Scale accuracy monitoring

Abstract

A belt scale monitoring and / or computing system, which comprises a conveyor belt scale of the known kind operatively installed relative to a conveyor belt system and providing a first data feed a data-driven calculation model system operatively installed relative to the conveyor belt system and providing a second data feed and a computational module operatively receiving in real-time the first and second data feeds and configured to process the first and second data feeds.

Description

[0001] SCALE ACCURACY MONITORING

[0002] BACKGROUND TO THE INVENTION

[0003] This invention relates to scale accuracy monitoring. More particularly, the present invention relates to a system for monitoring the accuracy of a conveyor belt scale in real time.

[0004] Conveyor belt scales, also known as belt weighers or weightometers, are devices used to measure the weight of materials being transported on a conveyor belt, typically in the mining, manufacturing and food processing industries.

[0005] The primary function of a conveyor belt scale is to measure the rate of material flow, and the total quantity of material transported over a given period. This measurement involves two main variables: the weight of the material on the belt and the speed of the belt.

[0006] The weight of the material on the belt is measured by load cells and weigh idlers. Load cells are sensors that measure the force exerted by the material on the belt and are placed under the conveyor belt, typically within a scale frame or weighbridge. As material passes over the scale, the load cells detect the weight and convert it into an electrical signal. Weigh idlers, on the other hand, are specially designed rollers that support the conveyor belt and are strategically placed to ensure accurate weight measurement. The weigh idlers transfer the weight of the material on the belt to the load cells.

[0007] Speed is measured by a speed sensor which is usually a wheel that rides against the surface of the belt, and which generates pulses that correspond to the belt’s speed. Each pulse represents a unit of belt travel, and the frequency of pulses is directly proportional to the belt speed.

[0008] An electronic integrator combines and processes the data from the load cells and the speed sensor to calculate the material flow rate and the total weight of material conveyed. It uses the formula:

[0009] Weight = Load Cell Output x Belt Speed

[0010] Ensuring the accuracy, repeatability and reliability of the measurements obtained from a belt scale is vitally important for the functioning of the belt scale. High accuracy is essential for inventory management, quality control, and regulatory compliance. Some systems offer accuracy within ±0.25% or better, which is crucial for applications requiring precise weight measurements.

[0011] Repeatability refers to the scale’s ability to produce consistent results under the same conditions. Repeatable measurements ensure reliability over time and are critical for maintaining process consistency and quality.

[0012] Transient fluctuations in measured outputs may occur and may take the form of temporary deviations, discrepancies, anomalies, inaccuracies, instability and noise.

[0013] Longer term or steady-state inaccuracy may occur as a result of system drift caused by environmental factors, mechanical wear, and load cell fatigue. Due to the importance of accuracy, repeatability and reliability of belt scales, these systems need to be calibrated on a regular basis to rectify and compensate for system drift, ensuring longterm accuracy and compliance with industry standards.

[0014] Different forms of calibration are required. These include zero calibration and span calibration. Zero calibration entails adjusting the scale to read zero when the belt is empty. This step removes any residual loads or drifts. Span calibration ensures the scale measures a known weight accurately across its entire range. This involves placing a test weight on the belt and adjusting the scale settings to match the known weight.

[0015] Despite the importance of accuracy of the measurements, calibration is often a costly and time-consuming exercise and is associated with system down-time. A need exists for a mechanism for providing a real-time indication of accuracy or at least an early warning of potential deviations.

[0016] Furthermore, even a properly calibrated system may suffer from transient fluctuations, which cannot be addressed by means of calibration. These transient fluctuations are furthermore often difficult or even impossible to identify or predict.

[0017] A belt scale as detailed above will be referred to herein as a “belt scale of the known kind”.

[0018] Various measuring and monitoring systems and components for monitoring various parameters of conveyor belts are known in the art. For example, belt rip events are typically measured, monitored or detected using sensor loops which are installed in the belt. Measuring and monitoring systems of this kind have known drawbacks. Sensor loops require a minimum bottom cover thickness to be used and easily get damaged due to material impact and the like. Such damage may lead to false alarms, unplanned downtimes or incorrect or inaccurate monitoring.

[0019] Of late, data driven calculation models are utilised to replace certain physical measuring and monitoring systems. Such data driven calculation models receive inputs, such as electrical power draw, mass flow rates, conveyor speeds, idler dimensions, belt dimensions, and the like, and utilises artificial intelligence, machine learning and computational algorithms to serve as early warning systems by providing predictive analytics and anomaly detection. These systems are capable of inferring mass flow by using the data as mentioned together with average bulk densities of materials transported by the conveyor.

[0020] A need also exists for means of more accurately predicting failure or inaccuracy of individual load cells of a belt scale, to allow a belt scale to continue operating accurately despite such failure or inaccuracies, and to more easily install, calibrate and commission individual load cells of a belt scale.

[0021] It is an object of the invention to provide scale monitoring and processing systems and methods that will, at least partially, address the above disadvantages.

[0022] It is also an object of the invention to provide scale monitoring and processing systems and methods which will be a useful alternative to existing systems and methods. SUMMARY OF THE INVENTION

[0023] In accordance with a first aspect of the invention there is provided a belt scale monitoring and / or computing system, comprising: a conveyor belt scale of the known kind operatively installed relative to a conveyor belt system and providing a first data feed; a data driven calculation model system operatively installed relative to the conveyor belt system and providing a second data feed; and a computational module operatively receiving in real-time the first and second data feeds and configured to process the first and second data feeds.

