FI91919C - Procedure for analyzing paper - Google Patents

Description

91919

METHOD FOR ANALYZING PAPER - FORFARANDE F0R ANALYZING PAPPER

The present invention relates to a method 5 for analyzing paper to locate errors in the machine members of a paper machine, wherein at least one quality quantity of paper is detected from the paper web as a function of web length and a quality quantity signal characterizing the quality quantity is generated.

A method is now known for locating defects in the machine members of a paper machine by analyzing some quality of the paper web and determining the operating frequencies of the machine members of the paper machine simultaneously and continuously. This is the so-called synchronized mean 15 method. This results in a rather complex error monitoring system, which is characterized in particular by the simultaneous and continuous detection of paper errors and the detection of the operating frequency of the machine elements, as well as their continuous comparison with each other. It should be noted that in the system, the phase shift of the error signals caused by different machine members as a function of the paper length of the same quality signal remains unresolved. Likewise, in the old system, no quality magnitude signal can be assigned to the paper so that the value of the quality magnitude signal at a particular point in the paper could be expressed. The system is also roon complicated, difficult to implement and potassium.

The object of the present invention is to obviate the above-mentioned drawbacks. In particular, it is an object of the invention 30 to provide a novel method for locating errors in the machine members of a paper machine, so that in the method it is not necessary to continuously and simultaneously monitor the values of the paper quality quantity and the operating frequencies of the machine members.

It is a further object of the invention to provide a method by which the machine members acting on defects of a quality of some quality of paper can be located each 2 91919 separately.

With respect to the features of the invention, reference is made to claim 1.

The invention is based on the premise that at least one quality variable of paper is detected from the paper web as a function of web length to generate a quality variable signal, and the quality variable signal is mathematically analyzed for regular repetitive error rates of a certain frequency.

The method according to the invention thus differs from the previously known one in that the detected quality variable signal is analyzed mathematically without the synchronization signals of the machine elements.

15 It is new and surprising that a quality variable signal can be analyzed mathematically without synchronization signals. It should be noted that the variation in paper quality is small. In addition, the number of paper machine machines that affect the quality of the paper is large. In addition, the operating frequencies of the machine elements are very close to each other. However, the measurements performed have yielded very good results when analyzing the paper quality quantity signal in a mathematically new way.

The analysis of the paper quality quantity signal according to the invention can be performed by statistical-mathematical methods known per se, applied in a new way, provided that a length quantity is also detected in addition to the paper quality quantity signal. The signal can be analyzed, e.g., first at a certain test frequency n Hz and then at at least one other test frequency (n ± p) Hz, where p is a fraction of 1 / 10,000 to 1/2, e.g.

0.001 - 0.1. In connection with the analysis, recurring errors of the quality variable and / or their period numbers are regularly determined by mathematical means at the experimental frequencies. The analysis can be performed, for example, at intervals of 0.0001 to 0.001 to 0.01 to 0.1 Hz, depending on the number of cycles in question and the number of machine elements of the paper machine corresponding to the number of cycles of 3,91919, i.e. according to the desired resolution. When a period number is found in the method with a continuous uniform characteristic variation of the quality variable in question, the period number in question determines the machine element, i. the machine element causing the error has the same period number. At the same time, the frequency of the error signal is obtained. The operating frequencies or period numbers of the machine elements can thus be determined completely separately from the detection and analysis of the paper quality signal.

After the number of repetitive errors in a paper quality quantity or quality quantity signal at a certain frequency, i. the machine element causing the error has been determined as described above, the analysis 15 can be continued with other period numbers to locate other sources of errors of the same quality in the same way. Thus, the method can be used to determine the error signals of a certain quality signal of the paper, i.e. the different aerial frequencies of the errors of the quality variable in question, as well as the forms of the error signals of these 20 and thus the causes of the errors of the paper quality variable.

Further, the analysis can be continued mathematically by summing the errors caused by different machine elements in the respective quality variable signal at different frequencies, whereby the maximum and minimum errors of the interaction of certain machine elements of the given quality quantity can be mathematically determined. In this way, the user of the paper machine can easily ascertain, e.g., whether he is able to produce a product of a certain quality for the 30 quality quantities in question with his machine.

Furthermore, when determining the paper quality quantity errors with the method according to the invention, and in particular the errors caused by different machine elements in that quality quantity, the paper machine user is informed how many 35% of the quality signal variation is explained by each machine element error, e.g. In this case, the user of the machine can schedule repairs to 4,91919, taking into account the quality requirements and the demand for different paper grades in an appropriate manner.

In the method according to the invention, any quality quality of the paper can be monitored, which is affected by the errors of the machine members of the pa-5 machine, e.g.

the basis weight of the paper, the thickness, the gloss, the transmission, the thickness of the coating, the roughness, etc. The detection of these quality quantities can be performed with any kind of detectors known per se. The operating frequencies of the machine members of the paper machine 10 can be determined with any kind of detectors known per se.

