LARSSON, Jan-Olof (Tolvmansvägen 3, Luleå, S-975 96, SE)
DU PLESSIS, Louwrens Theodor (Koningshoeven 1, BW's Hertogenbosch, NL-5235, NL)
LARSSON, Jan-Olof (Tolvmansvägen 3, Luleå, S-975 96, SE)
| CLAIMS 1. A system of monitoring a condition of a component of a machine, the component having a cyclic characteristic, characterized in that, the system comprises signal processing means arranged to: determine a cycle time of the component by means of a cycle time signal, - split vibration samples into cycles having a cycle time equal to the component cycle time, assigning each sample with a time stamp based on a sample's time position in its respective cycle, create a number of segments equal to a predetermined resolution, divide the number of segments over the component cycle time, assigning each segment with a time extension within the component cycle time, split the time stamped samples of each cycle into the segments by matching a sample's time stamp to be within a segment's time extension, for each segment, create a result by summing the samples within the segment and dividing the sum by the number of samples of the segment, provide the results of the segments as data of vibration over the component cycle. 2. The system according to claim 1 , characterized in that the system comprises vibration input means, cycle time input means, a sampler, and output means where the vibration input means is arranged to receive a vibration signal representing a magnitude of vibration a vibration sensor is subjected to when coupled to the machine, the cycle time input means is arranged to receive a cycle time signal from which the cycle time length of the component is determinable, the sampler is arranged to digitize the vibration signal into the vibration samples, the output means is arranged to visualize vibration information over a cycle of the component at the predetermined resolution by means of the results of the segments. 3. The system according to claim 2, characterized in that the vibration input means is arranged to receive one or more further vibration signals representing a magnitude of vibration one or more further vibration sensors are subjected to when coupled to the machine, and in that the sampler is arranged to digitize the one or more further vibration signals into the vibration samples. 4. The system according to claim 2 or 3, characterized in that the sampler is further arranged to normalize each vibration signal separately. 5. The system according to any previous claim, characterized in that the signal processing means is further arranged to normalize the results of the segments. 6. The system according to any previous claim, characterized in that the signal processing means is further arranged to normalize the results of the segments to a predetermined resolution. 7. A method of monitoring a condition of a component of a machine, the component having a cyclic characteristic, characterized in that, the method comprises: - determining a cycle time of the component, splitting vibration samples into cycles having a cycle time equal to the component cycle time, assigning each sample with a time stamp based on a sample's time position in its respective cycle, - creating a number of segments equal to a predetermined resolution, dividing the number of segments over the component cycle time, assigning each segment with a time extension within the component cycle time, splitting the time stamped samples of each cycle into the segments by matching a sample's time stamp to be within a segment's time extension, - for each segment, creating a result by summing the samples within the segment and dividing the sum by the number of samples of the segment, providing the results of the segments as data of vibration over the component cycle. |
TECHNICAL FIELD
The invention concerns condition monitoring systems and is directed in certain embodiments to condition monitoring of rotating components and other components having a cyclic behavior, such as gears and bearings and also to devices comprising such components, for example gearboxes.
BACKGROUND
A gear is a machine part that is designed to mesh with another similar machine part to transmit rotational motion. The most commonly used gears include planetary gears, spur gears, helical gears, bevel gears, worm gears, and rack and pinion gears. Gears mesh with each other in many different ways to transfer motion from one gear to another. In addition, gears can be used to increase or decrease the speed of rotation. For example, a smaller gear driven by a larger gear will have a greater speed of rotation than the larger gear. Conversely, a larger gear driven by a smaller gear will have a lower speed of rotation that the smaller gear. Gears may be housed in a gearbox. Gearboxes are used to transmit rotational motion in many different types of systems. A gearbox typically consists of at least one gear set and bearings to enable the gears to rotate.
The gears and bearings in a gearbox may have defects, or they may fail over time, or they may simply wear out. For example, the loads and stresses that are imposed on the bearings and gears may exceed acceptable limits, leading to failure of or damage to the gears and/or bearings. The damaged or failed components may be replaced once the existence of the damage or failure is known. Alternatively, the teeth of a gear or raceways/rollers of a bearing may simply begin to wear down through prolonged usage. Vibration analysis is an established non-intrusive method for monitoring the condition of mechanical components within rotating machinery. For example, the condition of a component may be determined by considering the frequency and magnitude of vibration signals produced by the component. Generally, components in good condition, such as gears with complete sets of teeth, produce smaller amplitude vibrations than components in poor condition, such as gears with chipped or missing teeth. There is still room for improvements in condition monitoring systems in general and in particular for detecting and identifying gear meshing faults and bearing faults in for example gearboxes.