[0024] The first data feed may relate to a real-time measurement of weight and / or mass flow rate of material operatively carried by the conveyor belt. The second data feed may relate to a data driven inferred calculation of weight, mass or mass flow rate.

[0025] The computational module may comprise a processing module which may be associated with a memory arrangement. The computational module may be configured for executing an algorithm in accordance with a predetermined program code.

[0026] The processing module may be integrated with a programmable logic controller (PLC) of the conveyor belt. Alternatively, the processing module may be a remote processing module associated with a backend, which may be provided in data-transfer communication with the conveyor belt scale.

[0027] The computational module may comprise a comparator. The system may be configured for monitoring accuracy of the belt scale. The comparator may be configured to determine discrepancies or differences between the first and second data feeds. Such discrepancies or differences may be classified relative to an acceptable tolerance level. The comparator may be configured for identifying an event when the discrepancy or difference exceeds the acceptable tolerance level.

[0028] The comparator may be configured for providing an output when the event is identified. The output may an alert communicated externally or an adjustment effected on a control system of the conveyor belt.

[0029] The system may be configured as a mass-flow monitoring and computing system. The computational module may operatively process the first and second data feeds into a third data feed, which may be provided as an output feed of the system.

[0030] The third data feed may relate to a normalised, compensated, composite, processed, aggregated, adjusted or smoothed computational mass flow rate.

[0031] In accordance with a second aspect of the invention there is provided a method of monitoring, measuring and / or processing data relating to a quantity of material carried by a conveyor belt system in real time, the method comprising the steps of: 51) receiving a first data feed from a conveyor belt scale of the known kind installed relative to or as part of the conveyor belt system;

[0032] 52) receiving a second data feed from a data driven calculation model system operatively installed relative to the conveyor belt system;

[0033] 53) continuously processing the first and second feeds utilising a processing module; and

[0034] 54) providing an output based on the processing of the first and second feeds.

[0035] The method may relate to monitoring an accuracy of the belt scale. The processing module may comprise a comparator module. The processing may comprise utilising the comparator module to compare the first and second feeds. Step S3 may be preceded by the step of establishing an acceptable tolerance level for discrepancies between values obtained from the first and second feeds.

[0036] Step S3 may comprises the sub-steps of: calculating a difference between values obtained from the first and second feeds; and identifying an event when the difference between the values obtained from the first and second feeds exceeds the acceptable tolerance level.

[0037] Step S4 may comprise providing an output when an event is identified. The output may be an alert or a control provided to a control system of the conveyor belt. The control may cause the belt to be stopped.

[0038] Alternatively, or in addition, the method may relate to monitoring and / or computing mass flow rate of the conveyor. Step S3 may comprise continuously processing the first and second data feeds into a third data feed. Step S4 may comprise providing the third data feed as an output. Step S3 may further comprises processing the third data feed into a normalised computational mass flow rate by aggregating the first and second data feeds. Aggregating of the first and second data feeds may occur by means of statistical methods, filtering techniques and / or machine learning algorithms that identify and mitigate anomalies.

[0039] In accordance with a third aspect of the invention there is provided a load cell signal monitoring and processing unit, comprising: a plurality of load cell input terminals each for operatively receiving a load cell input signal from one of a plurality of load cells; a processing unit associated with a memory arrangement, the processing unit provided in data-flow communication with the plurality of input terminals and configured for operatively receiving the load cell input signals and to process the load cell input signals into one or more processed output signals; an output terminal configured to transmit the one or more processed output signals.

[0040] The unit of the third aspect of the invention may furthermore comprise at least one further input terminal for operatively receiving at least one further input signal from at least one further sensor device Each further input terminal may be provided in data-flow communication with the processing unit. The further sensor device may comprise a speed sensor device. The further input signal may be a speed sensor input signal.

[0041] The unit may comprise between one and ten load cell input terminals. In some cases, the unit may more specifically comprise between one and four or between two and four load cell input terminals.

[0042] In a particularly preferred embodiment, the unit may comprise four load cell input terminals.

[0043] The processing unit may be configured for identifying a fault condition. The processing unit may be configured for: comparing a value associated with each load cell input signal to a predefined threshold range; and identifying a first fault condition when the value of one or more load cell input signal falls outside of the predefined threshold range. The first fault condition may typically relate to load cell failure.

[0044] Additionally, or alternatively, the processing unit may be configured for: comparing values associated with the complement of load cell input signals to identify differences between the values; comparing each identified difference to a predefined allowable difference value; and for identifying a second fault condition when an identified difference exceeds the predefined allowable difference value.

[0045] The second fault condition may relate to load cell imbalance.

[0046] A predefined threshold range and a predefined allowable difference value may be stored on the memory arrangement.

[0047] The processing unit may provide a fault output when an identified fault condition is identified. The fault output may be provided via one of the output terminal and a separate fault output terminal. The separate fault output terminal may typically comprise a solid state relay output, which may be overturned when the fault condition is removed.

[0048] The unit may further comprise an alarming means for providing an alarm when a fault output is provided by the processing unit.

[0049] In some embodiments according to the third aspect of the invention, the processing unit may be configured for: calculating an average value from the values obtained from the load cell input signals; and providing the calculated average value as the processed output signal.

[0050] The processing unit may be configured for calculating the average value based on a selection of values from a selection of input signals.