By web length is meant in this context any quantity relative to the length of the web or a quantity dependent on it, such as the absolute length of the web, the time (constant speed of the web), etc.

Comparing the method according to the invention with the previously known method for locating errors of paper machine machine members 20, it can be stated that the method according to the invention offers a substantially more versatile, cheaper device and operating method and more reliable variations in the quality of the paper produced by that paper machine.

A further advantage of the invention is that the explanation percentage, i.e. how much the error of a machine element explains the variation of the quality variable signal, can be calculated mathematically.

A further advantage of the invention is that the error signals caused by the different machine elements and their interaction correlate with the original quality variable signal. Based on the quality variable signal, faulty machine elements can be traced mathematically.

. >

A further advantage of the invention is that paper errors 5,91919 caused by different machine members can be located on the paper web on the basis of the combined effect of the machine edge error signals.

The invention will now be described in detail by means of exemplary embodiments with reference to the accompanying drawings, in which the figures 1a to d show the basis weight, thickness, gloss and resistance to paper length detected in a paper mill, i.e. as a function of the machine direction length of the paper, Fig. 2 shows the basis weight of the paper determined from the quality quantity signal of Fig. 1a as a function of paper length at a frequency, Figs. as a function of one frequency, and Figures 6-9 show Figures 2-5 of the paper basis weight, thickness, gloss, and transmission, respectively, determined from the quality variable signal of Figures 1a-d as a function of the length of the paper 20 at another frequency.

EXAMPLE 1

In one paper mill, printing paper was produced, with a basis weight of 57 g / m *. The printing paper was supercalendered. From a paper web distance of 4000 m, 25 basis weight, thickness, gloss and transmission were determined. The basis weight, thickness, gloss and resistance of the web are shown in Figures 1a to d as a function of the length of the web over a distance of 40 m. It can be seen from Figure 1a that the scattering of the basis weight of the paper was 1.22 g / m *. Mathematical analysis of the signal representing the basis weight at 6.35 Hz resulted in the variation of the basis weight shown in Figure 2 as a function of the longitudinal direction of the web.

Correspondingly, the determined paper thickness is shown in Fig. 3 as a function of web length, gloss 35 in Fig. 4 as a function of web length, and transmission in Fig. 5 as a function of web length. Comparing the images with each other, it can be seen that the basis weight, thickness, ---------- r ~ 91919 6 gloss correlate with each other, respectively the transmission correlates inversely with the previous ones. Upon further examination of the machine members of the paper machine, it was found that the counter roll of the first coating station had a running speed of 5.35 Hz, i.e. the poor condition of the coating roll clearly caused square weight, thickness, gloss and transmission errors.

In the same experiment, the characteristic variation of the quality variable signal of images 1a to d at 6.47 Hz was determined.

The basis weight is shown in Figure 6. Similarly, the variations in paper thickness, gloss and transmission at 6.47 Hz were determined, the results are shown in Figures 7-9. A more detailed analysis of the paper machine showed that the counter-roll speed of the second coating station was 6.47 Hz.

The embodiment shows that with the method according to the invention it is possible to locate the machine edge errors of a paper machine unambiguously and accurately by mathematically analyzing a quality variable signal of some quality of paper.

The embodiments are intended to illustrate the invention, and applications of the invention may vary within the scope of the appended claims.

Claims (1)

91919 CLAIMS A method for analyzing paper to locate defects in paper machine machine members, wherein at least one quality quantity of paper is detected from a paper-5 web as a function of web length, generating a quality quantity signal characteristic -10 to known period numbers of the machine members of a paper machine, characterized in that the quality variable signal is analyzed at a certain experimental frequency n Hz and then at at least one experimental frequency (n ± p) Hz, where p = fraction 1 / 10,000 - 1/2, and that ana - In connection with 15 lysings, regularly recurring errors in the quality of the signal are determined by mathematical means at the experimental frequencies. Forfare for analysis of papers for localization of papersmaskins maskinorgans fel, varvid åtminstone en qualitiesstorhet av papperet detekasas 5 ur pappersbanan såsom en function avans banans långd-stor-het, en qualetsstorheten characteristics of qualitative-torhetssignal bildas, up to the whole frequency and quality-10 tetsstorhetens Fels periodtal jamfors med pappersmaskinens maskinorgans kåda periodtal, kånne-t ecknat dárrav, att quality quadrature signal analysis at all forsox frequency n Hz and dhrefrequency 1 / 10,000 - 1/2, and a sample of the analysis with a mathematical method is best used for determining the frequency-sensitive signal quality.