200918381 九、發明說明： 【發明所屬之技術領域】 本發明係有關於一種用於計算軌道車輛中之車軸的方 法’而此軌道車輛具有一車軸計算感測器，其尤其被配置 在軌道中之一彎曲區域內，其中使一在此車軸計算感測器 之輸出端處之類比信號量變曲線與一第一切換界限作比較 ’而就在此第一切換界限之向上超限段與隨後的向下超限 段之間產生了一數位計算脈衝。 【先前技術】 車軸計算感測器在鐵道系統中被使用於無軌信號發送 及亦用於其他切換及信號發送功能。軌道車輛之鐵製車輪 的磁場影響效應主要係針對此目的而被使用。兩通道感測 器被提供以偵測此軌道車輛之運行方向。當車輛之車輪運 行於上時’此兩個感測器通道會相繼地產生用於偵測運行 方向之時序偏差信號。 這些依據感應工作原理而操作之車軸計算感測器可被 分成單通道或兩通道設計及被分割成若干近接開關，其可 感測諸鐵製車輪在一產生磁場之感測器上的反應，以及在 若干啣合於鐵路軌道周圍並具有獨立發射器與接收器之系 統上的反應。 所有以感應方式操作之感測器共同地具有以下所述之 事實：此諸感測器均對尤其發生在軌道彎曲區域中之轉向 架傾斜的錯誤很敏感。由於此所謂之側斜行進效應（Slde w a y s ι· u η n i n g e f f e c t)，一短距下沉可能在類比信號量變曲線 -6- 200918381 中發生在第一切換界線之下’以致產生數位計算脈衝之分 割。此g十算脈衝之分割則由於巳通過感測器之車軸數目而 導致錯誤之計算。此側斜行進效應係以特別明顯之方式發 生在具有相當小半徑之彎曲軌道中，諸如那些存在於區域 運輸系統中者。軌道車輛在當運行通過彎曲軌道時之運行 狀態尤被敘述於Fnednch，F發表在當地鐵路刊物Der Nahverkehr 1 9 8 5年二月號第52-61頁所揭之「軌道中之急 轉彎道的導引」一文中。 特別由於側斜行進效應之問題，其他原理已在區域運 fej系統（例如D . C .巡迴路線）中被用於無軌信號發送，其 中此側斜f了進效應並不會發生。 【發明内容】 本發明之目的在於揭示一種藉著車軸計算感測器來計 算車軸之方法，其中側斜行進效應並不會產生不良之後果 ’從而可避免車軸計算之錯誤。 此目的可 曲線中存在~ 按照本發明而被達成，因爲當類比信號量變 降至第一切換界限以下之短距下沉時，此數 丛π R衝之分割將可被避免，因爲該信號量變曲線在此 下/几之區域中會被提升’或被轉變爲一數位脈衝’並經由 OR運算而與原本被分割之脈衝進行邏輯結合，以便形成此 數位計算脈衝。 在第—變化型式中，該單—量變曲線可以說 在下沉區域中以—句絡曲被卡# ^ Μ 包絡曲線方式被理直。 -'、而，亦可使用此單一下沉而個別地產生一被疊置在 200918381 原已被分割的脈衝上之數位脈衝，因此使得一未經分割之 組合脈衝被形成，而其被進一步地評估爲一數位計算脈衝 〇 第一變化型式中之信號下沉的跨越可依據申請專利範 圍第2項而被實施，其中類比信號量變曲線被數位化，並 將與一位於第一切換界限下方之第二切換界限作比較，且 介於位在第二切換界限上方之數位化信號量變曲線的諸最 大値間之複數個信號區段將被複數個以數位方式產生之信 號區段所取代，而此諸以數位方式所產生之信號區段則藉 由一直線而將諸最大値相互連接，且藉此方式而被修正之 數位化信號量變曲線將被轉變爲類比之型式，並取代原本 之類比信號量變曲線。第二切換界限確保多個源自兩個不 同車軸之真實計算脈衝不會被結合而形成一個單一衝。此 信號下沉必須不得超過某一程度。 可依據申請專利範圍第3項實施第二變化型式，其中 該信號下沉可說是並不被修整，反而是爲了校正之目的而 被刻意地使用，其中此類比信號量變曲線在梯度方面係藉 由一微分元件予以估算’其中此數位脈衝係產生於諸轉折 點處，在此該梯度變爲負。此數位脈衝被形成於此信號下 沉區域中，並經由一邏輯OR組合而與原已被分割之脈衝 作邏輯結合，以便可形成該數位計算脈衝。一單獨之計算 脈衝因此藉由在其上疊置額外之脈衝而從原先之雙重脈衝 被產生’而此額外之脈衝起始在類比信號量變曲線之轉折 點處，亦即在其下沉之開始處。 -8- 200918381 申請專利範圍第4項界定該數位脈衝之時序長度係取 決於一先前時距中之平均梯度，以致使信號下沉之區域被 該額外數位脈衝所完全代替。 【實施方式】 下文中將配合參照圖式而更詳細地說明本發明。 第1至4圖係以上部信號曲線所呈現之可比較圖式顯 示對車軸計算感測器之類比影響信號1或1.1。 第1圖說明在一車軸計算感測器中之諸比率，而此車 "：' 軸計算感測器被安裝在一直線軌道上，而一軌道車輛之車 輪則運行於其上。顯然地，在此車軸計算感測器之輸出端 處會有一因而產生之信號量變曲線2，其啓動一開啓信號 以便在當達到一第一切換界限3時可產生一數位計算脈衝 4。當此車輪繼續運行時，將會首先達到影響信號1之最大 値且因此亦達到信號量變曲線2的輸出端側之最大値。此 影響信號1隨後將會減小，其中當有第一切換界限3之向 下超限段存在時，該產生數位計算脈衝4之信號將會被再 i 度地消除。因此，針對此軌道車輛之各車輪將會精密地產 生一數位計算脈衝4。一種可使得用於啓動計算脈衝4之 關斷界限稍微較低於開啓界限之滯後現象·的影響已被忽略 ，以便可簡化此圖式之呈現。 然而，如果此車軸計算感測器被安裝在一個彎曲軌道 上’則由於軌道車輛之轉向架的傾斜，亦即由於此軌道車 輛之側斜行進效應，將使得車軸計算錯誤可能發生，如第 2圖所示。在本文中，位於此車軸計算感測器之輸出端處 -9- 200918381 的is 5虎里變曲線2 _ 1係以一.小下沉5爲其特徵。如果此下 沉5掉落至第一切換界限3之下，則即使僅一車輪已通過 測量點，仍將會產生兩個數位計算脈衝4.丨及4.2。 第3圖顯示一種用於避免使計算脈衝4被分割成兩個 4.1/4.2之第一變化型式。在本文中’整個類比信號量變曲 線2. 1首先被數位化並儲存於一記憶體中。然後，經由一 在第一切換界限3之下的第二切換界限6 ,將會檢查複數 個最大値是否係位於11與12間之時距中，而此時距則被指 定爲第二切換界限6。如在此情形下，諸被數位化之數値 從第一最大値maxi起始至第二最大値max2止遞增地以（ max2 - maxi) /(t…2 - Um)之量被增加。以一此方式被 改變之信號量變曲線2. 1的數位化繪圖於是被再度地轉變 爲類比數値，在此情形下，下沉5可說是藉由位於諸最大 値m a X 1與m a x 2間之直線連接7而被跨越。此被予理直之 信號量變曲線被用以產生該數位計算脈衝4。 第4圖顯示一種用於避免形成雙重之計算脈衝4 . 1 / 4 2 之第二變化型式。在此’位於車軸計算感測器上之類比影 響信號1 _ 1係經由一微分元件而被估算。此微分信號係車 軸計算感測器上之類比影響信號1 . 1之陡峭度的測量標準 。在諸轉折點處，亦即當陡峭度=0時，此微分元件之符號 改變。在諸P點處（符號在此變爲負），延遲信號8. 1、8.2 、8 .3被形成’其時序延伸則視在其之前的平均陡峭度而定 °經由—邏輯◦ R運算以結合原已被分割之脈衝4 . 1 /4.2與 延遲信號8. 1/8.2/8.3，短距下沉5被抑制，從而使正確之 -10- 200918381 單獨計算脈衝4被形成。 【圖式簡單說明】 第1圖顯示計算脈衝在當車軸計算感測器被配置在直 線軌道時之產生情形； 第2圖顯示雙重之計算脈衝在當車軸計算感測器被配 置在一彎曲軌道時之產生情形； 第3圖顯示一用於避免第2圖中所示雙重之計算脈衝 的第一變化型式；及 第4圖顯示一用於避免第2圖中所示雙重之計算脈衝 的第二變化型式。 