[0051] The unit may include means for overriding one or more load cells manually. In such cases, the values associated with the load cell input signal(s) of the overridden load cells may be disregarded when calculating the average value. The processing unit may furthermore, or alternatively, be configured for, when calculating the average value, isolating and / or removing values associated with an input signal in respect of which a fault condition has been identified.

[0052] The unit may further include a display for displaying values associated with the inputs received via the input terminals, the one or more processed output signals and the like.

[0053] In accordance with a fourth aspect of the invention there is provided a method of monitoring and processing load cell input signals, the method comprising the steps of:

[0054] M1) receiving a plurality of load cell input signals via a plurality of load cell input terminals;

[0055] M2) utilising a processing unit to receive the load cell input signals and to process the load cell input signals into one or more processed output signals;

[0056] M3) providing the processed output signal to an output terminal.

[0057] The method may comprise the further step Mx of utilising the processing unit to identify a fault condition. This step Mx may comprise at least some of the following sub-steps:

[0058] Mx1) comparing a value associated with each load cell input signal to a predefined threshold range;

[0059] Mx2) identifying a first fault condition when the value of one or more load cell input signal falls outside of the predefined threshold range;

[0060] Mx3) comparing values associated with the complement of load cell input signals to identify differences between the values;

[0061] Mx4) comparing each identified difference to a predefined allowable difference value;

[0062] Mx5) identifying a second fault condition when an identified difference exceeds the predefined allowable difference value;

[0063] Mx6) providing a fault output when an identified fault condition is identified; and

[0064] Mx7) utilising alarming means for providing an alarm when a fault condition is identified.

[0065] Step M2 may comprise at least some of the sub-steps of: calculating an average value from values obtained from the load cell input signals; providing the calculated average value as the processed output signal; making a selection of input signals and wherein the average value is calculated considering the selection; manually overriding one or more load cells and wherein the average value is calculated by disregarding values associated with overridden load cells; and isolating and / or removing values associated with an input signal in respect of which a fault condition has been identified when calculating the average value.

[0066] In accordance with a fifth aspect of the invention there is provided a conveyor belt scale comprising: a scale subframe; at least a first weigh idler set supported by an idler subframe mounted to the scale subframe by means of a first pair of load cells; and a load cell signal monitoring and processing unit according to the third aspect of the invention wherein each load cell is connected to a load cell input terminal of the load cell signal monitoring and processing unit.

[0067] Furthermore, a tachometer (speed sensor) may be connected to the tachometer input terminal.

[0068] The first data feed may comprise a processed output signal obtained from a load cell signal monitoring and processing unit according to the third aspect of the invention.

[0069] The first data feed may comprise a processed output signal obtained in accordance with a method of monitoring and processing load cell input signals according to the fourth aspect of the invention.

[0070] BRIEF DESCRIPTION OF THE DRAWINGS

[0071] The invention will now be described in more detail, by way of example only, with reference to the accompanying drawings in which:

[0072] Figure 1 shows a sectioned front view of a belt scale to which a belt scale accuracy monitoring system in accordance with the invention is installed;

[0073] Figure 2 shows a schematic diagram of a method of using a belt scale accuracy monitoring system; and

[0074] Figure 3 shows a schematic front view of a load cell signal monitoring and processing unit in accordance with an aspect of the invention.

[0075] DESCRIPTION OF THE ILLUSTRATED EMBODIMENTS

[0076] Before any embodiments of the invention are explained in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," or "having" and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. Unless specified or limited otherwise, the terms "mounted", "connected", "engaged" and variations thereof are used broadly and encompass both direct and indirect mountings, connections, supports, and couplings and are thus intended to include direct connections between two members without any other members interposed therebetween and indirect connections between members in which one or more other members are interposed therebetween. Further, "connected" and "engaged" are not restricted to physical or mechanical connections or couplings. Additionally, the words "lower", "upper", "upward", "down" and "downward" designate directions in the drawings to which reference is made. The terminology includes the words specifically mentioned above, derivatives thereof, and words or similar import. It is noted that, as used in this specification and the appended claims, the singular forms "a," "an," and "the," and any singular use of any word, include plural referents unless expressly and unequivocally limited to one referent. As used herein, the term “include” and its grammatical variants are intended to be non-limiting, such that recitation of items in a list is not to the exclusion of other like items that can be substituted or added to the listed items.

[0077] Referring to the drawings, in which like numerals indicate like features, a non-limiting example of a belt scale accuracy monitoring system (or simply “system”), in accordance with the invention, is generally indicated by reference numeral 10.

[0078] The system 10 is installed to, relative to or as part of a conveyor belt system 12. The conveyor belt system 12 generally comprises a support frame 14 and various longitudinally spaced apart roller sets 16. The roller sets 16 support a belt 18, typically in troughed fashion.

[0079] The portion of the conveyor belt system 12 shown in figure 1 comprises a belt scale arrangement 20. The belt scale arrangement 20 comprises a belt scale of the known kind. As such, the roller set 16 shown in figure 1 comprises a weigh idler set which is supported by a roller subframe 22. The roller subframe 22, in turn, is supported by the support frame 14 by means of first and second load cells 24. The support frame 14 is generally supported on a surface 26. It will be appreciated that the belt scale arrangement 20 may comprise more than one weigh idler sets which are linearly spaced apart and provided in relatively close proximity to each other. Therefore, in some cases, the scale arrangement 20 comprises four load cells.

