System and method of determining occurrence of an event associated with an excavator wear member

Abstract

A method of detecting an event associated with a wear component comprising obtaining a reference acceleration of a reference component of an excavator to which the wear component is connected, determining an expected acceleration of the wear component from the reference acceleration based on a relationship between acceleration experienced by the reference component and acceleration experienced by the wear component, and comparing the expected acceleration to a corresponding measured acceleration of the wear component in order to determine whether an event associated with the wear component has occurred.

Description

SYSTEM AND METHOD OF DETERMINING OCCURRENCE OF AN EVENT ASSOCIATED WITH AN EXCAVATOR WEAR MEMBERTechnical Field

[0001] The present invention relates generally to monitoring systems and, more particularly, to monitoring systems for determining the occurrence of an event associated with an excavator wear member. Although the present invention will be described with particular reference to determining events associated with ground engaging tool components on mining or earthmoving machinery, it will be appreciated that the invention is not necessarily limited to this application.Background

[0002] During operation, all earthmoving or excavating mining machinery is subject to heavy impact and abrasion on the surfaces of the machinery that are in direct contact with the ground. To avoid having to constantly replace or refurbish the entire machinery, earthmoving or excavating mining machinery (such as, for example, face shovels, draglines, front end loaders and excavators) is usually fitted with replaceable Ground Engaging Tools (GETs) that are designed to absorb most of this direct impact on the parts of the machinery that are subject to the most amount of wear (typically at or around the digging edge of the bucket on earthmoving or excavating mining machinery). This way, the majority of the wear on the machinery occurs on the GETs and so the earthmoving or excavating mining machinery can be easily refurbished by simply replacing the GETs.

[0003] There are a number of types of GETs that are used on the buckets of earthmoving or excavating mining machinery. A first type of GET is designed to break up the ground and typically takes the form of a series of pointed protrusions or 'teeth' that extend out from the digging edges of the earthmoving or excavating mining machinery. In some earthmoving or excavating mining machinery, the teeth are directly welded onto or cast as part of the digging edge of the bucket. In other systems, thesetypes of GETs are mechanically attached to the digging edge of the bucket, often through the use of another component commonly known as an 'adapter'. As these GETs are directly responsible for digging into and breaking up the ground, they are subject to much greater impact (and therefore, much greater wear) than the parts of the digging edge in between the teeth.

[0004] A second type of GET, more commonly known as a 'shroud', protects the digging edge of the bucket between the teeth of the earthmoving or excavating mining machinery. Like adapters and teeth, shrouds are either directly welded or cast onto the digging edge, or otherwise mechanically attached to the bucket lip. As shrouds are not directly responsible for digging into the ground, they are subject to less wearing than teeth.

[0005] During excavating and earthmoving operations, the wear components on GET hardware (especially the teeth) experience gradual wearing and require periodic replacement in order to maintain efficient digging operations and to protect the bucket (and adapters, if used) from damage. As the GET wear components wear down, the penetration of the cutting edge reduces and the energy required to dig the same amount of material increases.

[0006] In other instances, the teeth and adapters can break and fall off during the dig and load cycle of the excavating operation and, as a result, "contaminate" the ore. GET components are typically made of hardened alloy steel and can weigh up to hundreds of kilograms, making them one of the worst tramp metal hazards in a mining operation, particularly in the downstream processing operations. They have the potential to create significant work place hazards, which can result in significant production losses through equipment damage, plant downtime and / or ore wastage.

[0007] Typically, if a breakage of a GET component is detected, the earthmoving or excavating mining machinery (including the associated haulage trucks) immediately cease production and the digging face is inspected. If the missing GET component (orfragment of that GET) cannot be readily found, several scoops of ore (in the context of a mining operation) are removed from the suspected location and all outbound haulage trucks carrying ore are re-routed to dump their loads in a "quarantine" stockpile area.