【主要元件符號說明】 1/1.1 2/2.1 3 4/4.1/4.2 5 6 7 類比影響信號 信號量變曲線 第一切換界限 數位計算脈衝 T-'f—r 讥 第二切換界限 直線連接 8.1/8.2/8.3 延遲信號 -11-200918381 IX. INSTRUCTIONS OF THE INVENTION: TECHNICAL FIELD The present invention relates to a method for calculating an axle in a rail vehicle, and the rail vehicle has an axle calculation sensor, which is especially configured in a track. In a curved region, wherein an analog semaphore curve at the output of the axle calculation sensor is compared with a first switching limit, and the upward overrun and the subsequent direction of the first switching limit are A digital calculation pulse is generated between the lower overrun segments. [Prior Art] The axle calculation sensor is used in the railway system for trackless signal transmission and also for other switching and signaling functions. The magnetic field effect of the iron wheels of rail vehicles is mainly used for this purpose. A two-channel sensor is provided to detect the direction of travel of the rail vehicle. When the wheels of the vehicle are running, the two sensor channels successively generate timing deviation signals for detecting the direction of travel. These axle calculation sensors, which operate according to the principle of induction operation, can be divided into single-channel or two-channel designs and divided into a plurality of proximity switches, which can sense the reaction of the iron wheels on a sensor that generates a magnetic field. And a reaction on a number of systems that are coupled around the railway track and have separate transmitters and receivers. All inductively operated sensors collectively have the fact that the sensors are sensitive to errors in the tilting of the steering wheel, particularly in the curved region of the track. Due to this so-called skewing effect (Slde ways ι· u η ning effect), a short-distance sinking may occur below the first switching boundary in the analog semaphore curve -6-200918381, so that the division of the digital calculation pulse is generated. . The division of this g-timed pulse results in an error calculation due to the number of axles passing through the sensor. This side-slope travel effect occurs in a particularly pronounced manner in curved tracks having a relatively small radius, such as those found in regional transportation systems. The operating state of a rail vehicle while running through a curved track is described in Fnednch, F. The local railway publication Der Nahverkehr, February 2, 1985, pp. 52-61, "The sharp turn in the orbit" Guided in the article. In particular, due to the problem of the skewed travel effect, other principles have been used for the transmission of trackless signals in regional fej systems (e.g., D. C. patrol lines), where the side effects do not occur. SUMMARY OF THE INVENTION An object of the present invention is to disclose a method for calculating an axle by means of an axle calculation sensor, wherein the side-slope traveling effect does not cause a defect, thereby avoiding an error in the axle calculation. This purpose can exist in the curve ~ according to the invention, because when the analog signal volume falls below the first switching limit, the segmentation of the number of π R rush can be avoided because the semaphore changes The curve is boosted or converted to a digital pulse in this lower/several region and logically combined with the originally split pulse via an OR operation to form this digitally calculated pulse. In the first variation mode, the single-quantity variation curve can be said