[0080] The belt 18 of the conveyor belt system 12 is operatively driven by drive motors (not shown) in known fashion. The conveyor belt system 12 comprises a programmable logic controller (PLC) which is used to control and regulate various inputs, such as power provided to the drive motors to control the travel speed of the belt. The PLC is typically housed in an electrical cabinet 28. The PLC is schematically shown by reference numeral 30.

[0081] The belt scale arrangement 20 is also associated with a belt speed measuring device of the known kind, typically taking the form of a wheel or roller running against a bottom surface of the belt.

[0082] The belt scale arrangement 20 is, at least upon installation and commissioning, calibrated in known fashion (by performing a zero calibration and a span calibration as aforementioned). At least initially, therefore, the belt scale arrangement 20 is assumed to provide a relatively accurate measurement of weight or mass flow rate of material carried by the conveyor. In some cases, as discussed in more detail below, the calibration is done using a load cell signal monitoring and processing unit 100.

[0083] Measurements taken by the belt scale arrangement 20 is provided continuously in real time to the PLC 30 as a first data feed. The PLC 30 is therefore equipped to record the mass and / or mass flow rate of material transferred or conveyed using the conveyor belt system 12. The first data feed therefore takes the form of a real time measurement of weight or mass flow rate of material carried by the conveyor belt system 12. It will be appreciated that references herein to the “first data feed” may relate to a single or multiple streams of data from one or more load cells 24 associated with the one or more weigh idler sets. As discussed more fully below, the first data feed may be received from a load cell signal monitoring and processing unit 100 and integrator combination.

[0084] The system 10 is also associated with a data driven calculation model system or module 32. The calculation model system 32 utilises data driven calculation models coupled with artificial intelligence and other analytical protocols and algorithms to provide predictive analytics, anomaly detection and data monitoring. The calculation model system 32 is provided in data-flow communication with the PLC 30 and receives from the PLC 30 various inputs, such as electrical power draw, mass flow rates, conveyor travel speed and the like (relating to the scale but also the conveyor as a whole). The calculation model system 32 is also provided with known dimensions and quantities such as idler dimensions, belt dimensions, average bulk densities of material conveyed by the belt, and the like. In use, the calculation model system 32 therefore performs data analysis to provide data-driven tools as inputs to the PLC 30.

[0085] In some implementations, the calculation model system 32 may, for example, provide inputs relating to general anomalies, potential belt rip events, drive monitoring, condition monitoring, energy efficiency monitoring, maintenance support, and electrical network analysis.

[0086] The calculation model system 32 may furthermore be configured to provide, based on data analysis and monitoring, mass flow calculations of material conveyed by the conveyor belt system 12. The mass flow calculations are based on inferred values based on average bulk densities of the material conveyed and monitoring of parameters of the conveyor. The calculation model system 32 may, in real time, provide the inferred mass-flow values as a second data feed to the system. The second data feed therefore relates to a data driven inferred mass flow calculation (and is therefore not an actual mass flow or weight measurement, as such). The second data feed may therefore lack accuracy. That said, since the mass flow calculation by the calculation model system 32 is an inferred calculation based on actual measured data inputs, the measurement is not likely to become inaccurate over time due to wear, misalignment, or other maintenance issues of the scale.

[0087] The system 10 further comprises a comparator module (which is not shown as such in the figures). It will be appreciated that the comparator module may take various forms and may comprise a standalone unit, may be a remote unit, such as a unit located in a control pulpit, may be cloud-based or backend related unit, or may be incorporated with, or made up by, one of the existing components of the system 10. For example, in some cases, the PLC 30 may fulfil the role of the comparator module. In other cases, the calculation model system 32 may fulfil the role of the comparator module. The comparator is associated with a processing module and a memory arrangement.

[0088] Irrespective of the embodiment thereof, the comparator module is configured to receive the first and second data feeds in real time and to compare them.

[0089] The system 10 may be set up such that, during calibration of the belt scale arrangement 20, the values measured by the belt scale arrangement 20 is compared to the inferred values obtained from the calculation model system 32, to establish a baseline comparison between the belt scale arrangement 20 and the calculation model system 32. Theoretically, these two values should be closely related and should correlate with each other. That said, since the values are determined or obtained using different hardware and methodologies, differences or deviations between the values may be present (with the measurements obtained from the belt scale arrangement 20 assumed to be the more accurate measurements). Therefore, an initial difference or deviation between the first and second data feeds is established and defined. Alternatively, or in addition, both the belt scale arrangement 20 and calculation model system 32 may simultaneously be calibrated, in which case, at least initially, substantially equal results should be returned by both systems.

[0090] Based on this initial difference or deviation between the first and second data feeds (if present), an acceptable tolerance is defined. The acceptable tolerance may, alternatively, be a predetermined acceptable tolerance. The acceptable tolerance is therefore an acceptable difference or deviation between the first and second feeds. Furthermore, a threshold value, which is above the acceptable tolerance may also be defined.

[0091] During normal operation of the conveyor belt system 12 and the system 10, the comparator continuously receives the first data feed (as shown at 34) and the second data feed (as shown at 36) and calculates a real-time difference between the first and second data feeds (as shown at 38).

[0092] The difference so calculated is furthermore compared, in real time to the acceptable tolerance (as shown at 40).

[0093] If the difference exceeds the acceptable tolerance, the difference is compared to the threshold (as shown at 42).