[0008] However, if a broken GET component is not detected or found within a relatively short period of time after breakage, there is a more significant risk that the broken GET or GET fragment may be delivered to the ore crusher, which is not designed to process such hard materials and which will commonly suffer significant (often catastrophic) mechanical damage if it attempts to process (i.e. crush) the GET or GET fragment. For example, one of the GET teeth, if broken free from the bucket of the earthmoving or excavating mining machinery, has the potential to jam the crusher causing severe damage and putting the crusher out of service and operation for hours or days at a time (depending on the degree of mechanical damage).

[0009] Further, the process of removing a jammed GET tooth or broken GET fragment from a crusher is a very dangerous procedure that can result in human injuries or even fatalities if not performed properly. A GET tooth or fragment that inadvertently enters a crusher also has the potential to be projected out at great speed due to the significant mechanical forces applied to it by, for example, the jaws of the crusher, which in turn poses significant dangers for nearby personnel and equipment.

[0010] An earthmoving or excavating mining machine that continues to operate with missing, or broken, GET wear components can significantly increase the risk of breakages or accelerated wearing / damage to other parts of the machine (for example, the bucket lip of a shovel) resulting in expensive equipment repairs and extended downtime.

[0011] Broken or detached GET components are a serious safety issue, and when combined with energy wastage, production losses, and equipment damage, they represent a significant operational cost to the global mining industry every year. The total cost of this problem to the global mining industry could be measured in billions ofdollars per annum, when considering both direct and indirect costs.

[0012] It is against this background that the present invention has been developed.

[0013] In this specification where a document, act or item of knowledge is referred to or discussed, this reference or discussion is not an admission that the document, act or item of knowledge or any combination thereof was at the priority date, publicly available, known to the public, part of the common general knowledge; or known to be relevant to an attempt to solve any problem with which this specification is concerned.

[0014] Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.Summary

[0015] The present disclosure relates to a method of detecting an event associated with a wear component, the method comprising:a. obtaining a reference acceleration of a reference component of an excavator to which the wear component is connected;b. determining an expected acceleration of the wear component from the reference acceleration based on a relationship between acceleration experienced by the reference component and acceleration experienced by the wear component; andc. comparing the expected acceleration to a corresponding measured acceleration of the wear component in order to determine whether an event associated with the wear component has occurred.

[0016] In an embodiment, the event is detachment or partial detachment of the wear component.

[0017] In an embodiment, obtaining a reference acceleration comprises obtaining a reference acceleration vector.

[0018] In an embodiment, determining an expected acceleration of the wear component from the reference acceleration based on a relationship between acceleration experienced by the reference component and acceleration experienced by the wear component, comprises calculating a dot product of a wear component vector and the reference acceleration vector.

[0019] In an embodiment, obtaining a reference acceleration comprises obtaining a reference acceleration from at least one accelerometer mounted to the reference component.

[0020] In an embodiment, obtaining a reference acceleration from at least one accelerometer comprises, obtaining reference accelerations from two accelerometers mounted to the reference component.

[0021] In an embodiment, the method further comprises transforming a first of the reference accelerations to be in a same reference frame as a second of the reference accelerations.

[0022] In an embodiment, the method further comprises averaging the first and second accelerations and using the averaged acceleration to determine the expected acceleration.

[0023] In an embodiment, the method further comprises comprising determining the relationship between wear component acceleration and reference component acceleration by:a. setting up a cost function comprising candidate wear component unit vectors for a plurality of data points, and differences between (a) an acceleration calculated from the candidate unit vector and a corresponding bucket acceleration; and (b) a corresponding measured wear component acceleration; andb. using a numerical solver to find a wear component vector that minimises cost across the plurality of data points.

[0024] There is also disclosed method of taking an action in respect of a wear component, comprising detecting an event using the above method, and initiating an action in response to the event, wherein the action comprises at least one of generating an alarm, generating a warning message, and stopping at least part of a machine.

