使用回火爐來對玻璃板進行回火是大家熟知的，其中將玻璃板加熱至回火溫度。在所謂的輥式輸送機上，玻璃板在回火爐中的旋轉的陶瓷輥上沿著一個方向上移動或來回移動。配備有輥式輸送機的爐通常被稱為輥式爐，其典型溫度為，例如，680℃至700℃。玻璃板從回火爐順次並排移動，或以一塊或多塊玻璃板的各種玻璃負載的形式沿著輥式輸送機傳送到爐子後部的回火冷卻單元，在此用噴氣嘴進行回火冷卻。通常，用於冷卻的空氣溫度與工廠外部或內部的空氣溫度大致相同。吹風機或壓縮機用於供給冷卻空氣。
代替輥式爐，可以使用基於噴氣懸浮的爐，通常也稱為噴氣懸浮爐。在基於噴氣懸浮技術的解決方案中，玻璃板在薄氣墊的支撐下漂浮並且僅在其橫向邊緣的一處接觸輸送線輥或其他輸送構件。可以將等效的空氣懸浮技術應用於安裝在爐子背面的回火冷卻裝置中的回火冷卻單元。
進入回火過程的玻璃板是所謂的浮法玻璃，其中內部應力分佈足夠低，以允許乾淨地切割玻璃板。浮法玻璃表面的壓縮應力通常為1MPa到4MPa。當玻璃板彎曲時，在其凸表面上會形成張應力，浮法玻璃表面平均承受的張應力高達30MPa。在相對較小的負載下可以達到此極限，並且如果超過此極限，浮法玻璃很容易破裂。回火過程的目的是充分提高浮法玻璃板的強度。
除了強度之外，回火玻璃板的另一個期望的性能是玻璃板破裂時的安全性。未回火的玻璃板會破碎成有割傷風險的大塊。回火玻璃板卻會破碎成幾乎無害的顆粒。
回火過程中在玻璃板表面形成的壓應力(硬化，即回火水準)取決於玻璃在通過玻璃板的特性溫度區域(大約600℃→500℃)冷卻時的厚度方向溫度曲線。通常，在回火冷卻中玻璃表面和中心厚度之間的溫差約為100℃。玻璃板越薄，達到相同的上述溫度差所需的冷卻能力越強。在回火期間，常形成呈抛物線形狀的穩定厚度方向應力分佈。玻璃板內部的穩定應力稱為殘餘應力。圖1顯示了應力分佈與玻璃厚度的關係。從圖1可以看出，在玻璃表面，即在抛物線的端部，應力是壓應力。在玻璃的中心，即在抛物線的中心，應力為拉應力。例如，對於厚度為4mm的玻璃板，在回火期間達到至少約100MPa的表面壓縮力，由此在玻璃中心的拉應力為約46MPa。中心層拉應力在此範圍內的4mm玻璃板在碎裂時會破碎成符合安全玻璃標準要求的顆粒。回火玻璃的破碎模式，即玻璃區域中顆粒的密度，特別取決於玻璃內的拉應力的大小。回火玻璃顆粒的密度與殘餘應力(即回火水準)之間的關係已在文獻中被廣泛研究。在出版物“Akeyoshi,K.和Kanai,E.，回火玻璃的機械性能，1965年第七屆國際玻璃會議論文集，論文編號為80”的圖中，給出了玻璃面積為50×50mm的顆粒數與所測量的玻璃回火水準的關係。從出版物中的圖可以得出結論，隨著玻璃板的變薄，獲得相同顆粒數所需的殘餘應力會增加。在出版物“Lee, H., Cho, S., Yoon, K. andLee, J.，應力玻璃中的玻璃厚度和碎裂行為,《New Journal of Glass and Ceramics》，第116-212頁”中，已經發現，當拉伸層的起始深度大於玻璃厚度的20%時，顆粒的密度急劇降低。因此，應力分佈的形狀也影響玻璃的碎裂。
當回火玻璃板彎曲時，僅當消除在回火過程中形成的玻璃板表面上的壓應力時，拉應力才會開始在其凸表面上形成。例如，上述的厚度為4毫米的回火玻璃板比未回火玻璃板承受大約((100MPa+30MPa)/30MPa=)4.3倍的彎曲。通常認為，回火玻璃的強度比浮法玻璃板強3至4倍。
已經建立了許多標準來定義可以用作安全玻璃的回火玻璃。根據標準EN12150-1，可以用作安全玻璃的回火玻璃在用尖銳的鋼制工具破碎時必須破碎成使得每個區域尺寸為50×50mm的玻璃，不包括破裂的部位和邊緣條，必須具有下表中的顆粒計數。在顆粒計數中，將整個處於50×50mm區域內的顆粒數計為1，將部分在上述區域內的顆粒數計為一半。
表1：根據標準EN12150-1，在破碎的安全玻璃中允許的最低顆粒數。 玻璃類型 標稱厚度(mm) 在50×50mm範圍內
的最小顆粒數
浮法和拉絲玻璃板 3 15
浮法和拉絲玻璃板 4至12 40
浮法和拉絲玻璃板 15至19 30
壓花型玻璃板 4至10 30
已知的問題是，對於回火玻璃，在不破碎回火玻璃的情況下無法看到玻璃是否會以安全的方式破裂以及玻璃的碎裂是否會滿足安全玻璃標準的要求。打破大型且厚的玻璃產品是很昂貴的，因此通常使用專門用於測試的小型測試玻璃進行測試。在這種情況下，將相同厚度的測試玻璃添加到第一個生產系列的第一個玻璃負載中，然後故意將玻璃破碎，並檢查其碎片是否滿足安全玻璃標準的要求。如果是滿足安全玻璃標準的要求，則玻璃負載中的所有玻璃均被判定是回火安全玻璃。然後使用相同的回火方法(即指示)對隨後的玻璃負載進行回火，假設所述回火方法將把它們回火成安全玻璃。然而，玻璃板和玻璃板中不同部分通常不被加熱到完全相同的回火溫度，因為加熱元件上的負載根據經過加熱元件的玻璃的量變化，以及在較小的負載下的區域內，玻璃板將被加熱更多。火爐中的某些加熱元件也可能損壞，從而減緩了對經過它們的玻璃塊的加熱。此外，為確保均勻加熱玻璃而採取的調節加熱的措施可能會失敗，並增加玻璃的溫度差。通常，未裝滿的玻璃負載加熱速度比裝滿的玻璃負載的加熱速度快，而大型玻璃的加熱速度則比小玻璃的加熱速度慢。因此，確認測試玻璃是安全玻璃並不能確保生產系列中的所有玻璃都是安全玻璃。為了確保這一點，要求測試玻璃的顆粒數明顯高於標準規定的顆粒數。即使採取這種措施，在所生產的玻璃中，也不能排除回火玻璃以危險方式破碎的可能性。即使成功加熱也不能保證。使用過低的吹氣壓力或過長的吹氣距離，例如，由於操作員錯誤，可能會導致生產的玻璃不能成為符合安全玻璃標準的安全玻璃。在典型的品質控制協議中，每次玻璃質量和厚度變化時都必須對新的測試玻璃進行回火，這在行業中的很多生產人員的生產中常見。因此，測試玻璃也帶來了明顯的增加成本。
有一些用於測量回火玻璃殘餘應力的商用設備。商品名稱為GASP的測量設備測量玻璃表面的壓縮應力，商品名稱為SCALP的測量設備測量玻璃的內部殘餘應力曲線。兩種設備均基於由殘餘應力決定的鐳射的不同折射。但是，這兩種設備僅適合在實驗室中進行測量，因為用它們進行測量相對較慢，無法監視生產玻璃中的殘餘應力。在兩種測量設備(GASP和SCALP)中，少量液體會滴在玻璃表面上，在GASP中是特殊油，而在SCALP中是特殊酒精。要使用的測量設備置於液體上方，並進行了調整，然後可以直接讀取應力或者使用輔助表讀取應力。裝置的精度是令人滿意的，即，可以以約5%的誤差範圍獲得絕對應力。但是，使用該設備進行的測量很難自動化，並且該設備的使用不適合要從移動的玻璃中測量應力的測量。在專利申請CN107677402A中，已經公開了一種利用現有技術中已知的設備的方法。
在專利申請US2003/0076487中還討論了測量回火玻璃的應力。其中的目的是基於玻璃中偏振光的行為來測量玻璃的殘餘應力。在這種方法中，不需要觸摸玻璃。
專利申請US2017/0221198A的方法通過自動分析回火玻璃的顆粒圖案來促進回火玻璃的品質控制。在該方法中，對破碎的回火玻璃進行照明和照相，並使用影像處理程式從照片中分析顆粒圖案的相關細節。It is well known to use a tempering furnace to temper a glass sheet, in which the glass sheet is heated to the tempering temperature. On so-called roller conveyors, glass plates move in one direction or back and forth on rotating ceramic rollers in the tempering furnace. A furnace equipped with a roller conveyor is generally called a roller furnace, and its typical temperature is, for example, 680°C to 700°C. The glass plates are moved side by side in sequence from the tempering furnace, or conveyed in the form of various glass loads of one or more glass plates along a roller conveyor to the tempering and cooling unit at the back of the furnace, where the air nozzles are used for tempering and cooling. Generally, the temperature of the air used for cooling is approximately the same as the temperature of the air outside or inside the factory. A blower or compressor is used to supply cooling air. Instead of roller furnaces, furnaces based on air jet levitation can be used, which are also commonly referred to as air jet levitation furnaces. In the solution based on the air jet suspension technology, the glass plate floats under the support of a thin air cushion and only touches the conveying line roller or other conveying member at one of its lateral edges. The equivalent air suspension technology can be applied to the tempering cooling unit installed in the tempering cooling device on the back of the furnace. The glass sheet that enters the tempering process is so-called float glass, in which the internal stress distribution is low enough to allow the glass sheet to be cut cleanly. The compressive stress on the surface of float glass is usually 1 MPa to 4 MPa. When the glass plate is bent, tensile stress is formed on its convex surface, and the average tensile stress on the float glass surface is as high as 30 MPa. This limit can be reached under a relatively small load, and if this limit is exceeded, the float glass is easily broken. The purpose of the tempering process is to fully increase the strength of the float glass sheet. In addition to strength, another desirable property of tempered glass sheets is safety when the glass sheets break. The untempered glass sheet will break into large pieces with a risk of cuts. The tempered glass plate will break into almost harmless particles. The compressive stress (hardening, tempering level) formed on the surface of the glass plate during the tempering process depends on the thickness direction temperature curve of the glass when it is cooled through the characteristic temperature region of the glass plate (approximately 600°C→500°C). Generally, the temperature difference between the glass surface and the