- 負載 6- 量測儀 7- 控制單元 8- 燃料供應單元20 1344718 BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a first preferred embodiment of a fuel supply method for a fuel cell of the present invention. ... Schematic diagram of the polarization curve of the battery under optimal conditions. • Figure 3 is a schematic diagram of the fuel cell system. Figure 4A is a schematic diagram of the current change in the fuel cell reaction process. Figure 4B is a schematic diagram showing the change in current during fuel cell reaction process load reduction. Figure 5 is a flow chart showing a second preferred embodiment of the fuel supply method for a fuel cell of the present invention. Figure 6A is a schematic diagram of current versus time in a fuel cell reaction. • Figure 6B is a graphical representation of power versus time over a fuel cell reaction. Figure 7 is a flow chart showing a third preferred embodiment of the fuel supply method for a fuel cell of the present invention. Figure 8 is a graph showing the relationship between power output and load change obtained by a fuel cell using the method of the second preferred embodiment of the present invention. [Description of main component symbols] 1 - Fuel cell fuel supply method UM4 - Step 2 - Fuel cell fuel supply method 201 to 211 - Step 20 to 28 - Step 301, 302 - Measurement point 1344718 401, 402 - Measurement point 501~508-measuring point 4 - fuel cell 4 0 - cathode 41 - anode 42 - proton exchange membrane 5 - load 6 - measuring instrument 7 - control unit 8 - fuel supply unit