[0025] There is also disclosed a monitoring system for detecting an event associated with a wear component, the monitoring system comprising:a. a wear component sensor for obtaining a measured wear component acceleration;b. a reference sensor for obtaining a reference acceleration of a reference component of an excavator to which the wear component is connected; andc. at least one processor for:i. determining an expected acceleration of the wear component from the reference acceleration based on a relationship between acceleration experienced by the reference component and acceleration experienced by the wear component; andii. comparing the expected acceleration to a corresponding measured acceleration of the wear component in order to determine whether an event associated with the wear component has occurred.

[0026] In an embodiment, the event is detachment or partial detachment of the wear component.

[0027] In an embodiment, the processor uses a reference acceleration vector obtained from the reference sensor.

[0028] In an embodiment, the processor determines the expected acceleration of the wear component by calculating a dot product of a wear component vector and the reference acceleration vector.

[0029] In an embodiment, the monitoring system comprises another reference sensor mounted to the reference component for obtaining another reference acceleration.

[0030] In an embodiment, the processor transforms a first of the reference accelerations to be in a same reference frame as a second of the reference accelerations.

[0031] In an embodiment, the processor averages the first and second accelerations and using the averaged acceleration to determine the expected acceleration of the wear component.

[0032] In an embodiment, the processor determines the relationship between wear component acceleration and reference component acceleration by:a. using a cost function comprising candidate wear component unit vectors for a plurality of data points, and differences between (a) an acceleration calculated from the candidate unit vector and a corresponding bucket acceleration; and (b) a corresponding measured wear component acceleration; andb. using a numerical solver to find a wear component vector that minimises cost across the plurality of data points.

[0033] In an embodiment, the monitoring system comprises a plurality of wear component sensors corresponding to respective wear components, and wherein the processor determines an expected acceleration for each wear component and compares, for each wear component, a corresponding measured acceleration of the wear component in order to determine whether an event associated with the respective wear component has occurred

[0034] In an embodiment, the processor initiates an action in response to the event, wherein the action comprises at least one of generating an alarm, generating a warning message, and stopping at least part of a machine.Brief Description of Drawings

[0035] Embodiments of the present invention will now be described with reference to the accompanying drawings. These embodiments are given by way of illustration only and other embodiments of the invention are also possible. In the drawings:

[0036] Figure 1 is a schematic diagram illustrating a monitoring system in accordance with a representative embodiment of the present disclosure;

[0037] Figure 2 is a partial view of a sensor assembly in accordance with a representative embodiment of the present disclosure;

[0038] Figure 3 is partial view of a sensor assembly, showing a cup of a cylindrical housing, in accordance with a representative embodiment of the present disclosure;

[0039] Figure 4 is a cross-sectional side elevation of a tooth for a ground engaging tool and a sensor assembly prior to the sensor assembly being secured to the tooth;

[0040] Figure 5 is a cross-sectional side elevation of the tooth depicted in Figure 4 after securing the sensor assembly to the tooth;

[0041] Figure 6 is an end elevation of the tooth and sensor assembly depicted in Figure 5;

[0042] Figure 7 is a front elevation of a front end loader machine that has a bucket on which is mounted a plurality of teeth of the type depicted in Figures 4, 5 and 6, and an sensor assembly reader for reading the sensor assemblies secured to the teeth;

[0043] Figure 8 is a schematic diagram depicting a machine that has a first alternative machine mounted sensor assembly reading station;

[0044] Figure 9 is a flow chart of a reference sensor calibration method;

[0045] Figure 10 illustrates the effect of reference sensor calibration;

[0046] Figure 11 is a flow chart of a wear sensor calibration;

[0047] Figure 12 illustrates the effect of wear sensor calibration;

[0048] Figure 13 is a flow chart of an event detection method; and

[0049] Figure 14 illustrates the effect of data filtering.Description of Embodiments

[0050] Representative embodiments of the present disclosure relate, generally, to various apparatus, methods, and systems for monitoring an event associated with a wear component and, more particularly, to monitoring systems for detecting an eventassociated with ground engaging tool components and tools used on, for example, earthmoving or excavating mining machinery. The disclosure has particular, but not necessarily exclusive, application to monitoring systems for detecting the detachment of ground engaging tool components from mining or earthmoving machinery. However, it should be understood that the disclosure is not limited to this representative embodiment, and may be implemented in other environments where similar earthmoving or excavation operations are conducted.

