200538492 九、發明說明： 【發明所屬之技術領域】 本發明係有關於一種一種製造聚酯或共聚酯的方法，它 經由對從終端反應器裏流出來的主流體的黏稠度進行測量 的方法來控制最終産品的黏稠度。 【先前技術】 製造聚酯或共聚酯的方法已爲人們所習知（專利號：us 3,499,873，EP 0 532 988，DE 35 44 551，US 6,582,080)。 傳統的方法是先將原料在一次酯化反應中變爲一種羥烷基 二碳酸單體或齊聚混合物。接著將這些已經部分酯化的齊 聚混合物進行一次先行的或預縮聚處理，這時會得到一種 預聚合物和縮聚産品以及基本上具有兩種功能的乙醇和水 組成的反應氣體。接著將這種預聚合物放到一個終端反應 器襄進行一次縮聚，以便將其聚合程度或聚酯的分子質量 調整到所期望的數值。US 3,617,225專利介紹的、適用於這 種用途的終端反應器就屬於一種習知的設備。 所期望的縮聚程度端視聚合物的用途而定。為了確保生 產的安全性，並使以這種聚合物熔體所製成最終産品能有 均勻的特性，必須嚴格按縮聚程度進行生産。 爲進行生産控制，通常採用直接對聚合物熔體的黏稠度 進行測量的方法，而不是測量其分子質量或縮聚程度。這 些數值的相互關係是人們習知的，很容易爲專業人員接 交’例如馬克修文（Mark-Houwink)關係式。（參見赫夫曼· 克略默·庫恩所著的《聚合物分析》斯圖加特1977年版） 100739.doc 200538492 通常經由在線（〇n_nne)使用黏稠度計的方法對從終端反 應器裏出來的熔體黏稠度進行控制。如果有必要，則根據 測量結果對終端反應器裏的條件，如壓力，溫度或停留時 間進行調整，使其達到所需要的數值。 這些在線使用的黏度計有著各種不同的測量原理·· 根據毛細管原理工作的黏度計測量的是從終端反應器裏 出來的溶體主流的其中-個分流的黏稠度，通常在測量後 使其重新並入主流。這種人們所習知的方法及其所使用的 裝置在：EP 0 595 276，WO 00166993 和 EP〇 7691569〇 等專 利中有詳細的介紹。這種方法的缺點在於： —需要購買昂貴的附加的測量儀器； ―其所測量的僅僅是一個小支流的特性； --受黏度計使用方法的影響（強力剪切，不同的停留時 間，不同的溫度）測量值會發生變化，並因此造成測量不準 確； —不容易馬上發現測量儀器的細微的調整變化，因而能 造成對反應器運行方式的負面影響； —需要對測：!：儀器進行附加的校正和維護。 專利US 6,405,579所介紹的、使用流體黏度計進行測量的 方法，同樣是人們所習知的一種方法。使用這種方法時， k安裝在主生産官道上的一個專用噴嘴裏不斷地有熔體的 -個小支流流出。舉例來說’可以經由一台天平對流出的 熔體進行連續不斷地測量。因爲從喷嘴流出的這一部分支 流始終具有和熔體黏稠度相同的條件，所以同樣可以使用 100739.doc 200538492 這種方法對熔體 調整。 的點稠度進行監控並對反應 g進行相應的 這種方法的缺點是·· —它所測量的同樣σ | . 像，、疋一個小支流的特性； —使用連續工作的天平， 八卞冋樣需要昂貴的配套儀器； -_由於支流是連續不斷从 、 斷也，爪出的，需要對天平和噴嘴經 常性的監控和清潔； 、、200538492 IX. Description of the invention: [Technical field to which the invention belongs] The present invention relates to a method for manufacturing polyester or copolyester by measuring the viscosity of the main fluid flowing out of the terminal reactor. To control the viscosity of the final product. [Prior art] Methods for manufacturing polyesters or copolyesters are well known (patent numbers: us 3,499,873, EP 0 532 988, DE 35 44 551, US 6,582,080). The traditional method is to change the raw material into a hydroxyalkyl dicarbonate monomer or oligomer mixture in one esterification reaction. These partially esterified oligomer mixtures are then subjected to a prior or pre-polycondensation process, at which point a prepolymer and polycondensation product and a reaction gas consisting of ethanol and water with essentially two functions are obtained. This prepolymer is then placed in a terminal reactor to perform a polycondensation in order to adjust the degree of polymerization or molecular weight of the polyester to a desired value. The terminal reactor described in the US 3,617,225 patent and suitable for this purpose is a conventional device. The desired degree of polycondensation depends on the use of the polymer. In order to ensure the safety of production and make the final product made of this polymer melt have uniform characteristics, it must be produced strictly according to the degree of polycondensation. For production control, it is common to measure the viscosity of a polymer melt directly, rather than measuring its molecular mass or degree of polycondensation. The correlations between these values are well known, and it is easy to transfer them for professionals, such as the Mark-Houwink relationship. (See Huffmann-Klimmer Kuhn's Polymer Analysis, Stuttgart, 1977.) 