center thickness during tempering cooling is about 100°C. The thinner the glass plate, the stronger the cooling capacity required to achieve the same above-mentioned temperature difference. During tempering, a stable thickness direction stress distribution in a parabolic shape is often formed. The stable stress inside the glass plate is called residual stress. Figure 1 shows the relationship between stress distribution and glass thickness. It can be seen from Figure 1 that on the glass surface, that is, at the end of the parabola, the stress is compressive. At the center of the glass, that is, at the center of the parabola, the stress is tensile stress. For example, for a glass sheet with a thickness of 4 mm, a surface compressive force of at least about 100 MPa is reached during tempering, and thus the tensile stress in the center of the glass is about 46 MPa. The 4mm glass plate with the central layer tensile stress within this range will be broken into particles that meet the requirements of safety glass standards when it is broken. The breaking mode of tempered glass, that is, the density of particles in the glass area, depends in particular on the magnitude of the tensile stress in the glass. The relationship between the density of tempered glass particles and the residual stress (ie tempering level) has been extensively studied in the literature. In the figure in the publication "Akeyoshi, K. and Kanai, E., Mechanical Properties of Tempered Glass, Proceedings of the Seventh International Conference on Glass in 1965, Paper No. 80", the glass area is given as 50×50mm The relationship between the number of particles and the measured tempering level of the glass. It can be concluded from the figures in the publication that as the glass sheet becomes thinner, the residual stress required to obtain the same number of particles will increase. In the publication "Lee, H., Cho, S., Yoon, K. and Lee, J., Glass Thickness and Fragmentation Behavior in Stressed Glass, "New Journal of Glass and Ceramics", pages 116-212" It has been found that when the initial depth of the stretched layer is greater than 20% of the thickness of the glass, the density of the particles decreases sharply. Therefore, the shape of the stress distribution also affects the breakage of the glass. When the tempered glass sheet is bent, only when the compressive stress on the surface of the glass sheet formed during the tempering process is eliminated, the tensile stress will begin to form on its convex surface. For example, the above-mentioned tempered glass sheet with a thickness of 4 mm can withstand approximately ((100MPa+30MPa)/30MPa=) 4.3 times the bending of the untempered glass sheet. It is generally believed that the strength of tempered glass is 3 to 4 times stronger than that of float glass. Many standards have been established to define tempered glass that can be used as safety glass. According to the standard EN12150-1, tempered glass that can be used as safety glass must be broken into a glass with a size of 50×50mm in each area when broken with sharp steel tools, excluding broken parts and edge strips, and must have The particle count in the table below. In the particle count, the number of particles in the entire 50×50mm area is counted as 1, and the number of particles partially in the above area is counted as half. Table 1: According to the standard EN12150-1, the minimum number of particles allowed in broken safety glass. Glass type Nominal thickness (mm) The smallest number of particles in the range of 50×50mm
Float and brushed glass panels 3 15
Float and brushed glass panels 4 to 12 40
Float and brushed glass panels 15 to 19 30
Embossed glass plate 4 to 10 30
The known problem is that for tempered glass, it is impossible to see whether the glass will break in a safe manner without breaking the tempered glass and whether the fragmentation of the glass will meet the requirements of the safety glass standard. It is expensive to break large and thick glass products, so small test glass specially used for testing is usually used for testing. In this case, add the test glass of the same thickness to the first glass load of the first production series, and then deliberately break the glass and check whether the fragments meet the requirements of the safety glass standard. If it meets the requirements of the safety glass standard, all glass in the glass load is judged to be tempered safety glass. The subsequent glass loads are then tempered using the same tempering method (i.e. instructions), assuming that the tempering method will temper them into safety glass. However, different parts of the glass plate and the glass plate are usually not heated to exactly the same tempering temperature, because the load on the heating element varies according to the amount of glass passing through the heating element, and in the area under a smaller load, the glass The plate will be heated more. Certain heating elements in the furnace may also be damaged, thereby slowing the heating of the glass blocks passing through them. In addition, measures taken to adjust the heating to ensure uniform heating of the glass may fail and increase the temperature difference of the glass. Generally, the heating speed of the underfilled glass load is faster than the heating speed of the filled glass load, and the heating speed of the large glass is slower than the heating speed of the small glass. Therefore, confirming that the test glass is safety glass does not ensure that all the glasses in the production series are safety glass. In order to ensure this, the number of particles in the test glass is required to be significantly higher than the number specified