[0051] Embodiments of the invention use sensors to measure accelerations experienced by wear components and reference components to which the wear components are connected, for example, by being mounted to the reference components. Techniques are disclosed that enable a relationship to be established between accelerations experienced by the reference components and accelerations experienced by wear components connected to the reference components. The relationship between an individual wear component and at least one corresponding reference component can be used to determine an expected acceleration for the wear component. A comparison of the expected acceleration to a measured acceleration can be used to determine whether an event, such as detachment of the wear component from the reference component, has occurred. For example, on the basis of the difference between the measured acceleration and the expected acceleration being outside a defined tolerance.

[0052] Sensors are mounted to the wear and reference components to obtain sensor data and output that sensor data for processing. WO2021 / 026597, the disclosure of which is incorporated herein by reference, describes one solution for deploying sensors in a harsh environment. The solution described in WO2021 / 026597 allows sensed data detected by a sensor embedded in a metal object (e.g. a GET component such as a tooth or adapter) to broadcast / transmit the sensed data to a remote receiver by exploiting the principle of coupling. Coupling, or mutual coupling, is a Radio Frequency term referring to an undesirable condition in which a first antenna within close proximity to a second antenna absorbs the energy being broadcast by the second antenna, therebyreducing the performance of the first antenna. Techniques described in the WO2021 / 026597 allow the metal object (e.g. the GET component such as a tooth or adapter), in combination with a powered sensor and a metal disc antenna, to act as one mutually coupled and matched antenna for the transmission of sensed data to a remote receiver.

[0053] WO2021 / 026597 discusses, amongst other things, system parameters for delivering an impact resistant sensor system able to operate in difficult RF environments. For example, WO2021 / 026597 describes very small sensor systems that can be embedded within a metal object (e.g. a GET component such as a tooth or adapter) for the detection and broadcasting of physical and / or operational characteristics of the GET component (and the excavating or earthmoving machine that it is associated with). Small form factor, impact resistance, and transmission capabilities are particularly useful when visual detection of a physical characteristic, such as wear, is impractical.

[0054] Figure 1 is a schematic diagram of a monitoring system 30 for monitoring for an event associated with a GET wear component that has been adapted from the sensor system described in WO2021 / 026597.

[0055] Referring to Figure 1, a monitoring system 30 includes a plurality of sensor assemblies 31. Each sensor assembly 31 is mounted on a respective wear component 32 of a ground engaging tool (GET) of a mining or earthmoving machine 33 such that the sensor assembly 31 is secured to the wear component 32 or to a reference component 34 of the mining or earthmoving machine 33. The machine 33 may, for example, be a loader such as a front end loader, a shovel, or an excavator. Depending on the type of mining or earthmoving equipment, the wear component 32 may, for example, be a tooth, adapter, shroud, protective plate, or a lip of a bucket or scoop. The reference component 34 is a component to which the wear component 32 is connected so as to move with the reference component 34. Depending on what type of mining or earthmoving equipment, the reference component 34 may be, for example, a bucketor scoop. As described in WO2021 / 026597, a sensor system may be able to detect the material characteristics (including for example, the gradual wearing, or complete loss) of the component 32 from the machine 33. However, the present disclosure focuses on the monitoring system 30 operating to detect the loss of a wear component 32 from the machine 33.

[0056] According to a representative embodiment, as shown in Figures 2 and 3 of the drawings, the sensor assembly 31 includes a protective cylindrical housing 40 (or outer casing). The cylindrical housing 40 includes a cylindrical cup 41 (or outer casing top portion), and a circular lid / end plug 42 (or outer casing bottom portion) with a closed bottom end for covering an opening 43 in an end of the cylindrical cup 41 and enclosing the space within the cylindrical housing 40.