SUMMARY
It is an object to disclose a method of detecting and identifying gear meshing faults and bearing faults in for example gearboxes and to disclose a condition monitoring system executing the method.
In a first aspect of the present disclosure, a condition monitoring system is provided for condition monitoring a machine comprising a multiple of interacting cyclical components, such as a gearbox. The system receives vibration data from one or more sensors that are attached to the machine. For example attached to the housing of the machine. The system will focus on individual components based on the cycle times of the components. The vibration data, in the form of digital samples, is broken down into cycles/periods, each corresponding in time length to the cycle time of the component focused on. Each sample is given a time stamp within the cycle that the sample belongs to, thus ranging between zero and the cycle time of the component focused on. A number of segments are created, each with a time extension within the cycle time of the component focused on. The number of segments being a desired output resolution over a cycle of the component focused on. All samples from each cycle/period are split into the segments. The splitting is based on matching a sample's time stamp to be within a segment's time extension. The samples of each segment are summed and divided by the number of samples split into the segment. This will allow speed fluctuations of a monitored component. The result, one component cycle of vibration information, can be further processed and/or presented in for example a visual circle representing the component.
In another aspect of the present disclosure a condition monitoring system is provided that receives vibration data from one or more sensors that are attached to a machine having cyclical components, such as a gearbox. Samples of the vibration data are time stamped based on a cycle time of a component focused on. A number of segments are created each with a time extension based on the cycle time of the component focused on. The samples are split into the segments based on matching a sample's time stamp to be within a segment's time extension. For each segment, the samples of the segment are summed and divided by the number of samples split into the segment. This will allow a degree of speed fluctuation of a monitored machine. The results are suitably presented as a cycle of the component focused on.
In another aspect of the present disclosure a system of monitoring a condition of a component of a machine is provided. The component has a cyclic characteristic. The system comprises signal processing means arranged to determine a cycle time of the component by means of a cycle time signal. The signal processing means is further arranged to split vibration samples into cycles having a cycle time equal to the component cycle time, and assigning each sample with a time stamp based on a sample's time position in its respective cycle. The signal processing means is also arranged to create a number of segments equal to a predetermined resolution, and divide the number of segments over the component cycle time, assigning each segment with a time extension within the component cycle time. The signal processing means is even arranged to split the time stamped samples of each cycle into the segments by matching a sample's time stamp to be within a segment's time extension, and for each segment, create a result by summing the samples within the segment and dividing the sum by the number of samples of the segment. The signal processing means is finally arranged to provide the results of the segments as data of vibration over the component cycle.
In a preferred embodiment, a system preferably comprises vibration input means, cycle time input means, a sampler, and output means. The vibration input means is arranged to receive a vibration signal representing a magnitude of vibration a vibration sensor is subjected to when coupled to the machine. The cycle time input means is arranged to receive a cycle time signal from which the cycle time length of the component is determinable. The sampler is arranged to digitize the vibration signal into the vibration samples. The output means is arranged to visualize vibration information over a cycle of the component at the predetermined resolution by means of the results of the segments.
Suitably the vibration input means is arranged to receive one or more further vibration signals representing a magnitude of vibration one or more further vibration sensors are subjected to when coupled to the machine. The sampler is then preferably also arranged to digitize the one or more further vibration signals into the vibration samples. In some embodiments the sampler is further arranged to normalize each vibration signal separately.
In some embodiments the signal processing means is further arranged to normalize the results of the segments and possibly to a predetermined resolution. The resolution can be used for matching a limited number of different colors or shades of gray. In another aspect of the present disclosure a method of monitoring a condition of a component of a machine is provided. The component has a cyclic characteristic. The method comprises a number of steps. A first step determines a cycle time of the component. A second step splits vibration samples into cycles having a cycle time equal to the component cycle time. A third step assigns each sample with a time stamp based on a sample's time position in its respective cycle. A fourth step creates a number of segments equal to a predetermined resolution. A fifth step divides the number of segments over the component cycle time. A sixth step assigns each segment with a time extension within the component cycle time. A seventh step splits the time stamped samples of each cycle into the segments by matching a sample's time stamp to be within a segment's time extension. An Eighth step creates, for each segment, a result by summing the samples within the segment and dividing the sum by the number of samples of the segment. A ninth step provides the results of the segments as data of vibration over the component cycle.