to be straightened in the sinking region by the -sentence curve by the card #^ 包 envelope curve. - ', but this single sinking can also be used to individually generate a digital pulse that is superimposed on the originally split pulse of 200918381, thus causing an undivided combined pulse to be formed, which is further Evaluating the crossing of the signal sink in the first variation of the digitally calculated pulse 可 can be implemented in accordance with item 2 of the scope of the patent application, wherein the analog semaphore curve is digitized and will be below a first switching limit The second switching limit is compared, and the plurality of signal segments between the maximum turns of the digitized semaphore curve above the second switching limit are replaced by a plurality of digitally generated signal segments, and The signal segments generated by the digital method are connected to each other by a straight line, and the digitized signal amount curve corrected by this method is converted into an analogy type and replaces the original analog signal. Quantitative curve. The second switching limit ensures that a plurality of real calculated pulses originating from two different axles are not combined to form a single punch. This signal must not sink more than a certain degree. The second variation can be implemented according to item 3 of the scope of the patent application, wherein the sinking of the signal can be said to be not trimmed, but is deliberately used for the purpose of correction, wherein such a ratio of the semaphore curve is borrowed in terms of gradient It is estimated by a differential component where the digital pulse is generated at the turning points where the gradient becomes negative. This digital pulse is formed in the sinking region of the signal and is logically combined with the originally split pulse via a logical OR combination so that the digitally calculated pulse can be formed. A separate calculated pulse is thus generated from the original double pulse by superimposing additional pulses thereon, and this additional pulse starts at the turning point of the analog semaphore curve, ie at the beginning of its sinking . -8- 200918381 The scope of the patent application defines that the timing length of the digital pulse depends on the average gradient in a previous time interval, so that the area where the signal sinks is completely replaced by the extra digital pulse. [Embodiment] Hereinafter, the present invention will be described in more detail with reference to the drawings. Figures 1 through 4 show the analogy of the above signal curves showing the analog effect of the sensor on the axle calculation signal 1 or 1.1. Figure 1 illustrates the ratios in an axle calculation sensor that is mounted on a linear track while the axle of a rail vehicle is running. Obviously, at the output of the axle calculation sensor there is a resulting semaphore curve 2 which initiates an open signal to produce a digital calculation pulse 4 when a first switching limit 3 is reached. When this wheel continues to run, it will first reach the maximum 影响 that affects the maximum value of signal 1 and therefore also the output side of semaphore curve 2. This influence signal 1 will then be reduced, wherein the signal generating the digital calculation pulse 4 will be cancelled again when there is a