[0057] Lid 42 includes an outer portion 44 for resting on a rim 45 of the cup 41 which surrounds the opening 43, and an inner portion 46 for inserting into the opening 43 when enclosing the space within the cylindrical housing 40. When inserted into the opening 43, the fit between the inner portion 46 of the lid 42 and the cup 41 is preferably a press fit. Alternatively, the interface between the inner portion 46 of the lid 42 and the cup 41 may include mating threads (not shown) and an internal bevel (not shown) to create a seal between the lid 42 and cup 41.

[0058] Alternatively, or in addition, the interface between the lid 42 and the cup 41 is sealed with a sealant such as, for example, a silicone sealant to prevent the ingress of undesirable materials (e.g. dust, liquid) into the space within the cylindrical housing 40.

[0059] The cup 41 of the cylindrical housing 40 includes a cylindrical side wall 47 which defines the opening 43 at the end of the cup 41 as well as the rim 45 of the cup 41. The opposing end of the cup 41 is preferably closed and comprises a base 48 from which the side wall 47 extends.

[0060] In a representative embodiment of the present disclosure, both the cup 41 and the lid 42 are made from a plastic material, such as polyetherimide plastic, that issubstantially transparent to radio frequency electromagnetic signals. Synthetic polyetherimide polymers have numerous benefits in addition to their permissibility / transparency to RF, like their durability and manufacturing options like being able to be printed with a 3D printing device. Using a 3D printer with a suitable base material, like a synthetic polymer, may also allow the cup 41 and the lid 42 to be printed in a single step around the sensor assembly 31. A unified casing or shell made from a single bottom and top outer casing portion can be beneficial when additional water proofing is desired.

[0061] It should be appreciated that other materials, having similar properties (i.e. substantially transparent to radio frequency electromagnetic signals), may also be used and are envisaged within the scope of the present disclosure.

[0062] Alternatively, instead of 3D-printing the cup 41 and the lid 42, the protective housing 40 may be moulded, extruded or machined / turned to achieve the same overall structure.

[0063] In a representative embodiment, the cylindrical housing 40 may also include a silicone rubber layer 49 that is adhered or bonded to a bottom surface of the lid 42. The silicone rubber layer 49 may preferably serve to dampen impact forces (e.g. forces transferred through the wear component 32) on the sensor assembly 31, and particularly the sensor component 51, during operation of the machine 33.

[0064] The sensor assembly 31 also includes a battery 50 that is situated inside the lid 42. As illustrated in Figure 2 of the drawings, the battery 50 does not need to be fully contained within the lid 42 (or outer casing bottom portion), merely inside the circumference of the inner portion 46 of the lid 42 in order to allow the inner portion 46 to be inserted into the opening 43 of the cup 41. In a representative embodiment of the present disclosure, shown in Figure 2 of the drawings, the battery is a lithium cell battery, preferably a lithium cell coin battery, having a diameter that substantially meets an inside diameter of the lid 42 (or outer casing bottom portion).

[0065] The sensor assembly 31 further comprises a sensor component 51 that includes a circuit board 52 on which various electronic components 53 are mounted. The circuit board 52 is adapted to be powered by the battery 50 that is connected to the circuit board 52 and that is also contained within the cylindrical housing 40. The circuit board 52 may include an epoxy resin coating (not shown) for additional protection from dust, fluid, and / or impact during operation. The electronic components 53 mounted to the circuit board 52 include an inertial measurement unit (IMU). Example embodiments employ IMUs that have a combination of accelerometers and gyroscopes, however, other embodiments may employ IMUs that only incorporate an accelerometer. The electronic components 53 mounted to the circuit board 52 may also include a temperature sensor, a magnetometer, a capacitive sensor, a piezoelectric microphone, and / or a MEMS piezoelectric microphone. However, it should be appreciated that various combinations of one or more of these electronic components 53 are also envisaged by the present disclosure depending on the specific application of the monitoring system 30. In an example, a first type of sensor component may be used in a reference component and a second type of sensor component may be used in a wear component. In an example, only a one-axis accelerometer is needed in each wear component, however, in practice, it may be easier to source three-axis accelerometers. In an example, measurements may only be obtained from one-axis of a three-axis accelerometer. In another example two 3-axis accelerometer outputs may be compared. In another example, the process could work in reverse having a singleaxis accelerometer for reference and 3-axis accelerometer on the wear component.