100739.doc 200538492 The melt from the end reactor is usually processed online (On_nne) using a viscometer. Body viscosity is controlled. If necessary, adjust the conditions in the end reactor, such as pressure, temperature, or residence time, to the required value based on the measurement results. These on-line viscometers have a variety of different measuring principles. The viscometers that work according to the capillary principle measure the viscosity of one of the mainstream streams of the solution coming out of the terminal reactor, usually after the measurement, they are re-established. Merged into the mainstream. This well-known method and the device used in it are described in detail in patents such as EP 0 595 276, WO 00166993 and EP 07691569. The disadvantages of this method are:-the need to purchase expensive additional measuring instruments;-it only measures the characteristics of a small tributary stream;-affected by the use of the viscometer (strong shear, different residence time, different (Temperature) measurement value will change, and therefore cause inaccurate measurement;-It is not easy to immediately find the slight adjustment changes of the measuring instrument, which can cause a negative impact on the operation mode of the reactor;-Need to measure:!: Additional corrections and maintenance. The method described in the patent US 6,405,579 using a fluid viscometer for measuring is also a method known to the people. When using this method, a small branch of melt is constantly flowing out of a special nozzle installed on the main production channel. For example, 'outgoing melt can be continuously measured via a balance. Because this part of the tributary flowing from the nozzle always has the same conditions as the melt viscosity, the same method can be used to adjust the melt using 100739.doc 200538492. The disadvantage of this method of monitoring the point consistency and corresponding reaction g is that it measures the same σ |. Like, the characteristics of a small tributary;-using a continuously working balance Requires expensive supporting instruments; -_ As the tributaries are continuously coming from, breaking off, and coming out of the claws, the balance and nozzles need to be monitored and cleaned frequently;
―會始終造成產品損失。 —種能避免支_量方法缺點的方法，是直接在溶體主 流裏使用旋轉式或㈣式液流儀（參見赫夫曼·克略默庫 恩所著的《聚合物分析》斯圖加年版）。這裏所採用 的是將由外部驅動的旋#體直接裝Α生産管道的方法，這 樣它就可以直接對流過其周圍的熔體進行測量。然而由於 生産管道中的溫度（>28〇。〇和壓力（>2〇巴），這種方法仍然 會在岔封，避免溶體流出，和死區形成等方面遇到各種各 樣的問題。甚至可能在圍繞旋轉體的層流層裏産生熱鏈結 終止作用。與熔體主流相比、這個層流層的物理特性仍可 能有改變並因此産生錯誤的測量結果。因此這種方法同樣 需要昂貴的儀器購置費用和維護費用，並且很容易受到干 擾0 【發明内容】 本發明之目的在於提供一種可以避免上述缺點的終端反 應器控制方法。 本發明採用了另外一種測量方法來實現對終端反應器的 100739.doc 200538492 &制知Γ…、這種對溶體主流進行測量的方法，最少要對一 個以上測點進行溫度和壓力差測量，並對溶體主流的體積 *里進订，則篁。在這種方法裏溶體的主流被稱爲炫體流， 匕私的疋在起縮聚作用的終端反應器和成型設備之間流動 的炼體流。所謂的成型設備是指使聚合物溶體經由凝固來 成型的设備。這裏所說的成型形態可以是纖維、膠片和薄 膜，也可以是要接著製成顆粒的條束和所有其他由聚合物 春 '熔體成5L的物體。可以作爲成型設備使用的設備有孔狀漏 板、扁平或圓形的隙縫式噴嘴、薄膜洗禱機或粒化器的逢 鑄頭。同樣可以使用壓鑄機來作爲成型設備，其成型形狀 可以是任意形狀。 從終端反應器裏出來熔體是經由一個輸送泵來進行傳送 的，在大多數情況下使用的是一個大型齒輪泵。