[0094] In cases where the difference is below the threshold, an alert is provided (as shown at 44) and the conveyor system 12 and system 10 continues to operate as normal. In this situation, the difference, although above the acceptable tolerance, is still small enough that an immediate stoppage of the conveyor belt system 12 is not yet warranted. However, the alert provided by the system 10 informs maintenance personnel or operators that potential issues may be present, allowing them to investigate. This may involve conducting a remote calibration of the system or running diagnostic investigations to determine whether the issues may originate from malfunctioning hardware or software. The maintenance personnel may decide to terminate operation of the conveyor belt system 12. In cases where the difference is above the threshold, control is provided (as shown at 46) and to the conveyor belt system 12 by means of the PLC 30. In this situation, the difference is severe enough to warrant immediate stoppage of the conveyor belt system 12, to allow maintenance personnel to address issues with the belt scale arrangement 20 or to recalibrate same.

[0095] By utilising two independent data feeds, the system 10 has built in redundancy. The likelihood of both the belt scale arrangement 20 and the calculation model system 32 failing simultaneously and to the same degree is considered negligible or at least acceptably small.

[0096] The system 10 therefore has the potential to provide increased accuracy, in that the system has the potential to reduce the likelihood of erroneous readings by cross-verifying data. The system furthermore has the ability to identify discrepancies early, allowing for prompt corrective action. The system 10 enhances the overall reliability of the measurement system and provides a backup in case one measurement system fails. Furthermore, it is believed that the system 10 could result in a cost reduction by eliminating the need for regular costly calibrations, and instead, providing a mechanism with which calibrations only need to be conducted when the system 10 calls for same.

[0097] By implementing this dual-feed approach, systems can achieve higher accuracy and reliability, which is essential in this critical application as small discrepancies in weight can amount to large variances in values in this application - measuring high tonnage of bulk material (valuable commodities) over an extended period. The ability to monitor and cross-verify the two feeds on a continuous basis and detect deviations automatically is believed to provide a significant improvement over existing systems.

[0098] Independent Load Cell Monitoring:

[0099] The invention further includes the capability to monitor each load cell independently. This is discussed in more detail below with reference to the load cell signal monitoring and processing unit 100. By continuously tracking each load cell, the system can identify specific inaccuracies or deviations within individual cells. If deviations exceed acceptable tolerances, the system can automatically flag these cells for investigation, maintenance, or recalibration. This ensures precision in the weight measurements, even when individual load cells malfunction or degrade over time.

[0100] Weight Distribution Per Load Cell:

[0101] Additionally, the system includes a feature for measuring weight distribution per load cell. This is discussed in more detail below with reference to the load cell signal monitoring and processing unit 100. This feature enables the system to detect lopsided or uneven material loading on the conveyor belt, which may cause measurement inaccuracies. By identifying uneven loading patterns, the system can notify operators and recommend adjustments to ensure proper weight distribution, improving the overall accuracy of the mass flow measurement.

[0102] Aggregation of Load Cell Readings and Compensation for Faulty Load Cells: A critical improvement over prior systems is the system’s ability to aggregate load cell readings and compensate for faulty cells. When one or more load cells fail or provide inaccurate readings, the system uses artificial intelligence and machine learning algorithms to calculate an accurate mass flow based on the remaining functional cells. By compensating for malfunctioning cells, the system can continue to provide reliable mass flow data, thus minimizing downtime and preventing the need for immediate maintenance intervention.

[0103] The use of the system 10 as detailed above is primarily aimed at identifying longer term or steady state inaccuracy caused by system drift.

[0104] In some implementations, or as a standalone aspect of the invention, the system may include or comprise a mass-flow monitoring or computing system, which, in addition to the belt scale of the known kind and the data driven calculation model, includes a calculation module which receives the first and second data feeds operatively in real time.

[0105] In some cases, this calculation module may be incorporated with the comparator module. Alternatively, the calculation module may be incorporated with or formed by the PLC 30, the calculation model system 32 or a backend or cloud-based calculation arrangement.

[0106] The calculation module processes the first and second data feeds into a third data feed and provides the third data feed as an output.

[0107] The calculation module creates the third data feed as a normalised computational mass flow rate. This normalised computational mass flow rate may be displayed and captured in parallel with the mass flow rates by any of the other capturing systems.

[0108] More particularly, the calculation module aggregates the first and second data flows and creates a normalised, compensated, composite, processed, aggregated, adjusted or smoothed data feed.

[0109] The calculation module therefore receives both the first and second data feeds and applies mathematical or statistical methodologies, filtering techniques, or machine learning algorithms to the first and second data feeds. In this way, the calculation module may create an output which is less susceptible to transient instability and fluctuations of either one of the first and second data feeds and which is more likely to represent an accurate reflection of an actual mass flow rate of the conveyor system.

[0110] Furthermore, by utilising machine learning algorithms, specific trends and causes for deviations may be identified, which may again be used to ensure accuracy of the measurements obtained by the measuring systems and which may indicate the need for preventative or curative maintenance.

[0111] Also, the third output, and the intelligence provided by the machine learning algorithms may be used to verify, expand upon or supplement the input obtained by the comparator module (if used simultaneously).

[0112] The load cell signal monitoring and processing unit 100 is shown in figure 3. It comprises a plurality of load cell input terminals 102. As shown in figure 3, the unit 100 may in some example cases comprise four input terminals 102. That said, examples containing more or fewer input terminals 102 are feasible. Each of these terminals are connected to a distinct load cell in use, such that a load cell input signal is received from the respective load cell via the load cell input terminal. The unit 100 further comprises a processing unit associated with a memory arrangement (not shown). Signals are transmitted to the processing unit via the terminals 102. The processing unit receives the load cell input signals and processes the load cell input signals into one or more processed output signals. At least a first output terminal 104 is provided for transmitting the output signals. As discussed before, the output terminal 104 may provide a (first) data feed to the system 10.