[0066] The sensor assembly 31 further comprises a metal disc antenna 54 that is connected to the circuit board 52 of the sensor component 51, via a metal connector element 55, and positioned at a predetermined distance above the circuit board 52 of the sensor component 51. The metal disc antenna 54 can be made from a variety of metallic materials that have properties making them suitable for use as an RF antenna. In a representative embodiment of the present disclosure, the metal disc antenna 54 is made from a copper beryllium alloy. In an embodiment, the metal disc antenna 54 andthe metal connector element 55 are integrally formed from a single piece of metallic material. Such a configuration enables the metal disc antenna 54 to be positioned (and resiliently retained) at a predetermined distance above the circuit board 52 of the sensor component 51 without the need for additional apparatus.

[0067] The sensor assembly 31 further comprises a cushioning element 56 that is interposed between the battery 50 and the sensor component 51. The cushioning element 56 is adapted to dampen impact forces on the sensor assembly 31, and particularly the sensor component 51, during operation of the machine 33. In a representative embodiment of the present disclosure, the cushioning element 56 may be a low-density foam or similar impact dampening / adsorbing material. In a particularly preferred embodiment of the present disclosure, an adhesive (such as, for example, a silicone adhesive or bonding agent) may be used to bond the circuit board 52 of the sensor component 51 to the cushioning element 56, and / or to bond the cushioning element 56 to the battery 50. However, it should be appreciated that a variety of similar adhesives or bonding agents may be used based on the desired impact performance of the system 30.

[0068] It should be understood that when the cup 41 and lid 42 of the cylindrical housing 40 are brought into engagement (e.g. via a press fit, or other sealing mechanism), that the battery 50, sensor component 51, metal disc antenna 54, metal connector element 55, and cushioning element 56 are all enclosed within the cylindrical housing 40.

[0069] In a representative embodiment of the present disclosure, the sensor assembly 31 is adapted to fit into a recess 76 in a wear component 32. on machine 33. A ground engaging tool (GET) or wear component 32 which is in the form of a replaceable tooth / point 70 for a bucket or scoop is depicted in Figures 4, 5 and 6 of the drawings. Tooth 70 has a generally tapered profile and includes an upper side 71, a lower side 72, a leading end 73, and a trailing end 74. A cavity 75 for receiving a projection of anadaptor 82 (as shown in Figures 8 and 9 of the drawings) that is secured to the bucket or scoop extends into the tooth 70 from the trailing end 74.

[0070] A cylindrical recess / hole 76 is created in the tooth 70 at a base 77 of the cavity 75. The recess 76 may, for example, be created in the tooth 70 by casting, boring, drilling, or milling it into the tooth 70 which is made out of metal, typically high-strength steel. The diameter of the recess 76 is slightly larger than the outer diameter of the cylindrical housing 40 so that the housing 40 is able to be inserted into the recess 76. The depth of the recess 76 is such that the sensor assembly 31 is able to be inserted into the recess 76 such that the sensor assembly 31 (including the cylindrical housing 40) does not protrude from the recess 76.

[0071] An adhesive agent such as, for example, a silicone sealant which is located between the bottom of the recess 76 and the inner end of the cylindrical housing 40 which includes the lid 44, secures the sensor assembly 31 to the tooth 70 so that the sensor assembly 31 is retained in place relative to the tooth 70. This adhesive agent may be in addition to, or as an alternative to, the silicone rubber layer 49 that is adhered or bonded to a bottom surface of the lid 42. Inserting the sensor assembly 31 into the recess 76 in this manner assists in protecting the housing 40, and exposes the base 48 of the cup 41 to the wearface / base 77 of the recess 76 (proximate the adapter 82 when the tooth 70 is brought into engagement with the adapter 82).