齒輪泵是 按照容積進行工作的，也就是說它所輸送的容積與被輸送 的熔體的黏稠度無關，而是由泵的轉速決定的。 • 爲了排除其中的雜質，熔體主流是經過-個聚合物熔體 篩檢程式導出的。 根據本發明，可以從兩個按前後順序佈置在熔體主流中 的儲壓器上所測出的測量值、兩個儲壓器之間的熔體溫度 和已知的齒輪泵輸送的體積流量中計算出熔體黏稠度。如 果發現黏稠度偏離終端反應器的給定值，則對其進行調整 使其重新回到給定值。例如：在發現黏稠度低於給定值時 可以降低反應器裏壓力，而在黏稠度高於給定值時則相應 升高壓力。當然也不排除其他的調整方法。 100739.doc 200538492 按照哈根泊蕭流（HagenPoisseuille)方程進行推論，層狀 粘質流的壓力損耗Elp是和與流量vt相乘所得出的動態黏 稠度成正比的。可以從幾何體裏推導出正比例常數Kn的近 似值（比方說，在一個内徑爲0，長度爲乙的管子裏，這個常 數爲128*L/(ff*D4))。然而，更好的方法則是在設備起動時 經由同樣的方法從測量值裏確定這個常數，就象在使·用毛 細管原理的黏度計時測出設備常數的方法一樣。因爲許多 聚合物都具有一個與剪切有關的黏稠度，所以在已知的流 動定理上有時需要有一個附加的常數Kscher。 7 =〜*芋咖，似，…） V t 士果在輸送泵的後面（或在增壓泵的前面）只有一段未分 支的路段’在這種簡單的情況下，可以經由對遂力差，平―It will always cause product loss. —A method to avoid the disadvantages of the support method is to use a rotary or tritium flow meter directly in the mainstream of the solution (see Polymer Analysis Stuttgarian Edition by Huffman Klemerkuhn ). The method used here is to directly load the rotating body with external drive into the A production pipeline, so that it can directly measure the melt flowing around it. However, due to the temperature (> 28.0 and pressure (> 20 bar) in the production pipeline), this method still encounters various problems in the aspects of bifurcation, avoiding the outflow of the solution, and the formation of dead zones. The problem. Thermal termination may even occur in the laminar layer surrounding the rotating body. Compared with the melt mainstream, the physical properties of this laminar layer may still change and therefore produce incorrect measurement results. Therefore this method It also requires expensive equipment purchase and maintenance costs, and is easily susceptible to interference. [Abstract] The purpose of the present invention is to provide a terminal reactor control method that can avoid the above disadvantages. The present invention uses another measurement method to achieve 100739.doc 200538492 of the terminal reactor & Manufacturing Γ ... This method of measuring the mainstream of the solution requires at least one measurement point to measure the temperature and pressure difference, and the volume of the mainstream of the solution * In this method, the mainstream of the solution is called dazzling body flow, and the puppet is between the terminal reactor and the molding equipment that play a role in condensation polymerization. The moving smelting stream. The so-called molding equipment refers to the equipment that solidifies the polymer solution through solidification. The molding form mentioned here can be fibers, films and films, or strands and pellets to be made into particles. All other objects made of polymer spring's melt into 5L. Equipment that can be used as molding equipment are perforated leak plates, flat or round slot nozzles, film washing machines or granulators. The die-casting machine can be used as the molding equipment, and the molding shape can be any shape. The melt from the terminal reactor is transferred through a conveying pump. In most cases, a large gear pump is used. Gear pump It works by volume, which means that the volume it transports has nothing to do