The different features according to the present teachings can be combined in any desired manner.
BRIEF DESCRIPTION OF THE DRAWING
The invention will now be explained in more detail with reference to the following figures, in which Fig. 1 illustrates a representative application of the present invention,
Fig. 2 illustrates two meshed gears, Fig. 3A, 3B, 3C illustrate a waveform and how cycles of different components map onto such a waveform. Fig. 4A, 4B illustrate splitting of samples into segments of a component over different waveforms and/or different cycles. Fig. 5 illustrates a flowchart of a method of according to the invention.
Fig. 6 illustrates an example of a visualization of one gear. DETAILED DESCRIPTION
A condition monitoring system according to the following preferred embodiments may be used, without limitation, for automotive, industrial, energy generating and any other applications comprising components having a repetitive cyclic behavior. Such components can be, but not limited to, shafts, gears and bearings. The system is especially advantageous for use on machines and machine units, such as gearboxes, as it can focus on one cyclic /periodic component of such a machine or machine unit at a time. The system is especially suited for use on applications that have a fluctuating rotational speed, creating varying cycle times, even fluctuations within a cycle/period, such as wind or water energy generating applications.
Representative examples will now be described in connection with Figures 1 to 6. Figure 1 illustrates a presently preferred application of a condition monitoring system according to the present teachings. The condition monitoring system comprises a digital processing unit 120, to which suitably an input device 124 such as a keyboard, and an output device 122 such as a display unit. Coupled with the system is a speed sensor 112 such as a tacho, and one 110 or more 111 vibration sensors. The one 110 or more 111 vibration sensors are suitably coupled to a signal conditioner 114 that will do any necessary analog amplification and filtering in addition to digitizing the vibration signals from the sensors to digital samples. The digitization will at least time stamp the beginning of a sequence of samples of a waveform. If more than one sensor is utilized, then it is important that the digitizations of the vibration signals from each sensor are synchronized with each other. Further units 126 for communication or further processing may also be present.
The vibration sensors 110, 111 are attached/coupled to a machine 100 that comprises a component 105 having cyclic characteristics, such as full turn rotation or another type of repetitive and cyclic movement such as part turn movements. The vibration sensors 110, 111 are used to sense vibrations 107, 109 generated by the component 105 whose condition is to be monitored.
The speed sensor 112 generates a speed signal that is used to determine the cycle time length, period, of a component 105 that is to be monitored. The speed signal can determine either directly or indirectly the cycle length of the component 105. If the speed sensor is directly sensing a speed of the component, then the determination is done directly. On the other hand, if the speed sensor measures a speed of another component, then the relationship between the component whose speed is measured and the component 105 that is currently monitored determines a speed relationship between the components, through for example gear ratios, which speed relationship is then used to indirectly determine the cycle time length of the component 105.
Referring to Figure 2, if a speed is measured at a first gear 210 and the monitored component is a second gear 220 directly meshed with the first gear 210, then the gear ratio between the first gear 210 and the second gear 220 will determine the cycle time length, period, of the second gear 220. To be observed is that cycle times can vary between cycles and also during cycles. Depending on the method of determining the speed of a monitored component, it might not be possible to make adjustment for changes of cycle times during the cycles or even catch every variation between cycles. The cycle time of a component might only be updateable every 10 cycles for example. Part of the novel signal processing can accommodate at least part of these possible speed variations. Changes of cycle times can be due to wind gusts on a wind- powered generator causing speed variations in a monitored gearbox. To be able to describe the digital signal processing in more detail, Figure 3A illustrates a typical vibration signal 300 picked up by one sensor, here just illustrated as amplitude 304 variations over time 302. We are making an assumption that we are looking at a machine comprising two meshing gears, for example as illustrated in Figure 2. The gears have a 5:2 ratio, meaning that when the big gear has two cycles, the small gear will have five cycles. As can be seen in Figure 3B, the waveform of Figure 3A has been divided 310, 312, 314, 316, 318 into five cycles 320, 322, 324, 326, 328, corresponding to the cycles of the small gear. As mentioned, the cycle time lengths, periods, 320, 322, 324, 326, 328 can be the same or different, in dependence on how the cycle time lengths are determined. As can be seen, there is some signal fluctuations 330, 332, 334, 336, 338 towards the end of 310, 312, 314, 316, 318 each cycle 320, 322, 324, 326, 328. In a corresponding manner in Figure 3C, the waveform of Figure 3A has been divided 350, 352 into the cycle lengths 360, 362 of the bigger gear. As can be seen in Figure 3C, there are also some recurring signal fluctuations 370, 