[0113] The unit 100 comprises at least one further input terminal 108 which may be connected to a tachometer (speed sensor).

[0114] The unit 100 furthermore comprises a display 114 and input keys 116 with which inputs may be provided to the unit. In use, the display 114 displays values associated with the signals received via the input terminals 102. The values received from the load cells may be expressed in mV. By displaying these values simultaneously and in real time, a technician may very easily and conveniently install, calibrate and commission the load cells of the a scale 20. These values may furthermore be provided as part of the output provided by the unit, to enable storing and further processing if necessary. The unit 100 also has the capability of measuring, recording and / or displaying measurements relating to the tachometer, load cell excitation voltage and the like. A tachometer output 110 and load cell output 112 are furthermore provided. The unit 100 is typically arranged near the scale 20 and may therefore be exposed to adverse environmental conditions. As such, the unit 100 is typically housed in a weatherproof housing 118.

[0115] The unit 100 is furthermore configured to be used in at least two use configurations: a normal output configuration, and a fault identification configuration. The unit 100 may simultaneously be configured in both of these configurations.

[0116] When configured in the fault identification configuration, the processing unit is configured for identifying a fault condition. In this configuration, the processing unit compares input values obtained from the various load cells to a predefined threshold range. A user may set the predefined threshold range before use of the unit 100. The threshold range may typically be a range expressed in mV in which the load cells are operating normally. A first fault condition may be identified in respect of a specific load cell when the value associated with the load cell falls outside of the predefined threshold range.

[0117] Another implementation of the fault identification configuration relates to the identification of differences in load cell values. This could point to load cell failure or load imbalance. Here, the processing module compares the values of each load cell to that of the others (rather than a predefined range) to identify differences or deviations therebetween. A predefined acceptable or allowable difference or deviation is provided to the unit. A second fault condition is identified in respect of a specific load cell if the difference between it and the other load cells exceeds the predefined acceptable or allowable difference or deviation.

[0118] An output is provided when a fault condition is identified in respect of one of the load cells. The output can be provided via the first output terminal 104 to a PLC or other backend operating system of the conveyor. Alternatively, the fault output is provided to a second output terminal 106 which is a solid-state relay output. This output 106 may sound an alarm when a fault condition is present and persisting.

[0119] When the unit 100 is configured in the normal output configuration, the processing unit calculates an average value from the values obtained from the load cell input signals and provides the calculated average value as the processed output signal and transmit the average value to the belt scale integrator.

[0120] The unit 100 is configured such that the average value may be based on a selection of values from the various input signals. This allows the normal output configuration to be used in cases where fault conditions had been identified. The selection may be a manual selection - in other words, a user may disable some load cell values from being used in the calculation of the average (for example, when such a load cell is faulty resulting in a skewed average). Alternatively, the system may automatically disregard values obtained from a load cell in respect of which a fault condition had been identified. In this way, the unit 100 may ensure that the data provided as output during a normal output configuration, is as accurate as possible.

[0121] It will be appreciated that the use of the unit provides the distinct advantage of being able to monitor values obtained from load cells before being integrated. This allows and facilitates easy calibration and commissioning, early identification of faults, prevention of inaccurate readings and planning of preventative maintenance. Furthermore, by facilitating the isolation or removal of values associated with faulty load cells, the conveyor and its scale may continue operating. This creates additional redundancy in the system, not previously possible. Furthermore, by facilitating monitoring of imbalanced loads, preventative measures may be taken, thereby improving accuracy and reducing the likelihood of damage to the belt and associated hardware.

[0122] It will be appreciated that the unit 100 may be incorporated with a belt scale of the known kind. The unit 100 may advantageously be used with the system 10 disclosed hereinbefore. Such a combination is believed to be particularly beneficial and may result in particularly accurate results.

[0123] In one example implementation, the system is installed on a troughed belt conveyor. The system continuously monitors individual load cells for deviations, and the calculation module calculates a normalized computational mass flow output by aggregating data from both sources.

[0124] In another example implementation, the system is used in a flat belt conveyor for food processing, where contamination prevention is critical. Here the system is designed to monitor for even weight distribution across the belt, ensuring that the material is loaded uniformly, which is vital for maintaining hygiene standards and preventing material buildup that could lead to contamination.

[0125] In yet a further example implementation, the system is used on conveyor systems in industries like pharmaceuticals, where precision is crucial. By adjusting load cell calibrations and optimizing belt speed autonomously, the system ensures continuous accuracy reducing the need for manual recalibration.

[0126] It will be appreciated that the third output represents a more accurate and reliable measurement by combining the strengths of both the first and second data feeds by smoothing out anomalies, temporary inaccuracies, noise, spikes and drops in the data. In cases where these deficiencies in the one of the feeds would cause a significant impact on the accuracy of the measurement, the normalised feed would allow continued reliable operation by mitigating the effect of the deficiency. By combining the first and second data feeds, the system improves the accuracy of mass flow measurements, reducing the impact of errors or inaccuracies present in a single data feed.

[0127] It is believed that the use of the system could lead to better inventory management, billing accuracy, and compliance with regulations due to more precise measurements. Furthermore, by smoothing out anomalies and providing a reliable measurement, the system could potentially further reduce the need for frequent calibrations or maintenance interventions.