with the viscosity of the melt being transported, but is determined by the speed of the pump. • In order to exclude impurities, the melt is mainly subjected to a polymerization process. Derived from the melt screening program. According to the present invention, the measured values measured on two pressure accumulators arranged in the main flow of the melt The melt viscosity is calculated from the melt temperature between the compressors and the known volumetric flow from the gear pump. If the viscosity is found to deviate from the given value of the end reactor, adjust it to return it to For example: when the viscosity is found to be lower than the given value, the pressure in the reactor can be reduced, and when the viscosity is higher than the given value, the pressure is increased accordingly. Of course, other adjustment methods are not excluded. 100739.doc 200538492 According to the Hagen Poisseuille equation, the pressure loss Elp of the layered viscous flow is proportional to the dynamic viscosity obtained by multiplying the flow rate vt. The proportional constant Kn can be derived from the geometry. Approximate value (for example, in a tube with an inner diameter of 0 and a length of B, the constant is 128 * L / (ff * D4)). However, a better method is to determine this constant from the measured value by the same method when the device is started up, just like the method of measuring the device constant by using a capillary principle viscosity timer. Because many polymers have a shear-related viscosity, an additional constant Kscher is sometimes required on known flow theorems. 7 = ~ * Taro, like, ...) V t Shiguo has only one unbranched section behind the delivery pump (or in front of the booster pump). In this simple case, the difference between ,level
句/皿度進行測量的方法並根據已知的栗的轉速的體積流量 很準確地計算出黏稠度。 在已知主體積流量分爲兩個或多個相互平行的支流的情 況下:如平行佈置了多個連續錢鑄機，則可以在確定各 4 ^丁路線的參數K7*Kseher後，經由在連續式料機前 測量多個點的壓力和溫度的方法計算出黏稠度。從各個平 :路段的壓力損耗關係和已知主流中各個支流的和中可以 ;確疋黏稠度和支流關係的方程式。比方說，準確 :知道各支流的情況，就可以更好地控制各支 度，以產生更均勾的顆粒。 100739.doc 200538492 令人驚奇的是本發明無須對反應器進行質量監控，對壓 力和溫度的測量可以直接在終端反應器的出料口或直接在 輸送泵的後面進行。即使是在熔體主流運動後期所發現的 _ 變動，在考慮了延長的調整滯後時間的情況下仍然適用於 對終端反應器進行監控。 在當今通常的聚酯生産設備裏多使用日生産能力爲1〇〇〇 噸聚酯的終端反應器。這通常就有可能用一台終端反應器 鲁裏生産出來的聚合物爲若干台成型設備供料。這樣做需要 使用專用的熔體分配器。可以使用一台熔體分配器將熔體 分爲若干個支流，也可以經由分支佈置的若干台分配器來 進行分流。在EP 615 006專利中就有對各類適用的熔體分配 器的說明 在大型設備上，如在使用直接紡紗設備的紡紗廠就省略 了先將從終知反應器出來的溶體進行造粒的中間工序，直 接對聚S曰纖維或聚酯長絲進行紡紗。由於在溶體的輸送過 • 程中會産生較大的壓力損耗，爲了更好地爲所有的成型設 備進行供料，至少要裝一個增壓果。這個增壓系可以裝在 熔體分配器的前面，也可以裝在分配器的後面。比較受歡 迎的實例是測量在聚合物篩檢程式和增壓泵之間的壓力 差。 爲了能使溶體均勻地被分配到各個成型設備，目前通用 的方法是在熔體分配器的後面加裝儲壓器，並用它來對熔 體分配器的測量值進行控制。而在製造聚酯顆粒的設備 上，通常是將儲壓器裝在造粒裝置裏。熔體的溫度是在熔 100739.doc -10- 200538492 體分配器之前進行測量的。另外，也可以在熔體分配器的 後面對溫度進行測量。 本發明另一個比較受歡迎的實例是使用所有儲壓器的測 .量值’熔體溫度以及輸送泵的已知體積流量來求出ι·ν·。如 果求出的這個Ι·ν·與給定值有偏差，則需要對終端反應器進 行再次調整。 下面對本發明的兩種實施方法進行詳細的說明，然而這 鲁並不意味著本發明僅局限於這兩種用途： 第1圖所示爲一種製造聚酯顆粒的設備，這種聚酯顆粒可 以用於製造飲料瓶。終端反應器（1)裏的熔體經由一個輸送 泵（2)被輸送出來，經過一個聚合物熔體篩檢程式（3)，在裝 入到管道裏的溫度感測器Ti上測量溫度後，被引導到一台 溶體分配器（4)上。熔體的流量可以從已知的輸送泵（2)每一 轉的輸送容積和其轉速計算出來。在熔體分配器4裏將熔體 -、、、〇右干σ (圖上爲四台）造粒機（5)。提供壓力測量值 • ρ1—ρ4的儲壓器？1、Ρ2、ρ3*Ρ4直接坐落在被成爲連續洗 鑄機組成部分的造粒機裏，並對經由澆鑄口壓力和大氣壓 力之間的壓力差進行測量。