in the standard. Even if such measures are taken, in the glass produced, the possibility of the tempered glass breaking in a dangerous manner cannot be ruled out. Even if the heating is successful, there is no guarantee. Using too low blowing pressure or too long blowing distance, for example, due to operator error, may cause the produced glass to fail to meet safety glass standards. In a typical quality control agreement, a new test glass must be tempered every time the glass quality and thickness change, which is common in the production of many production personnel in the industry. Therefore, the test glass also brings a significant increase in cost. There are some commercial equipment for measuring the residual stress of tempered glass. The measuring device with the trade name GASP measures the compressive stress of the glass surface, and the measuring device with the trade name SCALP measures the internal residual stress curve of the glass. Both devices are based on the different refraction of the laser which is determined by the residual stress. However, these two devices are only suitable for measurement in the laboratory, because they are relatively slow to perform measurements and cannot monitor the residual stress in the produced glass. In the two measuring devices (GASP and SCALP), a small amount of liquid will drip on the glass surface, in GASP it is a special oil, and in SCALP it is a special alcohol. The measuring device to be used is placed above the liquid and adjusted, and then the stress can be read directly or using an auxiliary table to read the stress. The accuracy of the device is satisfactory, that is, the absolute stress can be obtained with an error range of about 5%. However, the measurement performed with this device is difficult to automate, and the use of this device is not suitable for the measurement to measure the stress from the moving glass. In the patent application CN107677402A, a method using equipment known in the prior art has been disclosed. The measurement of the stress of tempered glass is also discussed in the patent application US2003/0076487. The purpose is to measure the residual stress of the glass based on the behavior of polarized light in the glass. In this method, there is no need to touch the glass. The method of patent application US2017/0221198A promotes the quality control of tempered glass by automatically analyzing the grain pattern of tempered glass. In this method, the broken tempered glass is illuminated and photographed, and an image processing program is used to analyze the relevant details of the grain pattern from the photograph.
圖3示出了根據本發明實施例的回火設備。回火設備20包括根據圖3在玻璃板的行進方向上連續排列的回火爐1和回火冷卻單元2。圖3所示的回火爐1配備有水平旋轉輥。旋轉輥形成輸送線，即所謂的輥線。可替代地，可以使用具有懸浮噴氣嘴的工作台(圖3中未示出)。回火設備還配備有安裝在回火爐和回火冷卻裝置之間的輸送線上方或下方的溫度掃描器5和高溫計8。溫度掃描器5安裝到回火設備上使得溫度掃描器5可以在沒有鏡子的情況下通過其透鏡看到在傳送線上移動的玻璃板。根據基於玻璃的特性選擇的回火指示、回火方法或類似的預定指示(其中，例如，定義了回火爐的溫度和在回火爐中待回火的玻璃板的加熱時間)，在回火爐中以預定速度沿相同方向或來回移動待加熱的玻璃板G。回火爐在輸送線上的傳送速度是可調的。加熱後的玻璃板以傳送速度W從回火爐1移動至回火冷卻單元2，該傳送速度W通常高於回火爐1內的玻璃板的傳送速度。通常，從回火爐到回火冷卻單元的傳送速度W在200mm/s至800mm/s之間的範圍內。
圖3所示的回火冷卻單元2配備有水平旋轉輥3。這些旋轉輥形成玻璃板輸送線，該玻璃板輸送線可以是與回火爐中的輸送線類似的輸送線。輸送機通過回火冷卻單元提供。回火冷卻單元2還配備有在輸送線的上方和下方設置的空氣冷卻箱4。空氣冷卻箱4配備有吹氣孔7，冷卻空氣作為空氣射流從吹氣孔7朝玻璃板G釋放。吹氣孔7通常是圓形的孔。在空氣冷卻箱4中，吹氣孔7通常以連續的排的形式安置在空氣冷卻箱的方向上。吹氣孔7也可以具有其他形狀，例如狹縫。通過改變吹氣壓力Δp和/或吹氣距離H來調節從空氣冷卻箱釋放的冷卻空氣射流的冷卻能力。吹氣壓力，即空氣冷卻箱內部壓力與環境壓力之間的壓力差，是由空氣冷卻箱4內部的壓力感測器6測量。
當回火爐1是噴氣懸浮爐時，旋轉輥3或噴氣懸浮台及其傳送部件在回火冷卻單元2中通常相對於垂直於玻璃板G的運動方向的水平方向處於稍微傾斜的位置。
回火過程中，玻璃(例如，厚度為3毫米)的每個部分必須進行回火冷卻至少約3秒鐘。例如，在600mm/s的回火冷卻傳送速度下，這將需要長度至少為1800mm的直通型回火冷卻單元。在直通式回火冷卻裝置中，玻璃板僅在一個方向上以傳送速度W移動通過強化冷卻裝置。所謂的擺動回火冷卻裝置通常比允許的最長玻璃負載長度長約1m。因此，玻璃負載整體以傳送速度W移動到回火冷卻單元中。當玻璃負載的前端到達回火冷卻單元的另一端時，玻璃負載會回頭。此後，玻璃負載在回火冷卻單元2中來回移動，直到完成回火冷卻，通常還完成了最終的冷卻。在回火冷卻中，將玻璃板從回火溫度冷卻至約450℃的溫度，並且在最終冷卻中，從該溫度冷卻至約50℃的溫度。最終冷卻不會影響玻璃的殘餘應力。
圖4示出了根據本發明的另一實施例的回火設備。在該實施例中，溫度掃描器5已經安裝在回火冷卻單元的底部，使得溫度掃描器可以通過鏡子9看到在傳送線上移動的玻璃板G，該鏡子9以一定角度安裝在溫度掃描器鏡頭的前面，並與鏡頭保持一定距離。圖3和4所示的底部溫度掃描器的問題在於其光學表面的污染。佈滿灰塵的鏡頭或鏡子可能會干擾測量和/或導致讀數不準確。然而，為了防止光學表面的污染，溫度掃描器5可以配備有自動的透鏡或鏡子清潔設備，該設備吹和/或擦拭乾淨其表面上的污垢和灰塵。測量玻璃板頂表面溫度的高溫計8的資料用於連續校準溫度掃描器5。每當未塗覆的玻璃G進入其測量半徑時，它就玻璃板的實際溫度給出準確的資訊。在上述校準中，將該讀數與來自玻璃板上相應測量位置的底部溫度掃描器5的讀數進行比較，以及底部溫度掃描器的設置，例如反射鏡9的反射係數，根據該比較而改變。
圖5示出了根據本發明實施例的回火設備的框圖。在圖5中，示出了該方法中回火設備線的組裝零件和主要資料流程的組成部分。圖5示出了回火爐1，在其前面是裝載台13，在該裝載台13上放置有待回火的玻璃負載，即玻璃板或多個玻璃板，以及回火冷卻單元2，在它後面的是一個卸貨台14，從卸貨台將回火玻璃板舉到玻璃架上。回火線控制單元11通常位於鄰近裝載台的起點的位置。在圖5中，警報裝置12位於卸貨台14的末端附近。警報裝置12可以是諸如警告燈、蜂鳴器或顯示終端。為了清楚地描述資料流程，在圖5中示出了計算單元10的位置。通常，計算單元10位於與控制單元11相同的空間中，或者甚至位於玻璃廠外部設置的所謂的雲伺服器上。
在計算單元10中，針對玻璃負載的至少一個玻璃板，優選地針對每個玻璃板，計算回火水準。回火水準包括玻璃板表面的壓應力、內部最高的拉應力或整個厚度方向的殘餘應力分佈。玻璃板的熱性能和機械性能是用於計算的初始資料，該計算被程式設計到計算單元所使用的計算程式中。因此，在對玻璃板進行回火之前已經知道了這些性能。計算中使用的玻璃或玻璃負載的特定主要特性，用於計算的初始資料，是玻璃板厚度L、回火冷卻單元中的時間ttc
，玻璃板的回火溫度場Ti, y-z
以及通過玻璃板表面的冷卻空氣射流獲得的傳熱係數h。傳熱係數h尤其取決於空氣射流的吹氣壓力Δp。玻璃板的厚度L通常在3mm至19mm的範圍內，並且幾乎總是在2mm至25mm的範圍內。
圖6示出了根據本發明實施例的流程圖。在圖6的第一步中，識別要回火的玻璃板的屬性，其中最重要的是玻璃板厚度L。玻璃板厚度L由使用者輸入到控制單元11中，例如，由操作員17，並基於識別出的玻璃板特性，確定回火指令，即加熱參數配置有關值的指令，例如回火爐的溫度和玻璃板加熱時間，通過這些指令在回火爐中對玻璃板進行回火，以及冷卻參數配置有關值的指令，例如在回火冷卻單元中的吹氣壓力和吹氣距離，通過這些指令玻璃板將會被冷卻，以及玻璃板從回火爐到回火冷卻單元的輸送速度S和/或回火冷卻時間ttc
是多長時間。可以使用例如鍵盤18或無線應用程式進行資料登錄，然後通過它們將資料傳送到回火設備。可替代地，可以自動識別關於玻璃板特性的資訊，例如，可以從玻璃板厚度自動測量裝置讀取關於厚度的資訊，從該玻璃板厚度自動測量裝置將讀取的資訊傳送到控制單元11。在選擇玻璃板的回火指令時，即加熱和回火的冷卻參數配置，操作員還將有關玻璃板類型的資訊輸入控制單元。玻璃板可以是普通透明玻璃類型的玻璃，也可以是另一種類型的玻璃：未塗覆的玻璃、塗覆的選擇性玻璃或塗覆有另一種塗層的玻璃。玻璃板的類型也是一種屬性，其特別影響加熱參數配置，但通常也影響冷卻參數配置。此外，基於玻璃板的類型，計算單元10將在步驟P1中從其材料特性庫中選擇正確的玻璃板熱特性和在步驟P2中選擇正確的機械材料特性。關於玻璃板，特別是玻璃板的每個部分，在回火冷卻單元中的停留時間ttc
的資訊可以從控制系統11傳送到計算單元10。在其最簡單的形式中，ttc
=S/W，其中，S是回火冷卻單元的長度(常數)，W是傳送速度。在上述的擺動溫度冷卻單元中，停留時間ttc
的計算僅稍微複雜些。停留時間也可以通過諸如光電管來測量。
上面列出了一些與爐子加熱參數配置有關的控制變數。該方法基於測量出的回火爐加熱結果，即玻璃板的溫度，確定玻璃板的回火水準。回火溫度是在加熱之後，玻璃板進入回火冷卻之前測量的。如果要增加對隨後的玻璃負載的加熱，例如基於計算出的回火水準，那麼，例如，增加爐子的加熱時間。與增加加熱時間相比，提高回火爐的溫度水準是稍微慢一些的控制手段。回火爐的溫度也可以本地調節，因為回火爐配備了可單獨調節的矩陣狀加熱電阻器場。在配備有對流加熱元件的回火爐中，也可以通過改變熱風的吹氣壓力或吹氣時間來調節加熱。玻璃板的尺寸(長度和寬度)是通常會影響加熱參數配置的特性。玻璃板的尺寸通常幾乎不會影響冷卻參數配置。