[0128] A further aspect of the system involves its ability to learn from historical operational data. The system continuously analyses past performance and deviations, allowing it to develop a knowledge base of recurring deviations that occur under specific conditions, such as during conveyor startup or shutdown. By accounting for these deviations, the system can adjust its tolerance levels dynamically, preventing false alarms and ensuring more accurate mass flow measurements over time.

[0129] The system combines real-time measurements with data-driven calculations to produce a superior output.

[0130] The system leverages historical operational data to refine tolerance levels and set thresholds dynamically over time, thus enabling more accurate detection of sustained deviations or gradual changes before they become critical.

[0131] The system is capable of detecting and tracking deviations in real-time and adjusting system tolerances based on historical patterns, external factors (e.g., environmental conditions), or prior operational data. This feature allows operators proactively to adjust system parameters to avoid critical measurement errors.

[0132] It will be appreciated that the above description only provides example embodiments of the invention and that there may be many variations without departing from the spirit and / or the scope of the invention.

[0133] It will be easily understood from the present description that the particular features of the present invention, as generally described and illustrated in the figures, can be arranged and designed according to a wide variety of different configurations. In this way, the description of the present invention and the related figures are not provided to limit the scope of the invention but simply represent selected embodiments.

[0134] The skilled person will understand that the technical characteristics of a given embodiment can in fact be combined with characteristics of another embodiment, unless otherwise expressed or it is evident that these characteristics are incompatible. Also, the technical characteristics described one embodiment can be isolated from the other characteristics of this embodiment unless otherwise expressed.