所測出的測量值被送到過程引 導系統，並在那裏計算出現實的I.V.值。根據這些測量值與 預先確定的I.V·給定值的之間的偏差，對反應器控制裝置進The method of measuring the sentence / plate degree and calculating the viscosity from the volume flow rate of the known chestnut speed is very accurate. When the main volume flow is known to be divided into two or more parallel tributaries: If a plurality of continuous casting machines are arranged in parallel, after determining the parameters K7 * Kseher of each 4 ^ D route, The viscosity is calculated by measuring the pressure and temperature at multiple points in front of the continuous feeder. From the relationship between the pressure loss of each road section and the sum of the known tributaries in the mainstream, the equation of the viscosity and the tributary relationship can be determined. For example, accurate: knowing the conditions of each tributary, you can better control each branch to produce more uniform particles. 100739.doc 200538492 Surprisingly, the present invention does not require quality monitoring of the reactor, and the measurement of pressure and temperature can be performed directly at the discharge port of the terminal reactor or directly behind the transfer pump. Even the _ changes found in the later stages of the mainstream movement of the melt are still suitable for monitoring end reactors, taking into account the extended adjustment lag time. In today's common polyester production facilities, terminal reactors with a daily production capacity of 1,000 tons of polyester are mostly used. This usually makes it possible to feed several molding machines with polymers produced in one terminal reactor Lurie. This requires the use of a dedicated melt distributor. A melt distributor can be used to divide the melt into several substreams, or it can be split via several distributors arranged in branches. In the EP 615 006 patent there are descriptions of various applicable melt distributors. In large-scale equipment, such as spinning mills using direct spinning equipment, the solution from the known reactor is first omitted. The intermediate step of pelletizing is to directly spin the poly- or polyester filaments. Due to the large pressure loss during the transport of the solution, at least one pressurized fruit must be installed in order to better feed all the molding equipment. This booster can be installed in front of the melt distributor or behind the distributor. A more popular example is the measurement of the pressure difference between a polymer screening program and a booster pump. In order to enable the solution to be evenly distributed to each molding equipment, the current common method is to install a pressure accumulator behind the melt distributor and use it to control the measured value of the melt distributor. In the equipment for manufacturing polyester granules, the accumulator is usually installed in a granulating device. The melt temperature is measured before melting the 100739.doc -10- 200538492 volume distributor. Alternatively, the temperature can also be measured behind the melt distributor. Another more popular example