372 around three fifths into each cycle. Traditionally for each component, each gear in this example, all corresponding cycles would be averaged, sample-by-sample. For the smaller gear, all the cycles 320, 322, 324, 326, 328 of Figure 3B would be averaged and for the larger gear, all the cycles 360, 362 of Figure 3C would be averaged. This works well in most circumstances when the cycle lengths are the same and that there are no speed fluctuations within the cycles. However, if there are fluctuations in speed, then the signal fluctuations of interest will also fluctuate in time, meaning that the fluctuations of interest will end up in different samples within a cycle between cycles. For example, if there is a signal fluctuation of interest in sample thirty-five, as counted from a cycle start, in a first cycle, maybe, due to speed fluctuations, the signal fluctuation of interest ends up in sample thirty-eight, as counted from a cycle start, in a second cycle. This means that when the two cycles are averaged, then the signal fluctuation of interest will disappear because, due to the change of position between cycles, it is seen as noise. To enable condition monitoring of equipment that have speed fluctuations, the inventors propose a method of combining samples of cycles in such a manner that signal fluctuations of interest are enhanced, even if there is not a perfect match, sample to sample, between the cycles. According to the invention, each cycle of samples of a component, i.e. collection of samples over one or more time periods each corresponding to a cycle time of the component, is de-sampled into a predetermined number of segments. The predetermined number of segments is less than the number of samples of a cycle of the component. Each cycle is divided into the predetermined number of segments. Samples of a cycle are split into the segments in dependence of the time stamp of the segment and the time stamp, stamped or calculated, of the sample. When all the samples of all the cycles have been split into their appropriate segments, then all the samples of a segment are summed and then divided by the number of samples that were added in that segment. Unless the number of samples of a cycle is evenly dividable into the segments, then not all of the segments will comprise the same number of samples. This is the principle, which of course mathematically can be done in other ways. Figures 4A and 4B illustrate splitting of samples into segments of a component over different waveforms and/or different cycles. Figure 4A and 4B illustrate how 8 samples 431 , 432, 433, 434, 435, 436, 437, 438, of a first cycle and 8 samples 441 , 442, 443, 444, 445, 446, 447, 448 of a second cycle are split into three segments 410, 412, 414. As a note, the fifth sample 435 of the first cycle in Figure 4A and the fourth sample 444 of the second cycle in Figure 4B in this example illustrate fluctuations of interest with the same origin. According to the timeline 402 of Figure 4A, the first three samples 431 , 432, 433 are put in a first segment 410, the next two samples 434, 435 are put in a second segment 412, and the last three samples 436, 437, 438 are put in a third segment 414. According to the timeline 402 of Figure 4B, which represents a different cycle, be it from a same sensor or a different sensor. The first two samples 441 , 442 are put in the first segment 410, the next three samples 443, 444, 445 are put in the second segment, 412 and the last three samples 446, 447, 448 are put in the third segment 414. To be noted is that the samples of the fluctuations of interest 435, 444 both end up in the second segment 412, 413 even though they are different samples from each respective cycle.
The samples in each segment are then added together and divided by the number of samples comprised in that segment. In the example, the first segment 410 will comprise five samples 431, 432, 433, 441 , 442, the second segment 412 will comprise five samples 434, 435, 443, 444, 445, and the third segment 414 will comprise six samples 436, 437, 438, 446, 447, 448. Thus the five samples of the first segment are summed and then divided by five, the five samples of the second segment are summed and then divided by five and the six samples of the third segment are summed and then divided by six. The order of summing and division can of course be done mathematically in a different order/manner, this order is just for illustrative purposes.
In another numerical example, consider a component having a cycle time length of 15 seconds. Further, the predetermined number of segments is five. Thus, the first segment will have samples with time stamps, stamped or calculated, of 0 to 2.9 seconds. The second segment will have samples with time stamps, stamped or calculated, of 3.0 to 5.9 seconds. The third segment will have samples with time stamps, stamped or calculated, of 6.0 to 8.9 seconds. The fourth segment will have samples with time stamps, stamped or calculated, of 9.0 to 11.9 seconds. Finally, the fifth segment will have samples with time stamps, stamped or calculated, of 12 to 14.9 seconds. Further consider that during a cycle of 15 seconds 17 samples were taken from a sensor. The first four samples (0, 0.8, 1.8, 2.7) will be in the first segment, the next three samples (3.5, 4.4, 5.3) will be in the second segment, the following four samples (6.2, 7.1 , 7.9, 8.8) will be in the third segment, the next three samples (9.7, 10.6, 11.5) will be in the fourth segment, and the last three samples (12.4, 13.2, 14.1) will be in the fifth segment. The segments do not have the same number of samples. This also enables samples of non- complete cycles to be used.