通過玻璃板表面的冷卻空氣射流獲得的傳熱係數h是計算回火水準時必不可少的資訊。確定它的最佳方法是通過測量。傳熱係數可以例如通過在回火冷卻單元中冷卻加熱的厚銅板來測量。因此，銅內部的溫度感測器將產生一條冷卻曲線，根據該冷卻曲線可以相對精確地確定傳熱係數。有關此類測量的更多資訊，請參見出版物Rantala, M 2015, Heat Transfer Phenomena in Float Glass Heat Treatment Processes. Tampere University of Technology. Publication, Volume 1355, Tampere University of Technology.(Rantala，M2015，浮法玻璃熱處理工藝中的傳熱現象。坦佩雷工業大學。坦佩雷工業大學出版物1355卷。)。在不同的吹氣壓力和吹氣距離下對銅進行冷卻的一系列測量提供了足夠的資訊可用於與吹氣壓力和吹氣距離有關的傳熱係數的製錶。因此，計算單元將從表格中(直接或通過從最接近的列表值進行插值)確定玻璃頂面和底面的傳熱係數h(值可以不同，上表面hu
和下表面hl
)，基於由壓力感測器6(在玻璃上方和下方的噴嘴外殼中)測得的吹氣壓力Δp和設置的吹氣距離H(由操作員使用控制單元11選擇)。回火冷卻單元還可以包括用於測量吹氣距離H的測量裝置，由此計算單元10從吹氣距離測量裝置15接收吹氣距離H。通常，當在部署時校準了控制吹氣距離的設備時，吹氣距離H的設定值與實際吹氣距離緊密對應。通常，吹氣壓力為10Pa至20000Pa，具體取決於玻璃厚度。上述壓力範圍的下限通常用於厚度為19mm的玻璃板，壓力範圍的上限壓力用於厚度為3mm的玻璃板。通常，吹氣距離，即，氣孔和玻璃板之間的最短距離為10mm至40mm，取決於玻璃板的厚度。傳熱係數h也可以基於文獻中發現的基於測量的相關方程來確定。方程中的變數是吹氣壓力和吹氣距離。此外，還需要有關系統尺寸的資訊(氣孔直徑、氣孔之間的距離等)。然而，可以使用與本發明的回火冷卻單元很不同的系統來定義相關方程。因此，相關方程的精度可能不足。代替上述測量方法，可以使用流體動力學的數值模型(CFD)來對傳熱係數製表(與Δp和H有關)，只要通過至少一些上述測量結果確認建模的準確性即可。通常，h=40W/(m2
K)的傳熱係數足以對厚度為19mm的玻璃板進行回火，而h=650W/(m2
K)的傳熱係數足以對厚度為3mm的玻璃板進行回火，但這是否足夠還取決於玻璃板的回火溫度。但是，通常，傳熱係數在40W/(m2
K)至650W/(m2
K)之間的範圍內。通常，當要回火的玻璃的厚度改變時，調節吹氣壓力Δp和吹氣距離H以改變傳熱係數h。通常僅改變吹氣壓力，以及在某些情況下僅改變吹氣距離就足夠了。當要增加傳熱係數時，增加吹氣壓力。當要增加傳熱係數時，縮短吹氣距離。
在回火設備部署期間仔細部署吹風機並確定吹風機曲線，從而在計算回火水準時減少了對測量的吹氣壓力Δp的需求。基於吹風機曲線，可以基於吹風機葉輪轉速來確定吹氣壓力，並且由操作員設定的吹氣壓力與實際吹氣壓力緊密對應。因此，計算單元10可以將操作者在控制單元11中設置的值用作吹氣壓力Δp。在一個優選的解決方案中，計算單元10從壓力感測器6接收吹氣壓力資訊。
在根據圖6的優選方案中，從玻璃板G的兩側測量吹氣壓力，因為例如當吹氣來自不同的吹風機時，頂部和底部吹風的吹氣壓力可能發生顯著變化。在這種情況下，壓力測量裝置在頂部和底部測量吹氣壓力(Δpu
，Δpl
)。此外，吹氣距離可以顯著變化，因此，在優選的方案中，它們也從玻璃板的兩側測量。因此，距離測量裝置測量頂部和底部的吹氣距離(Hu
，H1
)。
在圖6中，在計算單元中確定傳熱係數的方法(上文描述了該方法的不同替代方法)，根據吹氣壓力Δpu
和吹氣距離Hu
確定頂部的傳熱係數hu
，並且基於吹氣壓力Δpl
和吹氣距離H1
確定底部的傳熱係數hi
。
計算單元還將需要有關回火空氣溫度Tair
的資訊。可以使用典型的25℃的工廠車間溫度，尤其是在回火吹風機的吸氣入口位於工廠內部的情況下。回火空氣溫度也可能根據季節和天氣而顯著變化，從而使用測得的溫度可以提高計算的準確性。在一個優選的解決方案中，該資訊由控制單元10從吸氣入口處或吹風機管道中的溫度感測器16接收，如圖6所示。
在計算回火水準的方法中，使用在爐子之後和回火冷卻開始之前用溫度掃描器從玻璃板測量的溫度場Ti, y-z
，該資訊從溫度掃描器5傳遞至計算單元10。在一個優選方案中，溫度掃描器從玻璃板的底表面測量溫度。還優選的是，玻璃板的頂表面的溫度用高溫計測量，該溫度計的讀數用於連續校準底部溫度掃描器。頂部的溫度掃描器也可以用於連續校準溫度掃描器，但是使用高溫計的成本較低。
圖2a和2b示出了根據本發明的測量玻璃板中的溫度的方法。如圖2a所示，使用溫度掃描器測量的溫度場由小圖元組成，其尺寸取決於掃描器的屬性。每個圖元都有自己的測量溫度Ti, y-z
。在最簡單的解決方案中，整個玻璃負載中的圖元的平均溫度，即整個玻璃負載的平均溫度 Tm, loading
，被例如計算單元的計算程式選擇，作為玻璃板測量的回火溫度Tmea
。在一個優選的解決方案中，在圖2a中，玻璃板內所有圖元的平均溫度，即整個玻璃板的平均溫度Tm
，被例如計算單元的計算程式選擇，作為玻璃板的測量的回火溫度Tmea
。如圖2b所示，玻璃板也可以分成較小的子部分，其中子部分的參考溫度Tm, Y-Z
是其中圖元的平均溫度。子部分的大小通常約為4×4cm至30×30cm。該子部分的尺寸可以取決於玻璃板的尺寸，並且不必是正方形的。在計算平均值時，可以從圖元溫度中濾除看起來不正確的值。選擇參考溫度Tm, Y-Z
中的最小值Tm, Y-Z, min
作為玻璃板的測定回火溫度Tmea
。以上述方式，針對玻璃負載中的每個玻璃板，選擇代表玻璃板的測量的回火溫度Tmea
(=Tm, loading
或Tm
或Tm, Y-Z, min
)。還可以以其他方式基於溫度掃描器的測量資料選擇回火溫度。溫度選擇必須基於通常由Tmea
表示的溫度掃描器測量資料，這一點至關重要。因此，計算單元的溫度資料分析程式從由溫度掃描器測量的溫度場Ti, y-z
中確定代表玻璃板的測量的回火溫度Tmea
，該回火溫度由計算單元的回火水準計算程式用作玻璃板回火溫度T0
。
溫度掃描器也可以位於玻璃上方，在這種情況下，它可以測量玻璃頂面的表面溫度場。但是，對於頂面發射率低(選擇的，即低輻射玻璃)且未知的頂面上具有塗層的玻璃，無法進行從頂部進行的測量。在優選的解決方案中，用底部的溫度掃描器測量溫度。
在一些連續的回火設備中，玻璃板順序地並且在大致相同的位置移動跨越線寬。在這種情況下，可以使用一個或至少幾個高溫計相當全面地測量玻璃溫度。一台高溫計在玻璃移動的方向沿一條線測量玻璃溫度。在最簡單的解決方案中，溫度掃描器是高溫計或其他以可比較方式測量玻璃溫度的測量設備。在這種情況下，代表玻璃板的測得的回火溫度Tmea
是沿玻璃運動方向的一條或多條線中的玻璃的平均溫度。在優選的解決方案中，用測量玻璃負載中每個玻璃的整個表面積的溫度的溫度掃描器測量溫度。
在計算單元中計算玻璃溫度的程式根據能量方程式(1)確定回火冷卻期間玻璃厚度方向溫度分佈T(x，t)的變化：
其邊界條件為等式(2)和(3)：
其中T=溫度，L=玻璃板的厚度，k=玻璃的導熱率，ρ=玻璃的密度，cp
=玻璃的比熱容。
此外，玻璃板(或玻璃板的子部分)的初始溫度T(x，0)=T0
。例如以上述方式，玻璃板的回火溫度T0
是由溫度掃描器從玻璃板的表面測量並從玻璃板的測量溫度場確定的回火溫度Tmea
。回火溫度可以考慮溫度掃描器5和回火冷卻單元之間的玻璃板的自然冷卻ΔTm
。從溫度掃描器到回火冷卻單元的停留時間通常為0.1s到0.7s，具體取決於輸送線的傳送速度。因此，T0
=Tmea
-ΔTm
，其中Tmea
是基於溫度掃描器測量的玻璃板溫度。ΔTm
在0至10℃之間。在玻璃板的厚度L上進行計算。在玻璃的上表面，厚度方向座標x=0，在下表面x=L。
在傳熱研究中，求解能量方程是一個普遍的問題，文獻中對此提出了幾種解決方案，除了上述參考文獻(Rantala2015)之外，例如還有在以下出版物中：Field, R. E. and Viskanta, R., Measurement and prediction of the dynamic temperature distributions in soda-lime glass plates, Journal of American Ceramic Society, vol. 73, 7, pp. 2047-2053, 1990; Gardon, R., Calculation of temperature distribution in glass plates undergoing heat treatment, Journal of American Ceramic Society, vol. 41, 6, pp. 200-209, 1958; Siedow, N., Lochegnies, D., Grosan, T. and Romero, E., Application of a New Method for Radiative Heat Transfer to Flat Glass Tempering, Journal of American Ceramic Society, vol. 88, 8, pp. 2181-2187, 2005.
等式中要求的玻璃的熱材料特性是密度ρ、比熱容cp
和熱導率k。它們是眾所周知的，並已在計算程式中提供給計算單元。玻璃的這些和輻射特性在圖6中用P1表示。
在等式中，qr, u
是玻璃和環境之間向上(向下qr, l
)的淨輻射熱流。在上面提到的出版物中已經公開了確定它的方法，並且已經公開了確定它的方法所需的玻璃輻射特性。
術語qc, u
和qc, l
是通過回火冷卻單元的空氣射流在玻璃的頂面和底面實現的對流冷卻功率，由等式qc, u
=hu
[Tair
-T(0,t)]和qc, l
=hl
[Tair
-T(L,t)]計算。在等式中，Hu
和hl
是玻璃頂面和底面的對流傳熱係數。上面介紹了確定它們的三種替代技術。重要的是，必須與吹氣壓力和吹氣距離相關地將傳熱係數製錶或以方程式將傳熱係數列出。因此，計算單元能夠基於所測量的吹氣壓力和距離以足夠的精度來確定它們。在等式中，Tair
是回火空氣的溫度。底面的邊界條件中還包括術語qct
，它考慮了從玻璃(在玻璃與滾筒之間的接觸點)傳遞到滾筒的熱量。通常，導熱性差的細繩已經繞在回火冷卻裝置的旋轉輥上。在這種情況下，接觸傳熱的影響可以四捨五入為qct
=0。
利用能量方程式(1)，計算在時間間隔0≤t≤ttc
期間玻璃板的厚度方向溫度分佈T(x，t)的變化，其中ttc
是玻璃板(玻璃中的點)在回火冷卻中的停留時間。通常，基於玻璃板的傳送速度和回火冷卻單元的長度，以上述方式獲得。但是，計算週期ttc
必須足夠長以使玻璃板的整個厚度的溫度下降到溫度極限ttcend
以下。該溫度以下的冷卻不再影響玻璃板上形成的殘餘應力分佈，即回火程度。因此，時間ttc
是玻璃板已經冷卻到溫度極限ttcend
以下的時間。典型的溫度極限ttcend
=450℃。如果在時間間隔0≤t≤ttc
內對流傳熱係數(hu
，hl
)的值發生變化，則將在計算中予以考慮。
作為能量方程式(1)的解，在回火冷卻時間(0-ttc
)上，玻璃板或玻璃板的子部分的厚度方向溫度分佈T(x，t)在時間間隔Δt處獲得。時間間隔Δt在0.001s與2s之間的範圍內,它取決於玻璃厚度。在玻璃厚度L上以間隔Δx的密度獲得溫度分佈。厚度間隔Δx為(0.01-0.2)L。然後，將從能量方程式(1)計算出的溫度分佈T(x，t)用於計算玻璃板的殘餘應力。根據以下公式計算在回火過程中在玻璃板上達到的回火水準：
初始條件為σij
(x，0)=0時，借助於由能量方程式(1)計算出的溫度分佈T(x，t)計算出由冷卻引起的由熱伸長率εth
(x，t)導致的載荷：
在熱伸長率方程中，虛擬溫度Tf
通過溫度分佈T(x，t)和材料參數Mp
來計算：
應力場必須滿足一般平衡條件：
和
求解玻璃的回火應力，即上述等式，是玻璃力學研究中的常見問題，在例如以下出版物中討論過：Aronen, A 2012, Modelling of Deformations and Stresses in Glass Tempering. Tampere University of Technology. Publication, Volume. 1036, Tampere University of Technology; Nielsen, J.H., 2009, Tempered Glass - Bolted Connections and Related Problems, Ph.D. thesis, DTU Civil Engineering, Lyngby, Denmark; Carrè, H., 1996, Numerical Simulation of Soda-Lime Silicate Glass Tempering, Journal de Physique IV, vol. 6, no. 1, pp. 175-185.