downward overrun of the first switching limit 3. Therefore, for each wheel of the rail vehicle, a digital calculation pulse 4 will be generated. One effect that allows the hysteresis of the turn-off limit for starting the calculation pulse 4 to be slightly lower than the turn-on limit has been ignored so that the presentation of this pattern can be simplified. However, if the axle calculation sensor is mounted on a curved track, the axle calculation error may occur due to the inclination of the bogie of the rail vehicle, that is, due to the side travel effect of the rail vehicle, such as the second The figure shows. In this paper, the is 5 Huli curve 2 _ 1 at the output of this axle calculation sensor -9-200918381 is characterized by a small sinking 5 . If this sinking 5 falls below the first switching limit 3, even if only one wheel has passed the measuring point, two digital calculation pulses 4. and 4.2 will be generated. Figure 3 shows a first variant for avoiding the division of the calculation pulse 4 into two 4.1/4.2. In this context, the entire analog semaphore curve 2.1 is first digitized and stored in a memory. Then, via a second switching limit 6 below the first switching limit 3, it will be checked whether the plurality of maximum chirps are in the time interval between 11 and 12, and the distance is then designated as the second switching limit. 6. As in this case, the digitized number 値 is incremented from the first maximum 値maxi to the second maximum 値max2 by increments of (max2 - maxi) / (t...2 - Um). The digitized plot of the semaphore curve 2.1 changed in this way is then again converted to an analog number 値, in which case the sink 5 can be said to be located at the maximum 値ma X 1 and max 2 The straight line connection 7 is crossed. This semaphore curve is used to generate the digital calculation pulse 4. Figure 4 shows a second variant for avoiding the formation of a double computed pulse 4 . 1 / 4 2 . Here, the analogy signal 1 _ 1 located on the axle calculation sensor is estimated via a differential component. This differential signal is a measure of the steepness of the signal that affects the signal 1.1 in the axle calculation sensor. At the turning points, that is, when the steepness = 0, the sign of the differential element changes. At points P (the sign becomes negative here), the delay signals 8.1, 8.2, 8.3 are formed 'the timing extension is determined by the previous average steepness. Combined with the originally split pulse of 4. 1 / 4.2 and the delayed signal of 8. 1 / 8.2 / 8.3, the short-range sinking 5 is suppressed, so that the correct -10 200918381 alone calculated pulse 4 is formed. [Simple diagram of the diagram] Figure 1 shows the calculation of the pulse when the axle calculation sensor is placed in a linear orbit; Figure 2 shows the double calculation pulse when the axle calculation sensor is configured in a curved orbit The situation of the time; FIG. 3 shows a first variation for avoiding the double calculation pulse shown in FIG. 2; and FIG. 4 shows a first calculation for avoiding the double calculation pulse shown in FIG. Two variants. [Description of main component symbols] 1/1.1 2/2.1 3 4/4.1/4.2 5 6 7 Analogy affects signal semaphore curve First switching limit digit calculation pulse T-'f-r 讥 second switching limit straight connection 8.1/8.2 /8.3 Delay signal-11-