Figure 5 illustrates a flowchart of a method of according to the invention. In a first step 510, one or more measurement waveforms are sampled. If more than one sensor is used, then the sampling of each sensor is synchronized. Suitably a tacho signal or other speed determining means is also acquired. In a second step 512, it is suitable to take an absolute value of each sample of the sampled waveform or waveforms. This is to increase the available dynamic range. In a third step 514 it is suitable to normalize each sampled waveform individually. In a fourth step 516 a gear/bearing/component counter is initiated. In a fifth step 518 a waveform counter is initiated. In a sixth step 520 a cycle counter for the component at hand is initiated (the component at hand is given by the component counter) in the waveform at hand (the waveform at hand is given by the waveform counter). In a seventh step 522 a cycle length for the cycle at hand is determined (the cycle at hand is given by the cycle counter), the cycle length being dependent on the component at hand (how fast it is turning) and the speed at the time of the samples taken for the cycle at hand. In an eighth step 524, all the samples taken during the cycle length are extracted/determined. In a ninth step 526, the extracted samples are split into a predetermined number of segments based on the timestamp of the individual extracted samples. The timestamp can be that given or calculated from a sequence start. The predetermined number of segments representing one cycle of the component at hand and being the same for all the cycles of all the waveforms for that component, each component having a separate set of segments. In a tenth step 528, the number of individual samples that are in each individual segment is updated. In an eleventh step 530, the cycle counter is updated. In a twelfth step 532, it is determined if all cycles have been done or not, i.e. it is determined if there are more complete or part cycles in the waveform at hand or not. If there are not, then the process continues to step thirteen 534, if there are more cycles, then the process returns to the seventh step 522. In a thirteenth step 534, the waveform counter is updated. In a fourteenth step 536, it is determined if all waveforms have been processed or not. If all the waveforms have been processed for the component at hand, then the process continues with the fifteenth step 538, otherwise the process returns to the sixth step 520. In a fifteenth step 538, all the samples split into each individual segment are summed, then each sum is divided with the number of samples actually summed in that segment as determined by the tenth step 520. The result is stored in each respective segment of the component at hand. In a sixteenth step 540, the component counter is updated. In a seventeenth step 542, it is determined if all components have been processed or not, if all the components have been processed then the process continues with an eighteenth step 544, otherwise the process returns to the fifth step 518. In an eighteenth step 544, further processing on the sequence of segments is done for each component, such as normalizing each sequence into a predetermined number of levels for output for even further processing or a visualization. Figure 6 illustrates an example of a typical one cycle circular 600 visualization output of one component such as a gear. In the example, the predetermined resolution of one cycle of the component in question is 64 segments 602. The predetermined output resolution, both as to number of segments and to a normalization of each segment, will be set according to the use of the output. If, as in this example a very low-resolution monochrome display is used, 64 segments, with 16 steps in each segment, might be the optimum. In this example, the darker the segment is, the larger the amplitude of vibration is. In other applications where a higher resolution color display is used, the preferred resolution might be 360 segments per cycle, each segment then representing one degree of the cycle of the component in question, with 128 or 256 steps per segment. If further processing is to be performed, a higher or lower output resolution might be chosen. The predetermined output resolution might also be influenced by sampling frequency and the resulting number of samples in a cycle for the component in question, that is the relationship between sampling frequency and cycle time. As can be seen in Figure 6, there are four areas/sections 610, 612, 614, 616 which show high vibration amplitudes and can thus be considered to be areas/sections of interest 610, 612, 614, 616. These sections of interest 610, 612, 614, 616 are evenly spaced out around the circle 600. It seems likely that a directly or indirectly coupled component with a cycle time of one quarter the cycle time of the component at hand, has a defect. This could be two gears that are meshed together.
In summary, a fluctuation of speed of the components monitored is possible with the method of de-sampling the samples of a cycle before any averaging of the samples is performed.
The invention is not restricted to the above-described embodiments, but may be varied within the scope of the following claims.