應力方程式中的伸長率是流體靜力學伸長率和偏向伸長率。等式中所需的玻璃的機械材料特性是與時間和溫度有關的壓縮模量K和剪切模量G以及熱膨脹係數α1
和αg
。它們例如從上述參考文獻Aronen 2012中眾所周知，並且已經在計算程式中提供給計算單元。玻璃板的機械性能在圖6中用P2表示。
作為計算的結果，獲得玻璃板的厚度方向殘餘應力分佈σ(x)，即，在玻璃厚度上的厚度間隔處的應力σ值。殘餘應力分佈決定了玻璃的回火水準。在對玻璃板進行回火時，強度的提高尤其由玻璃表面的殘餘應力來表示，即，值σ(0)或σ(L)中的較小者。當玻璃板破裂時，回火玻璃板的安全性(例如，按照標準EN12150-1計算的顆粒數)尤其取決於玻璃中心厚度x=0，5L處的拉應力(0，5L)。
基於計算出的殘餘應力分佈，選擇玻璃的回火水準的參考變數。參考變數基於計算的玻璃厚度方向殘餘應力分佈σ(x)。在優選情況下，參考變數是在玻璃的中心厚度處的殘餘應力，即σ(0，5L)。參考變數也可以是玻璃表面的殘餘應力，即σ(0)或σ(L)，或者是平均值的類型，或者是根據殘餘應力分佈σ(x)計算出的一些其他殘餘應力值。殘餘應力參考變數的一般符號為σcal
。玻璃中心厚度處的殘餘應力，即σ(0，5L)，也可以根據標準EN12150-1轉換為玻璃顆粒數，並用作參考變數。例如，可以根據上述參考文獻Akeyoshi 1965中發佈的結果圖像進行更改，其中對於幾種不同的玻璃厚度，已經提出了與應力σ(0，5L)有關的顆粒數。來自結果圖像的資訊可以提供給計算單元，例如通過使用多項式擬合。申請人對應力σ(0，5L)與顆粒數量之間關係的瞭解是基於文獻的基礎，也是申請人收集的大量研究結果。在研究中，對大量各種厚度的玻璃進行了回火，使用上述SCALP測量設備測量了它們的σ(0，5L)值，並根據標準EN12150-1對玻璃進行了顆粒計數測定。參考變數的通用符號為CV，覆蓋了值σcal
和基於其確定的粒子數Ncal
。
參考變數CV的值用於確保將玻璃負載的玻璃成功地回火處理成安全玻璃。操作者基於在警報裝置12的顯示器上顯示的玻璃的參考變數CV或由警報裝置12給出的警報來進行監視。上述選項之間的區別在於，在前者中，操作員能夠基於參考變數CV(例如σ(0，5L))斷定回火玻璃是否為安全玻璃。在後者中，通過計算單元得出結論，從而基於警報裝置給出的警報，發現玻璃板不符合安全玻璃的要求。當在報警設備顯示幕上顯示參考變數CV的值時，可以使用前者。在後面的內容中會更多地討論後者。
基於計算出的殘餘應力分佈，選擇玻璃板回火水準的參考變數CV，然後將其與預定的警報極限值AV進行比較。在參考變數和警報極限值之間的比較滿足預定標準的情況下，計算單元創建警報並將警報消息A發送給警報裝置。預定標準可以是參考變數和警報極限值之間的偏差。例如，參考變數與標準警報極限值的偏差或參考變數與另一預定警報極限值的偏差。例如，如果計算出的參考變數CV的值小於警報極限AV的值，則可以創建警報。在本發明的上下文中，警報極限值也可以被稱為閾值。
警報極限AV已被程式設計到計算單元中。例如，如果參考變數為應力σ(0，5L)，則警報極限值為45MPa。例如，警報極限值為60個粒子，如果參考變數根據EN12150-1的粒子計數為至少40個粒子。通過將參考變數與警報極限值進行比較，可以獲得有關玻璃板是否滿足安全玻璃標準要求的資訊。優選地，顆粒計數的警報極限比標準EN12150-1(和表1)中提出的最小值(例如40)高至少50%，例如60，當玻璃厚度為4mm至12mm時。通常，當玻璃厚度為3.8mm或更大時，警報極限值在50到80個顆粒之間。通常，當玻璃厚度小於3.8mm時，警報極限值在25到50個顆粒之間。優選地，應力σ(0，5L)警報極限值足夠高，基於此應力σ(0，5L)，與通過標準EN12150-1的最低要求相比，玻璃中至少要多形成50%以上的顆粒。通常，當玻璃厚度為3mm至4mm時，警報極限σ(0，5L)值在45MPa至60MPa(拉應力)之間的範圍內。通常，當玻璃厚度為5mm到8mm時，警報極限σ(0，5L)值在35MPa到55MPa之間。通常，當玻璃厚度為10mm-19mm時，警報極限σ(0，5L)值在30MPa到45MPa之間。當參考變數是玻璃表面的殘餘應力，即σ(0)或σ(L)時，警報極限通常為70MPa到110MPa(壓應力)，或者是上述提到的警報極限σ(0，5L)的絕對值的1.8倍。
當警報設備通過光、聲音、顏色或在顯示幕幕上顯示的消息發出警報時，玻璃負載或玻璃負載中至少一塊玻璃板的參考變數低於警報極限。優選的是，顯示幕識別這些玻璃板，使得在玻璃負載的玻璃板中容易找到它們。基於此，回火生產線操作員可以從生產的玻璃中取出引起警報的玻璃板。這些玻璃板的標準安全性測試提供了資訊，通過這些資訊可以完善上述確保回火玻璃板安全性的方法。基於此資訊，可以減小警報極限和標準最小值之間的差異。
通過使用該計算程式，可以在顯示終端上標示玻璃負載或該玻璃負載中的任何玻璃板是否不屬於安全玻璃或是否屬於安全玻璃，例如通過顏色說明。例如，可以通過顯示終端上顯示的玻璃負載圖案上的顏色(例如紅色)來發出警報，指示存在不符合安全玻璃標準的安全玻璃。該計算程式還可以，例如在顯示終端上，使用預選的顏色標示參考變數的級別。例如，綠色表示玻璃板或玻璃負載是安全玻璃。例如，黃色表示玻璃板或玻璃可能是安全玻璃。例如，紅色表示玻璃板或玻璃負載不是安全玻璃。上述使用三種顏色需要設置兩個極限閾值AV，即閾值AV1和AV2。因此，例如將參考變數高於上閾值AV2的玻璃板視為安全玻璃，而參考變數CV低於下閾值AV1的玻璃板視為不合格安全玻璃。參考變數CV在AV1和AV2之間的玻璃板被認為可能是安全玻璃。計算程式還可以用顏色指示參考變數的級別。例如，可以通過將為每種玻璃選擇的參考變數CV放置在玻璃板溫度掃描器圖像或指示玻璃板邊界的其他圖案的頂部，並以預定顏色將其高亮顯示來執行此操作。預定顏色可以是例如紅色、黃色和綠色。例如，如果參考變數值在49MPa到50MPa之間，並且該範圍內的參考變數可以用作安全玻璃，則可以由用戶選擇的顏色(例如綠色)顯示在溫度掃描器圖像中。
該方法可用於確保根據各種玻璃板厚度和玻璃板類型將玻璃負載成功回火成安全玻璃。因此，例如，當計算出的回火水準(CV)接近警報極限(AV)時，並且當滿足預定標準時，回火爐的選定溫度和/或加熱時間，即輸送線速度，被重新調整。分別地，當所計算的回火水準接近警報極限時，當滿足預定標準時，可以重新調整回火冷卻單元中的空氣噴嘴的選擇的吹氣壓力和/或吹氣距離。如果檢測到回火水準與警報極限之間的偏差，則回火設備的回火爐控制系統可以自動增加/減少爐中玻璃板的加熱時間，或建議增加/減少玻璃板加熱時間。此外，如果檢測到回火水準和警報極限之間的偏差，則回火設備的回火冷卻單元控制系統可以自動建議調節玻璃板冷卻噴嘴的吹氣壓力和/或吹氣距離。
對於本領域技術人員顯而易見的是，本發明不限於本文描述的解決方案，而是可以在由申請專利範圍設定的範圍內以多種方式來應用本發明的想法。Figure 3 shows a tempering device according to an embodiment of the invention. The tempering equipment 20 includes a tempering furnace 1 and a tempering cooling unit 2 that are continuously arranged in the traveling direction of the glass sheet according to FIG. 3. The tempering furnace 1 shown in FIG. 3 is equipped with horizontal rotating rollers. The rotating rollers form a conveying line, a so-called roller line. Alternatively, a workbench with suspended air nozzles (not shown in Figure 3) can be used. The tempering equipment is also equipped with a temperature scanner 5 and a pyrometer 8 installed above or below the conveying line between the tempering furnace and the tempering cooling device. The temperature scanner 5 is mounted on the tempering device so that the temperature scanner 5 can see the glass plate moving on the conveying line through its lens without a mirror. Tempering instructions, tempering methods or similar predetermined instructions selected based on the characteristics of the glass (where, for example, the temperature of the tempering furnace and the heating time of the glass plate to be tempered in the tempering furnace are defined), in the tempering furnace The glass plate G to be heated is moved in the same direction or back and forth at a predetermined speed. The conveying speed of the tempering furnace on the conveyor line is adjustable. The heated glass sheet moves from the tempering furnace 1 to the tempering cooling unit 2 at a conveying speed W, which is generally higher than the conveying speed of the glass sheet in the tempering furnace 1. Generally, the transfer speed W from the tempering furnace to the tempering cooling unit is in the range of 200 mm/s to 800 mm/s. The tempering cooling unit 2 shown in FIG. 3 is equipped with a horizontal rotating roller 3. These rotating rollers form a glass sheet conveying line, which may be a conveying line similar to the conveying line in the tempering furnace. The conveyor is provided by the tempering cooling unit. The tempering cooling unit 2 is also equipped with an air cooling box 4 arranged above and below the conveying line. The air cooling box 4 is equipped with blowing holes 7, and cooling air is released from the blowing holes 7 toward the glass plate G as an air jet. The blowing hole 7 is usually a circular hole. In the air cooling box 4, the blowing holes 7 are usually arranged in a continuous row in the direction of the air cooling box. The blowing holes 7 may also have other shapes, such as slits. The cooling capacity of the cooling air jet released from the air cooling box is adjusted by changing the blowing pressure Δp and/or the blowing distance H. The blowing pressure, that is, the pressure difference between the internal pressure of the air cooling box and the ambient pressure, is measured by the pressure sensor 6 inside the air cooling box 4. When the tempering furnace 1 is an air jet levitation furnace, the rotating roller 3 or the air jet levitation table and its conveying parts are usually in a slightly inclined position with respect to the horizontal direction perpendicular to the moving direction of the glass sheet G in the tempering cooling unit 2. During the tempering process, each part of the glass (for example, with a thickness of 3 mm) must be tempered and cooled for at least about 3 seconds. For example, at a tempering cooling transfer speed of 600mm/s, this would require a straight-through tempering cooling unit with a length of at least 1800mm. In the straight-through tempering cooling device, the glass sheet moves through the strengthening cooling device at the conveying speed W in only one direction. The so-called swing tempering cooling device is usually about 1 m longer than the longest allowed glass load length. Therefore, the entire glass load moves to the tempering cooling unit at the conveying speed W. When the front end of the glass load reaches the other end of the tempering cooling unit, the glass load will turn back. Thereafter, the glass load moves back and forth in the tempering cooling unit 2 until the tempering cooling is completed, and the final cooling is usually completed. In the tempering cooling, the glass sheet is cooled from the tempering temperature to a temperature of about 450°C, and in the final cooling, it is cooled from this temperature to a temperature of about 50°C. The final cooling will not affect the residual stress of the glass. Fig. 4 shows a tempering device according to another embodiment of the present invention. In this embodiment, the temperature scanner 5 has been installed at the bottom of the tempering cooling unit, so that the temperature scanner can see the glass plate G moving on the conveying line through the mirror 9, which is installed on the temperature scanner at a certain angle The front of the lens and keep a certain distance from the lens. The problem with the bottom temperature scanner shown in Figures 3 and 4 is the contamination of its optical surface. A dusty lens or mirror may interfere with the measurement and/or cause inaccurate readings. However, in order to prevent contamination of the optical surface, the temperature scanner 5 may be equipped with an automatic lens or mirror cleaning device that blows and/or wipes away dirt and dust on its surface. The data of the pyrometer 8 that measures the temperature of the top surface of the glass plate is used to continuously calibrate the temperature scanner 5. Whenever the uncoated glass G enters its measuring radius, it gives accurate information on the actual temperature of the glass sheet. In the above calibration, the reading is compared with the reading of the bottom temperature scanner 5 from the corresponding measurement position on the glass plate, and the settings of the bottom temperature scanner, such as the reflection coefficient of the mirror 9, are changed according to the comparison. Fig. 5 shows a block diagram of a tempering device according to an embodiment of the present invention. In Figure 5, the assembly parts and main data flow components of the tempering equipment line in this method are shown. Figure 5 shows the tempering furnace 1, in front of which is a loading platform 13, on which is placed a glass load to be tempered, that is, a glass plate or multiple glass plates, and a tempering cooling unit 2, behind it It is an unloading station 14 from which the tempered glass panel is lifted onto the glass shelf. The tempering line control unit 11 is usually located adjacent to the starting point of the loading platform. In FIG. 5, the alarm device 12 is located near the end of the unloading dock 14. The alarm device 12 may be, for example, a warning light, a buzzer, or a display terminal. In order to clearly describe the data flow, the location of the calculation unit 10 is shown in FIG. 5. Generally, the computing unit 10 is located in the same space as the control unit 11, or even on a so-called cloud server installed outside the glass factory. In the calculation unit 10, for at least one glass plate loaded by the glass, preferably for each glass plate, the tempering level is calculated. The tempering level includes the compressive stress on the surface of the glass plate, the highest internal tensile stress or the residual stress distribution in the entire thickness direction. The thermal and mechanical properties of the glass plate are the initial data used for the calculation, and the calculation is programmed into the calculation formula used by the calculation unit. Therefore, these properties are already known before tempering the glass sheet. The specific main characteristics of the glass or glass load used in the calculation. The initial data used for the calculation are the thickness of the glass sheet L, the time t tc in the tempering cooling unit, the tempering temperature field of the glass sheet T i, yz and the passing glass The heat transfer coefficient h obtained by the cooling air jet on the surface of the plate. The heat transfer coefficient h depends in particular on