of the present invention is to use the measured values of all pressure accumulators to measure the melt temperature and the known volume flow of the transfer pump to determine ι · ν ·. If the obtained Ι · ν · deviates from the given value, the terminal reactor needs to be adjusted again. The two implementation methods of the present invention are described in detail below, but this does not mean that the present invention is limited to these two uses: Figure 1 shows a device for manufacturing polyester particles. Used in the manufacture of beverage bottles. The melt in the terminal reactor (1) is conveyed through a transfer pump (2), passes through a polymer melt screening program (3), and the temperature is measured on the temperature sensor Ti installed in the pipeline. Is directed to a solution dispenser (4). The melt flow rate can be calculated from the conveying volume of each revolution of the known conveying pump (2) and its speed. In the melt distributor 4, the pellets (5) are formed by melt-,-, and o-dry σ (four in the figure). Provide pressure measurement • Pressure accumulator for ρ1—ρ4? 1. P2, ρ3 * P4 are located directly in the granulator, which is part of the continuous washing machine, and measure the pressure difference between the pressure through the gate and the atmospheric pressure. The measured value is sent to the process guidance system, where the actual I.V. value is calculated. Based on the deviation between these measured values and the predetermined I.V. given value, the reactor control device is developed.
一員例如，經由真空控制的過程電腦來改變反應写室 内的壓力。 〜W 第2圖所不爲一直接紡紗廠。除了聚合物熔體篩檢程式（3) 外’匕和圖1所示設傷完全相同。纟篩檢程式（3)的後面是熔 100739.doc 200538492 體主抓中的>1力差測i ’它直接位於兩個提供測量值的 儲磨器Ρ5和Ρ6之間。在兩個儲璧器之間使用溫度感測器扣 來測量熔體主流的溫度。在這個壓力測量段之後有一台增 I泵⑹’它用來産生必要的|力’以便將熔體主流經由溶 體分配器(4)送到各㈣紗梁⑺。所測出的測量值被送到過 程引導系統，並在那襄計算出現實的Ι ν.值。根據這些測量 值與預先確定的I.V.給定值的之間的偏差，同樣可以對反應 益控制裝置進行干預。 在許多紡紗廠裏，在熔體分配器之前都有將聚合物熔體 導出的裝置。在本例裏這個導出裝置系指造粒機⑻，它將 紡紗廠暫時不用的聚合物製成聚合物顆粒，在以後需要時 可以將這些顆粒重新熔化，這些顆粒可以用於製造纖維、 薄膜或甑子。纟這個具體實㈣導出|置在聚合物筛檢程 式（3)和帛-個儲壓器Ρ5之間。彡個溶體流是經由果輸送流 的差值計算出來的，在使用儲壓器！>7和溫度感測器Ti2的情 況下，同樣可以用過程引導電腦計算出它的黏稠度。 【實施方式】 本發明之方法得以連續性或非連續性的方式來進行，其 經由下列兩實施例來進行說明之。 第1實施例 在一套只有一台造粒機的生産聚丁烯的設備裏，既可以 選擇使用黏度計的方法，也可以經由從輸送泵的轉速，輸 达官道裏的溫度和澆鑄頭之前的壓力裏來計算黏稠度。圖3 所顯示的就是在經過正比例常數Kscher的匹配後兩個數值 100739.doc -12- 200538492 在若干天内一直能良好地保持一致的情況。 第二實施例 足疋一台用於製造聚乙稀（ΡΕτ)瓶子的設備，它有兩台造 粒機本例中的黏稠度是從輸送泵的轉速，輸送管道裏的 溫度和涛鑄頭之前的壓力裏計算出來的。並用計算出來的 結果對黏稠度調節裝置進行控制。事實證明它的調整質量 可以和通常的黏度計控制一樣好。本例中的目標黏稠度爲 0’625 dl/g，可以保持在〇,〇〇15 dug的標準公差範圍内（絕對 值爲 ±0,004 dllg)。 【圖示簡單說明】 第1圖係為爲一種製造聚酯顆粒的設備； 第2圖係為一直接紡紗廠；以及 第3圖係為一顯示的就是在經過正比例常數以心以的匹 配後兩個數值在若干天内一直能良好地保持一致的情況。 【主要元件符號說明】 1 終端反應器 2 輸送泵 3 聚合物熔體篩檢程式 4 熔體分配器 5 造粒機 6 增壓泵 7 紡紗梁 8 造粒機 100739.doc ^One member, for example, changes the pressure in the reaction writing chamber via a vacuum-controlled process computer. ~ W Figure 2 is not a direct spinning mill. Except for the polymer melt screening program (3), the injury is the same as that shown in Figure 1.