Claims

CLAIMS1) A belt scale monitoring and / or computing system, comprising: a conveyor belt scale of the known kind operatively installed relative to a conveyor belt system and providing a first data feed; a data driven calculation model system operatively installed relative to the conveyor belt system and providing a second data feed; and a computational module operatively receiving in real-time the first and second data feeds and configured to process the first and second data feeds.2) The system according to claim 1 , wherein the first data feed relates to a real time measurement of one of weight and mass flow rate of material operatively carried by the conveyor belt.3) The system according to claim 1 , wherein the second data feed relates to a data driven inferred calculation of weight, mass or mass flow rate.4) The system according to claim 1 , wherein: the computational module comprises a processing module; the processing module is associated with a memory arrangement; and the computational module is configured for executing an algorithm in accordance with a predetermined program code.5) The system according to claim 4, wherein the processing module is one of: i) integrated with a programmable logic controller (PLC) of the conveyor belt; and ii) a remote processing module associated with a backend, which is provided in data-transfer communication with the conveyor belt scale.6) The system according to claim 1 , wherein the computational module comprises a comparator, wherein the system is configured for monitoring accuracy of the belt scale, wherein the comparator is configured to determine discrepancies or differences between the first and second data feeds and to classify the discrepancies or differences relative to an acceptable tolerance level, and wherein the comparator is configured for identifying an event when the discrepancy or difference exceeds the acceptable tolerance level.7) The system according to claim 6, wherein the comparator is configured for providing an output when the event is identified, and wherein the output is one of an alert communicated externally and an adjustment effected on a control system of the conveyor belt.8) The system according to claim 1 , configured as a mass-flow monitoring and computing system, wherein the computational module operatively processes the first and second data feeds into a third data feed, which is provided as an output feed of the system.9) The system according to claim 8, wherein the third data feed relates to one of a normalised, compensated, composite, processed, aggregated, adjusted and smoothed computational mass flow rate.10) A method of monitoring, measuring and / or processing data relating to a quantity of material carried by a conveyor belt system in real time, the method comprising the steps of:51) receiving a first data feed from a conveyor belt scale of the known kind installed relative to or as part of the conveyor belt system;52) receiving a second data feed from a data driven calculation model system operatively installed relative to the conveyor belt system;53) continuously processing the first and second feeds utilising a processing module; and54) providing an output based on the processing of the first and second feeds.11) The method according to claim 10, wherein the method relates to monitoring an accuracy of the belt scale, wherein the processing module comprises a comparator module, wherein the processing comprises utilising the comparator module to compare the first and second feeds, and wherein step S3 is preceded by the step of establishing an acceptable tolerance level for discrepancies between values obtained from the first and second feeds.12) The method according to claim 11 , wherein step S3 comprises the sub-steps of: calculating a difference between values obtained from the first and second feeds; and identifying an event when the difference between the values obtained from the first and second feeds exceeds the acceptable tolerance level.13) The method according to claim 12, wherein step S4 comprises providing an output when an event is identified and wherein the output is one of: i) an alert; and ii) a control provided to a control system of the conveyor belt.14) The method according to claim 13, wherein the control causes the belt to be stopped.15) The method according to claim 10, wherein the method relates to monitoring and / or computing mass flow rate of the conveyor, wherein step S3 comprises continuously processing the first and second data feeds into a third data feed and wherein step S4 comprises providing the third data feed as an output.16) The method according to claim 15, wherein step S3 comprises processing the third data feed into a normalised computational mass flow rate by aggregating the first and second data feeds.17) The method according to claim 16, wherein aggregating of the first and second data feeds occurs by means of one of statistical methods; filtering techniques; and machine learning algorithms that identify and mitigate anomalies.18) A load cell signal monitoring and processing unit, comprising: a plurality of load cell input terminals each for operatively receiving a load cell input signal from one of a plurality of load cells; a processing unit associated with a memory arrangement, the processing unit provided in data-flow communication with the plurality of input terminals and configured foroperatively receiving the load cell input signals and to process the load cell input signals into one or more processed output signals; an output terminal configured to transmit the one or more processed output signals.19) The load cell signal monitoring and processing unit according to claim 18, comprising at least one further input terminal for operatively receiving at least one further input signal from at least one further sensor device, each further input terminal provided in data-flow communication with the processing unit.20) The load cell signal monitoring and processing unit according to claim 19, wherein the further sensor device comprises a speed sensor device and wherein the further input signal is a speed sensor input signal.21) The load cell signal monitoring and processing unit according to claim 18, comprising between two and ten load cell input terminals.22) The load cell signal monitoring and processing unit according to claim 18, comprising four load cell input terminals.23) The load cell signal monitoring and processing unit according to claim 18, wherein the processing unit is configured for identifying a fault condition.24) The load cell signal monitoring and processing unit according to claim 23, wherein the processing unit is configured for: comparing a value associated with each load cell input signal to a predefined threshold range; and identifying a first fault condition when the value of one or more load cell input signal falls outside of the predefined threshold range.25) The load cell signal monitoring and processing unit according to claim 24, wherein the first fault condition relates to load cell failure.26) The load cell signal monitoring and processing unit according to claim 23, wherein the processing unit is configured for: comparing values associated with the complement of load cell input signals to identify differences between the values; comparing each identified difference to a predefined allowable difference value; and for identifying a second fault condition when an identified difference exceeds the predefined allowable difference value.27) The load cell signal monitoring and processing unit according to claim 26, wherein the second fault condition relates to load cell imbalance.28) The load cell signal monitoring and processing unit according to claim 23, wherein a predefined threshold range and a predefined allowable difference value are stored on the memory arrangement.29) The load cell signal monitoring and processing unit according to claim 23, wherein the processing unit provides a fault output when an identified fault condition is identified, the fault output provided via one of the output terminal and a separate fault output terminal.30) The load cell signal monitoring and processing unit according to claim 29, wherein the separate fault output terminal comprises a solid state relay output.31) The load cell signal monitoring and processing unit according to claim 29, further comprising an alarming means for providing an alarm when a fault output is provided by the processing unit.32) The load cell signal monitoring and processing unit according to claim 18, wherein the processing unit is configured for: calculating an average value from the values obtained from the load cell input signals; and providing the calculated average value as the processed output signal.33) The load cell signal monitoring and processing unit according to claim 32, wherein the processing unit is configured for calculating the average value based on a selection of values from a selection of input signals.34) The load cell signal monitoring and processing unit according to claim 33, including means for overriding one or more load cells manually, and wherein the values associated with the load cell input signal(s) of the overridden load cells are disregarded when calculating the average value.35) The load cell signal monitoring and processing unit according to claim 24 and 33, wherein the processing unit is configured for, when calculating the average value, isolating and / or removing values associated with an input signal in respect of which a fault condition has been identified.36) The load cell signal monitoring and processing unit according to claim 18, further including a display for displaying values associated with at least some of: i) inputs received via the input terminals; and ii) the one or more processed output signals37) A method of monitoring and processing load cell input signals, the method comprising the steps of:M1) receiving a plurality of load cell input signals via a plurality of load cell input terminals;M2) utilising a processing unit to receive the load cell input signals and to process the load cell input signals into one or more processed output signals;M3) providing the processed output signal to an output terminal.38) The method according to claim 37, comprising the further step Mx of utilising the processing unit to identify a fault condition.39) The method according to claim 38, wherein step Mx comprises at least some of the following sub-steps:Mx1) comparing a value associated with each load cell input signal to a predefined threshold range;Mx2) identifying a first fault condition when the value of one or more load cell input signal falls outside of the predefined threshold range;Mx3) comparing values associated with the complement of load cell input signals to identify differences between the values;Mx4) comparing each identified difference to a predefined allowable difference value;Mx5) identifying a second fault condition when an identified difference exceeds the predefined allowable difference value;Mx6) providing a fault output when an identified fault condition is identified; andMx7) utilising alarming means for providing an alarm when a fault condition is identified.40) The method according to claim 37, wherein step M2 comprises the sub-steps of:M2.1) calculating an average value from values obtained from the load cell input signals; andM2.2) providing the calculated average value as the processed output signal.41) The method according to claim 40, wherein step M2 comprises the sub-step of making a selection of input signals and wherein the average value is calculated considering the selection.42) The method according to claim 41 , wherein step M2 comprises the sub-step of manually overriding one or more load cells and wherein the average value is calculated by disregarding values associated with overridden load cells.43) The method according to claims 38 and 40, wherein step M2 comprises the sub-step of isolating and / or removing values associated with an input signal in respect of which a fault condition has been identified when calculating the average value.44) A conveyor belt scale comprising: a scale subframe; at least a first weigh idler set supported by an idler subframe mounted to the scale subframe by means of a first pair of load cells; and a load cell signal monitoring and processing unit according to claim 18, wherein each load cell is connected to a load cell input terminal of the load cell signal monitoring and processing unit.45) The system according to claim 1 , wherein the first data feed comprises a processed output signal obtained from a load cell signal monitoring and processing unit according to claim 18.46) A method according to claim 10, wherein the first data feed comprises a processed output signal obtained in accordance with a method of monitoring and processing load cell input signals according to claim 37.

Images