the blowing pressure Δp of the air jet. The thickness L of the glass plate is usually in the range of 3 mm to 19 mm, and is almost always in the range of 2 mm to 25 mm. Fig. 6 shows a flowchart according to an embodiment of the present invention. In the first step of Figure 6, the properties of the glass sheet to be tempered are identified, the most important of which is the thickness L of the glass sheet. The thickness L of the glass sheet is input by the user into the control unit 11, for example, by the operator 17, and based on the identified characteristics of the glass sheet, the tempering instruction is determined, that is, the instruction of the relevant value of the heating parameter configuration, such as the tempering furnace temperature and The heating time of the glass plate, the tempering of the glass plate in the tempering furnace through these instructions, and the instructions for the cooling parameter configuration related values, such as the blowing pressure and the blowing distance in the tempering cooling unit, through these instructions the glass plate will Will be cooled, and how long is the conveying speed S and/or the tempering cooling time t tc of the glass sheet from the tempering furnace to the tempering cooling unit. For example, the keyboard 18 or a wireless application can be used for data registration, and then the data can be transmitted to the tempering device through them. Alternatively, the information about the characteristics of the glass plate can be automatically recognized. For example, the information about the thickness can be read from an automatic glass plate thickness measuring device, and the read information can be transmitted to the control unit 11 from the automatic glass plate thickness measuring device. When selecting the tempering command of the glass plate, that is, the cooling parameter configuration of heating and tempering, the operator also inputs the information about the glass plate type into the control unit. The glass plate can be a common transparent glass type of glass, or another type of glass: uncoated glass, coated selective glass, or glass coated with another coating. The type of glass plate is also an attribute, which particularly affects the heating parameter configuration, but usually also affects the cooling parameter configuration. In addition, based on the type of the glass plate, the calculation unit 10 will select the correct thermal properties of the glass plate from its material property library in step P1 and the correct mechanical material properties in step P2. Regarding the glass sheet, especially each part of the glass sheet, information about the residence time t tc in the tempering cooling unit can be transmitted from the control system 11 to the calculation unit 10. In its simplest form, t tc = S/W, where S is the length (constant) of the tempering cooling unit and W is the conveying speed. In the above-mentioned swing temperature cooling unit, the calculation of the residence time t tc is only slightly more complicated. The residence time can also be measured by, for example, a photocell. Some control variables related to the furnace heating parameter configuration are listed above. This method determines the tempering level of the glass plate based on the measured heating result of the tempering furnace, that is, the temperature of the glass plate. The tempering temperature is measured after heating and before the glass sheet enters the tempering and cooling. If the heating of the subsequent glass load is to be increased, for example based on the calculated tempering level, then, for example, the heating time of the furnace is increased. Compared with increasing the heating time, increasing the temperature level of the tempering furnace is a slightly slower control method. The temperature of the tempering furnace can also be adjusted locally, because the tempering furnace is equipped with an individually adjustable matrix heating resistor field. In tempering furnaces equipped with convection heating elements, the heating can also be adjusted by changing the blowing pressure or blowing time of the hot air. The size (length and width) of the glass plate is a characteristic that usually affects the configuration of heating parameters. The size of the glass plate usually hardly affects the cooling parameter configuration. The heat transfer coefficient h obtained by the cooling air jet on the surface of the glass plate is essential information when calculating the tempering level. The best way to determine it is through measurement. The heat transfer coefficient can be measured, for example, by cooling a heated thick copper plate in a tempering cooling unit. Therefore, the temperature sensor inside the copper will generate a cooling curve, according to which the heat transfer coefficient can be determined relatively accurately. For more information on this type of measurement, see publication Rantala, M 2015, Heat Transfer Phenomena in Float Glass Heat Treatment Processes. Tampere University of Technology. Publication, Volume 1355, Tampere University of Technology. (Rantala, M2015, Float Heat transfer phenomenon in glass heat treatment process. Tampere University of Technology. Tampere University of Technology publication 1355 volume.). A series of measurements for cooling copper under different blowing pressures and blowing distances provide enough information for the tabulation of heat transfer coefficients related to blowing pressure and blowing distance. Therefore, the calculation unit will determine the heat transfer coefficient h of the top and bottom surfaces of the glass (values can be different, the upper surface h u and the lower surface h l ) from the table (directly or by interpolation from the closest list value), based on The blowing pressure Δp measured by the pressure sensor 6 (in the nozzle housing above and below the glass) and the set blowing distance H (selected by the operator using the control unit 11). The tempering cooling unit may further include a measuring device for measuring the blowing distance H, whereby the calculation unit 10 receives the blowing distance H from the blowing distance measuring device 15. Generally, when the device that controls the blowing distance is calibrated during deployment, the set value of the blowing distance H closely corresponds to the actual blowing distance. Generally, the blowing pressure is 10 Pa to 20000 Pa, depending on the thickness of the glass. The lower limit of the above pressure range is usually used for glass plates with a thickness of 19 mm, and the upper limit pressure of the pressure range is used for glass plates with a thickness of 3 mm. Generally, the blowing distance, that is, the shortest distance between the air hole and the glass plate, is 10 mm to 40 mm, depending on the thickness of the glass plate. The heat transfer coefficient h can also be determined based on correlation equations found in the literature based on measurements. The variables in the equation are blowing pressure and blowing distance. In addition, information about the size of the system (air hole diameter, distance between air holes, etc.) is also required. However, a system very different from the tempering cooling unit of the present invention can be used to define the relevant equations. Therefore, the accuracy of the correlation equation may be insufficient. Instead of the above measurement method, a fluid dynamics numerical model (CFD) can be used to tabulate the heat transfer coefficient (related to Δp and H), as long as the accuracy of the modeling is confirmed by at least some of the above measurement results. Generally, the heat transfer coefficient of h=40W/(m 2 K) is sufficient for tempering a glass plate with a thickness of 19mm, and the heat transfer coefficient of h=650W/(m 2 K) is sufficient for a glass plate with a thickness of 3mm. Tempering, but whether this is sufficient also depends on the tempering temperature of the glass sheet. However, in general, the heat transfer coefficient in the range of between 40W / (m 2 K) to 650W / (m 2 K) a. Generally, when the thickness of the glass to be tempered is changed, the blowing pressure Δp and the blowing distance H are adjusted to change the heat transfer coefficient h. It is usually sufficient to change only the blowing pressure, and in some cases only the blowing distance. When increasing the heat transfer coefficient, increase the blowing pressure. When increasing the heat transfer coefficient, shorten the blowing distance. The blower is carefully deployed and the blower curve is determined during the deployment of the tempering equipment, thereby reducing the need for the measured blowing pressure Δp when calculating the tempering level. Based on the blower curve, the blowing pressure can be determined based on the rotation speed of the blower impeller, and the blowing pressure set by the operator closely corresponds to the actual blowing pressure. Therefore, the calculation unit 10 can use the value set by the operator in the control unit 11 as the blowing pressure Δp. In a preferred solution, the calculation unit 10 receives blowing pressure information from the pressure sensor 6. In the preferred solution according to FIG. 6, the blowing pressure is measured from both sides of the glass plate G, because, for example, when the blowing comes from different blowers, the blowing pressure of the top and bottom blowing may change significantly. In this case, the pressure measuring device measures the blowing pressure (Δp u , Δp l ) at the top and bottom. In addition, the blowing distance can vary significantly, therefore, in a preferred solution, they are also measured from both sides of the glass plate. Therefore, the distance measuring device measures the top and bottom blowing distances (H u , H 1 ). In FIG. 6, a method of determining the heat transfer coefficient calculation unit (various alternative methods described above) of the method, determining the heat transfer coefficient h u according to top insufflation pressure and the blowing distance Δp u H u, and The heat transfer coefficient h i of the bottom is determined based on the blowing pressure Δp l and the blowing distance H 1 . The calculation unit will also need information about the tempering air temperature T air . A typical factory floor temperature of 25°C can be used, especially if the suction inlet of the tempering blower is located inside the factory. The tempering air temperature may also vary significantly depending on the season and weather, so using the measured temperature can improve the accuracy of the calculation. In a preferred solution, the information is received by the control unit 10 from the temperature sensor 16 in the suction inlet or the blower duct, as shown in FIG. 6. In the method of calculating the tempering level, the temperature field Ti, yz measured from the glass plate with a temperature scanner after the furnace and before the tempering