纟 The screening program (3) is followed by fused 100739.doc 200538492 in the main grip of the body > 1 force differential measurement i ′, which is directly located between the two mills P5 and P6 that provide measured values. Use a temperature sensor clasp between the two reservoirs to measure the temperature of the melt main stream. After this pressure measurement section, an I pump ⑹ 'is used to generate the necessary force to send the melt main stream to each reed beam ㈣ through the solution distributor (4). The measured value is sent to the process guidance system, and the actual I ν. Value is calculated there. Depending on the deviations between these measured values and predetermined I.V. setpoints, it is also possible to intervene in the reaction benefit control device. In many spinning mills, there is a device for discharging the polymer melt before the melt distributor. In this example, the lead-out device refers to a granulator. It will make polymers that are not used by the spinning mill into polymer particles. These particles can be remelted when needed later. These particles can be used to make fibers and films. Or bitch. This concrete implementation is derived between the polymer sieve inspection formula (3) and a pressure accumulator P5. A solution flow is calculated from the difference between the fruit transport flow and the accumulator is being used! > 7 and the temperature sensor Ti2, the process can also be used to guide the computer to calculate its viscosity. [Embodiment] The method of the present invention can be performed in a continuous or discontinuous manner, and it will be described through the following two examples. In the first embodiment, in a set of equipment for producing polybutene with only one granulator, the method of using a viscometer can be selected, or the speed of the conveying pump can be used to convey the temperature in the official channel and the casting head. To calculate the viscosity. Figure 3 shows the two values 100739.doc -12- 200538492 that have been consistently consistent for several days after matching with the proportional constant Kscher. The second embodiment is sufficient for a piece of equipment for manufacturing polyethylene (PET) bottles. It has two granulators. The viscosity in this example is from the speed of the conveying pump, the temperature in the conveying pipe and the casting head. Calculated from the previous pressure. The calculated results are used to control the viscosity adjustment device. It turns out that its adjustment quality can be as good as the usual viscometer control. The target viscosity in this example is 0’625 dl / g, which can be kept within the standard tolerance range of 〇〇01515 dug (the absolute value is ± 0,004 dllg). [Simplified illustration of the figure] Figure 1 is a kind of equipment for manufacturing polyester pellets; Figure 2 is a direct spinning mill; and Figure 3 is a display showing the matching after the proportional constant The last two values have been in good agreement for several days. [Symbol description of main components] 1 Terminal reactor 2 Delivery pump 3 Polymer melt screening program 4 Melt distributor 5 Granulator 6 Booster pump 7 Spinning beam 8 Granulator 100739.doc ^