NL09081 SPB
14 JANUARY 2010
FIGURE 1 - illustrates a representative application of the present invention:
100 a system that is investigated,
105 component with cyclic characteristics,
107 vibration to first vibration sensor,
109 vibration to optional second vibration sensor,
1 10 first vibration sensor,
1 1 1 optional second vibration sensor,
1 12 a speed sensor/tacho,
1 14 signal conditioning and sampling of sensor signals,
120 a digital signal processing unit,
122 output device such as a display unit,
124 input device such as a keyboard,
126 further processing unit/communication unit.
FIGURE 2 - illustrates two meshed gears:
210 a first larger diameter gear,
220 a second smaller diameter gear.
FIGURES 3A, 3B and 3C - illustrate a waveform and how cycles of different components map onto such a waveform:
300 waveform,
302 timeline,
304 amplitude,
310 end of first cycle of first component,
312 end of second cycle of first component,
314 end of third cycle of first component,
316 end of fourth cycle of first component,
318 end of fifth cycle of first component,
320 time length of first cycle of first component, 322 time length of second cycle of first component,
324 time length of third cycle of first component,
326 time length of fourth cycle of first component,
328 time length of fifth cycle of first component,
330 point of interest of first cycle of first component,
332 point of interest of second cycle of first component,
334 point of interest of third cycle of first component,
336 point of interest of fourth cycle of first component,
338 point of interest of fifth cycle of first component,
350 end of first cycle of second component,
352 end of second cycle of second component,
360 time length of first cycle of second component,
362 time length of second cycle of second component,
370 point of interest of first cycle of second component,
372 point of interest of second cycle of second component.
FIGURES 4A and 4B - illustrate splitting of samples into segments of a component over different waveforms and/or different cycles:
402 timeline,
410 segment N,
412 segment N+1 ,
414 segment N+2,
431 a sample of first splitting,
432 a sample of first splitting,
433 a sample of first splitting,
434 a sample of first splitting,
435 a sample of interest of first splitting,
436 a sample of first splitting,
437 a sample of first splitting,
438 a sample of first splitting,
441 a sample of second splitting,
442 a sample of second splitting, a sample of second splitting,
a sample of interest of second splitting,
a sample of second splitting,
a sample of second splitting,
a sample of second splitting,
a sample of second splitting. - illustrates a flowchart of a method of according to the invention: a first step, sampling one or more measurement waveforms, if a plurality of sensors are used, then these are synchronized, a tacho signal or other speed determining means are also acquired, a second step, taking an absolute value of sampled measured waveform or waveforms,
a third step, normalizes each waveform individually,
a fourth step, initiates a gear/bearing/component counter, a fifth step, initiates a waveform counter,
a sixth step, initiates a cycle counter for the component at hand (the component at hand is given by the component counter) in the waveform at hand (the waveform at hand is given by the waveform counter),
a seventh step, determines the cycle length for the cycle at hand (the cycle at hand is given by the cycle counter), the cycle length being dependent of the component at hand and the speed at the time of the cycle,
an eighth step, extracts all the samples taken during the cycle length,
a ninth step, splits the extracted samples into a predetermined number of segments based on timestamp of the extracted samples, the predetermined number of segments representing one cycle of the component at hand and being the same for all the cycles of all the waveforms for that component, each component having a separate set of segments, a tenth step, updates the number of individual samples that are in each individual segment,
an eleventh step, updates the cycle counter,
a twelfth step, determines if all cycles have been done or not, i.e. it is determined if there are more complete or part cycles in the waveform at hand or not, if there are not, then the process continues to step thirteen 534, if there are more cycles, then the process returns to the seventh step 522,
a thirteenth step, updates the waveform counter,
a fourteenth step, determines if all waveforms have been processed or not, if all the waveforms have been processed for the component at hand, then the process continues with the fifteenth step 538, otherwise the process returns to the sixth step 520, a fifteenth step, sums all the samples split into each individual segment, then divides each sum with the number of samples actually summed in that segment as determined by the tenth step
520, the result is stored in each respective segment of the component at hand,
a sixteenth step, updates the component counter,
a seventeenth step, determines if all components have been processed or not, if all the components have been processed then the process continues with an eighteenth step 544, otherwise the process returns to the fifth step 518,
an eighteenth step, does further processing on the sequence of segments for each component, such as normalizing each sequence into a predetermined number of levels.
FIGURE 6 - illustrates an example of a visualization of one
600 a circular one cycle illustration,
602 predetermined output resolution unit,
610 a first section of interest,
612 a second section of interest, a third section of interest, a fourth section of interest,
Next Patent: FOAMING OR FOAMABLE PREPARATIONS HAVING A CONTENT OF ALKYL LACTOSIDES