cooling starts is used, and the information is transmitted from the temperature scanner 5 to the calculation unit 10. In a preferred embodiment, the temperature scanner measures the temperature from the bottom surface of the glass plate. It is also preferred that the temperature of the top surface of the glass plate is measured with a pyrometer, and the reading of the thermometer is used to continuously calibrate the bottom temperature scanner. The temperature scanner on the top can also be used to continuously calibrate the temperature scanner, but the cost of using a pyrometer is lower. Figures 2a and 2b show a method of measuring the temperature in a glass sheet according to the present invention. As shown in Figure 2a, the temperature field measured by a temperature scanner is composed of small picture elements whose size depends on the properties of the scanner. Each picture element has its own measured temperature Ti , yz . In the simplest solution, the average temperature of the picture elements in the entire glass load, that is, the average temperature T m, loading of the entire glass load, is selected by a calculation program such as a calculation unit as the tempering temperature T mea measured by the glass plate . In a preferred solution, in Figure 2a, the average temperature of all picture elements in the glass plate, that is, the average temperature T m of the entire glass plate, is selected by a calculation program such as a calculation unit as the tempering measurement of the glass plate. Temperature T mea . As shown in Figure 2b, the glass plate can also be divided into smaller sub-parts, where the reference temperature T m, YZ of the sub-parts is the average temperature of the picture elements. The size of the sub-part is usually about 4×4cm to 30×30cm. The size of this sub-section may depend on the size of the glass plate, and does not have to be square. When calculating the average value, you can filter out values that seem incorrect from the element temperature. The minimum value T m, YZ, min among the reference temperature T m, YZ is selected as the measured tempering temperature T mea of the glass plate. In the above manner, for each glass plate in the glass load, the measured tempering temperature T mea (=T m, loading or T m or T m, YZ, min ) representing the glass plate is selected. The tempering temperature can also be selected in other ways based on the measurement data of the temperature scanner. The temperature selection must be based on the temperature scanner measurement data usually represented by T mea , which is very important. Therefore, the temperature data analysis program of the calculation unit determines the measured tempering temperature T mea representing the glass plate from the temperature field Ti, yz measured by the temperature scanner, and the tempering temperature is used by the tempering level calculation program of the calculation unit Set the tempering temperature T 0 of the glass plate. The temperature scanner can also be located above the glass, in which case it can measure the surface temperature field on the top surface of the glass. However, for glass with a low top surface emissivity (selected, low-e glass) and unknown coating on the top surface, measurement from the top cannot be performed. In the preferred solution, the temperature is measured with a temperature scanner at the bottom. In some continuous tempering equipment, the glass plates are moved across the line width sequentially and at approximately the same position. In this case, one or at least several pyrometers can be used to measure the glass temperature fairly comprehensively. A pyrometer measures the glass temperature along a line in the direction of glass movement. In the simplest solution, the temperature scanner is a pyrometer or other measuring device that measures the temperature of the glass in a comparable way. In this case, the measured tempering temperature T mea representing the glass sheet is the average temperature of the glass in one or more lines along the direction of glass movement. In a preferred solution, the temperature is measured with a temperature scanner that measures the temperature of the entire surface area of each glass in the glass load. The formula for calculating the glass temperature in the calculation unit determines the change in temperature distribution T(x, t) in the thickness direction of the glass during tempering and cooling according to the energy equation (1): The boundary conditions are equations (2) and (3): Among them, T=temperature, L=thickness of glass plate, k=thermal conductivity of glass, ρ=density of glass, c p =specific heat capacity of glass. In addition, the initial temperature of the glass plate (or a sub-part of the glass plate) T(x, 0)=T 0 . For example, in the above manner, the tempering temperature T 0 of the glass plate is the tempering temperature T mea measured by the temperature scanner from the surface of the glass plate and determined from the measured temperature field of the glass plate. The tempering temperature may consider the natural cooling ΔT m of the glass plate between the temperature scanner 5 and the tempering cooling unit. The residence time from the temperature scanner to the tempering cooling unit is usually 0.1s to 0.7s, depending on the conveying speed of the conveyor line. Therefore, T 0 = T mea- ΔT m , where T mea is based on the temperature of the glass plate measured by the temperature scanner. ΔT m is between 0 and 10°C. Calculate on the thickness L of the glass plate. On the upper surface of the glass, the thickness direction coordinate x=0, and on the lower surface x=L. In heat transfer research, solving the energy equation is a common problem. Several solutions have been proposed in the literature. In addition to the above reference (Rantala2015), there are, for example, the following publications: Field, RE and Viskanta , R., Measurement and prediction of the dynamic temperature distributions in soda-lime glass plates, Journal of American Ceramic Society, vol. 73, 7, pp. 2047-2053, 1990; Gardon, R., Calculation of temperature distribution in glass plates undergoing heat treatment, Journal of American Ceramic Society, vol. 41, 6, pp. 200-209, 1958; Siedow, N., Lochegnies, D., Grosan, T. and Romero, E., Application of a New Method for Radiative Heat Transfer to Flat Glass Tempering, Journal of American Ceramic Society, vol. 88, 8, pp. 2181-2187, 2005. The thermal material properties of glass required in the equation are density ρ, specific heat capacity c p and thermal conductivity k. They are well known and have been provided to the calculation unit in the calculation program. These and radiation characteristics of the glass are represented by P1 in FIG. 6. In the equation, q r, u are the upward (downward q r, l ) net radiant heat flow between the glass and the environment. The method of determining it has been disclosed in the above-mentioned publication, and the glass radiation characteristics required by the method of determining it have been disclosed. The terms q c, u and q c, l are the convective cooling power achieved by the air jet of the tempering cooling unit on the top and bottom surfaces of the glass, which is given by the equation q c, u = h u [T air -T(0, t)] and q c, l = h l [T air -T(L,t)]. In the equation, Hu and h l are the convective heat transfer coefficients of the top and bottom surfaces of the glass. Three alternative techniques for determining them are described above. What is important is that the heat transfer coefficient must be tabulated or listed in an equation in relation to the blowing pressure and the blowing distance. Therefore, the calculation unit can determine them with sufficient accuracy based on the measured blowing pressure and distance. In the equation, T air is the temperature of the tempering air. The boundary condition of the bottom surface also includes the term q ct , which takes into account the heat transferred from the glass (at the contact point between the glass and the roller) to the roller. Usually, a string with poor thermal conductivity has been wound around the rotating roller of the tempered cooling device. In this case, the effect of contact heat transfer can be rounded to q ct =0. Using energy equation (1), calculate the change in temperature distribution T(x, t) in the thickness direction of the glass sheet during the time interval 0≤t≤t tc , where t tc is the glass sheet (point in the glass) during tempering and cooling The stay time in. Generally, it is obtained in the above-mentioned manner based on the conveying speed of the glass sheet and the length of the tempering cooling unit. However, the calculation period t tc must be long enough so that the temperature of the entire thickness of the glass sheet falls below the temperature limit t tcend . Cooling below this temperature no longer affects the residual stress distribution formed on the glass sheet, that is, the degree of tempering. Therefore, the time t tc is the time during which the glass sheet has cooled below the temperature limit t tcend . The typical temperature limit is t tcend =450°C. If the value of the convective heat transfer coefficient (h u , h l ) changes within the time interval 0≤t≤t tc , it will be considered in the calculation. As the solution of the energy equation (1), the temperature distribution T(x, t) in the thickness direction of the glass sheet or the sub-portion of the glass sheet is obtained at the time interval Δt at the tempering cooling time (0-t tc ). The time interval Δt is in the range between 0.001s and 2s, which depends on the thickness of the glass. The temperature distribution is obtained at the density of the interval Δx over the glass thickness L. The thickness interval Δx is (0.01-0.2)L. Then, the temperature distribution T(x, t) calculated from the energy equation (1) is used to calculate the residual stress of the glass sheet. Calculate the tempering level achieved on the glass plate during the tempering process according to the following formula: When the initial condition is σ ij (x, 0) = 0, the thermal elongation ε th (x, t) caused by cooling is calculated with the help of the temperature distribution T(x, t) calculated by the energy equation (1) The resulting load: In the thermal elongation equation, the virtual temperature T f is calculated by the temperature distribution T (x, t) and the material parameter M p : The stress field must meet the general equilibrium conditions: with Solving the tempering stress of glass, that is, the above equation, is a common problem in the research of glass mechanics. It is discussed in the following publications, for example: Aronen, A 2012, Modelling of Deformations and Stresses in Glass Tempering. Tampere University of Technology. Publication , Volume. 1036, Tampere University of Technology; Nielsen, JH, 2009, Tempered Glass-Bolted Connections and Related Problems, Ph.D. thesis, DTU Civil Engineering, Lyngby, Denmark; Carrè, H., 1996, Numerical Simulation of Soda -Lime Silicate Glass Tempering, Journal de Physique IV, vol. 6, no. 1, pp. 175-185. The elongation in the stress equation is the hydrostatic elongation And deflection elongation . The mechanical material properties of the glass required in the equation are the compression modulus K and the shear modulus G, and the thermal expansion coefficients α 1 and α g in relation to time and temperature. They are well known from the aforementioned reference Aronen 2012, for example, and have been provided to the calculation unit in the calculation program. The mechanical properties of the glass plate are represented by P2 in FIG. 6. As a result of the calculation, the thickness direction residual stress distribution σ(x) of the glass sheet, that is, the stress σ value at the thickness interval in the glass thickness, is obtained. The residual stress distribution determines the tempering level of the glass. When the glass sheet is tempered, the increase in strength is particularly represented by the residual stress on the glass surface, that is, the smaller of the value σ(0) or σ(L). When the glass sheet is broken, the safety of the tempered glass sheet (for example, the number of particles calculated in accordance with the standard EN12150-1) depends in particular on the tensile stress (0,5L) at the center thickness of the glass x=0,5L. Based on the calculated residual stress distribution, the reference variable of the tempering level of the glass is selected. The reference variable is based on the calculated residual stress distribution σ(x) in the thickness direction of the glass. In a preferred case, the reference variable is the residual stress at the center thickness of the glass, that is, σ (0, 5L). The reference variable can also be the residual stress of the glass surface, namely σ(0) or σ(L), or the type of average value, or some other residual stress value calculated according to the residual stress distribution σ(x). The general symbol of the residual stress reference variable is σ cal . The residual stress at the center thickness of the glass, namely σ(0,5L), can also be converted into the number of glass particles according to the standard EN12150-1 and used as a reference variable. For example, it can be modified according to the result image published in the above-mentioned reference Akeyoshi 1965, in which for several different glass thicknesses, the number of particles related to the stress σ (0, 5L) has been proposed. Information from the resulting image can be provided to the calculation unit, for example by using polynomial fitting. The applicant's understanding of the relationship between stress σ (0, 5L) and the number of particles is based on the basis of literature and is also the result of a large number of researches collected by the applicant. In the study, a large number of glasses of various thicknesses were tempered, and their σ(0,5L) values were measured using the above-mentioned SCALP measuring equipment, and the glass particles were counted according to the standard EN12150-1. The common symbol of the reference variable is CV, which covers the value σ cal and the number of particles N cal determined based on it. The value of the reference variable CV is used to ensure that the glass loaded glass is successfully tempered into safety glass. The operator performs monitoring based on the reference variable CV of the glass displayed on the display of the alarm device 12 or the alarm given by the alarm device 12. The difference between the above options is that in the former, the operator can determine whether the tempered glass is safety glass based on the reference variable CV (for example, σ(0, 5L)). In the latter, the calculation unit draws a conclusion, and based on the alarm given by the alarm device, it is found that the glass plate does not meet the requirements of safety glass. When the value of the reference variable CV is displayed on the display screen of the alarm device, the former can be used. The latter will be discussed more in the following content. Based on the calculated residual stress distribution, the reference variable CV of the tempering level of the glass sheet is selected, and then it is compared with the predetermined alarm limit value AV. In the case that the comparison between the reference variable and the alarm limit value satisfies a predetermined standard, the calculation unit creates an alarm and sends an alarm message A to the alarm device. The predetermined criterion may be the deviation between the reference variable and the alarm limit value. For example, the deviation of the reference variable from a standard alarm limit value or the deviation of the reference variable from another predetermined alarm limit value. For example, if the calculated value of the reference variable CV is less than the value of the alarm limit AV, an alarm can be created. In the context of the present invention, the alarm limit value may also be referred to as a threshold value. The alarm limit AV has been programmed into the calculation unit. For example, if the reference variable is the stress σ (0, 5L), the alarm limit value is 45 MPa. For example, the alarm limit value is 60 particles, if the reference variable is at least 40 particles according to the particle count of EN12150-1. By comparing the reference variable with the alarm limit value, information about whether the glass panel meets the requirements of the safety glass standard can be obtained. Preferably, the alarm limit of the particle count is at least 50% higher than the minimum value (for example 40) proposed in the standard EN12150-1 (and Table 1), for example 60, when the glass thickness is 4mm to 12mm. Generally, when the glass thickness is 3.8 mm or more, the alarm limit value is between 50 and 80 particles. Generally, when the glass thickness is less than 3.8mm, the alarm limit value is between 25 and 50 particles. Preferably, the stress σ(0,5L) alarm limit value is high enough, based on this stress σ(0,5L), compared with the minimum requirement of the standard EN12150-1, at least 50% more particles should be formed in the glass. Generally, when the glass thickness is 3 mm to 4 mm, the alarm limit σ (0, 5L) value is in the range of 45 MPa to 60 MPa (tensile stress). Generally, when the glass thickness is 5mm to 8mm, the alarm limit σ(0,5L) value is between 35MPa and 55MPa. Generally, when the glass thickness is 10mm-19mm, the alarm limit σ(0,5L) value is between 30MPa and 45MPa. When the reference variable is the residual stress of the glass surface, namely σ(0) or σ(L), the alarm limit is usually 70MPa to 110MPa (compressive stress), or the absolute value of the above-mentioned alarm limit σ(0, 5L) 1.8 times the value. When the alarm device issues an alarm by light, sound, color, or a message displayed on the display screen, the glass load or the reference variable of at least one glass plate in the glass load is lower than the alarm limit. Preferably, the display screen recognizes these glass plates so that they are easy to find in glass-loaded glass plates. Based on this, the operator of the tempering production line can remove the glass sheet that caused the alarm from the produced glass. The standard safety tests of these glass panels provide information, which can be used to complete the above-mentioned methods of ensuring the safety of tempered glass panels. Based on this information, the difference between the alarm limit and the standard minimum can be reduced. By using the calculation program, it is possible to indicate on the display terminal whether the glass load or any glass plate in the glass load is not safety glass or whether it is safety glass, for example, by color description. For example, an alarm can be issued by the color (for example, red) on the glass load pattern displayed on the display terminal, indicating that there is safety glass that does not meet the safety glass standard. The calculation program can also, for example, use a preselected color to indicate the level of the reference variable on the display terminal. For example, green indicates that the glass plate or glass load is safety glass. For example, yellow indicates that the glass plate or glass may be safety glass. For example, red indicates that the glass plate or glass load is not safety glass. The use of the three colors mentioned above requires the setting of two limit thresholds AV, namely the thresholds AV1 and AV2. Therefore, for example, a glass plate whose reference variable is higher than the upper threshold value AV2 is regarded as safety glass, and a glass plate whose reference variable CV is lower than the lower threshold value AV1 is regarded as unqualified safety glass. The glass plate with the reference variable CV between AV1 and AV2 is considered to be safety glass. The calculation program can also use colors to indicate the level of the reference variable. For example, this can be done by placing the reference variable CV selected for each type of glass on top of the glass plate temperature scanner image or other pattern indicating the boundary of the glass plate and highlighting it in a predetermined color. The predetermined color may be, for example, red, yellow, and green. For example, if the reference variable value is between 49 MPa and 50 MPa, and the reference variable in this range can be used as safety glass, a color (for example, green) that can be selected by the user is displayed in the temperature scanner image. This method can be used to ensure that the glass load is successfully tempered into safety glass according to various glass plate thicknesses and glass plate types. Therefore, for example, when the calculated tempering level (CV) is close to the alarm limit (AV), and when a predetermined standard is met, the selected temperature and/or heating time of the tempering furnace, that is, the conveyor line speed, is readjusted. Separately, when the calculated tempering level is close to the alarm limit, when the predetermined standard is met, the selected blowing pressure and/or blowing distance of the air nozzle in the tempering cooling unit can be readjusted. If a deviation between the tempering level and the alarm limit is detected, the tempering furnace control system of the tempering equipment can automatically increase/decrease the heating time of the glass plate in the furnace, or suggest increasing/decreasing the heating time of the glass plate. In addition, if a deviation between the tempering level and the alarm limit is detected, the tempering cooling unit control system of the tempering equipment can automatically suggest adjusting the blowing pressure and/or blowing distance of the glass sheet cooling nozzle. It is obvious to those skilled in the art that the present invention is not limited to the solutions described herein, but the idea of the present invention can be applied in various ways within the scope set by the scope of patent application.