[0072] Referring to Figures 7, a mining / earthmoving machine 33 in the form of a front end loader 80 includes a ground engaging tool in the form of a bucket 81. A plurality of adapters 82 are mounted on a bottom lip 83 of the bucket 81, and a respective tooth 70 is secured to each adapter 82 in the usual manner. Each adapter 82 includes a projection 84 that is inserted into the cavity 75 of a respective tooth 70 such that at the interface of each projection 84 and tooth 70 there is sufficient clearance between the projection 84 and the sensor assembly 31.

[0073] In some examples, once embedded within the recess 76 of the wear component 32, the sensor assembly 31 (including, particularly, the metal disc antenna 54) may be fine-tuned to use the surrounding metal (of the wear component 32) as an amplifier or at least an extension of that antenna 54. In some examples, the sensor assembly tuning may be calibrated beforehand using a standard wear component.

[0074] Preferred frequencies, and / or ranges for amplification, for the antenna 54 are ideally within the Ultra High Frequency (UHF) range, although it should be appreciated that other frequencies and frequency ranges may be preferred depending on the application and / or the type of wear components 32 within with the sensor assembly 31 is located.

[0075] Further details as to the positioning of the sensor assembly 31 within a recess / hole 76 are described in WO2021 / 026597. The person skilled in the art will appreciate that a similar recess may be formed in a reference component to receive a sensor assembly. In some examples, two or more sensor assemblies may be employed within a reference component and / or a wear component in order to provide an element of redundancy.

[0076] It will also be appreciated that the sensor assembly may be deployed in other machines that have reference components and wear components, including in machines with multiple reference components such as in drag-line excavators which have a plurality of buckets.

[0077] The monitoring system 30 further comprises a remote radio frequency receiver 90 operable to receive sensor data wirelessly from the sensor component 51, transmitted to the remote radio frequency receiver 90 via the metal disc antenna 54. In a representative embodiment of the present disclosure, the remote radio frequency receiver 90 is mounted on the front end loader 80. The remote radio frequency receiver 90 preferably includes an antenna (or plurality of antennas, not shown) that are mounted on a suitable position on the front end loader 80 such as, for example, the topof a cab 92 of the front end loader 80. The antenna (not shown) allows the remote radio frequency receiver 90 to communicate with the sensor assemblies 31. In particular, it allows the remote radio frequency receiver 90 to detect / read sensor assemblies 31 that are within the range of the remote radio frequency receiver 90.

[0078] Referring again to Figure 1 of the drawings, the remote radio frequency receiver 90 is connected to a Wi-Fi transceiver 93. The remote radio frequency receiver 90 and the transceiver 93 are connected to each other so that they can communicate with each other. The reader 90 is able to transmit data to the transceiver 93. For example, the reader 90 is able to transmit to the transceiver 93 sensor data which the reader 90 reads from the sensor assembly 31. A transceiver antenna 94 is connected to the transceiver 93 so that the transceiver 93 is able to communicate with a wireless communication network such as a Wi-Fi communication network 95 of the monitoring system 30. The transceiver 93 is able to transmit the data (e.g. sensor data of the sensor assembly 31) that is transmitted to it by the remote radio frequency receiver 90 to the network 95. This arrangement assumes different communication protocols are used to communicate with the sensor assemblies and upstream to other processing equipment. In other embodiments, a single transceiver may handle all communications.

[0079] If the machine 33 (e.g. front end loader 80) includes a plurality of wear components 32 (such as shown in Figure 7 of the drawings with the front end loader 80) that each includes their own sensor assembly 31, the remote radio frequency receiver 90 reads the data of each of the sensor assemblies 31 in the wear components 32 and the sensor assembly in the reference components 34.

[0080] An Ethernet switch 96 is preferably connected to the remote radio frequency receiver 90 and the transceiver 93. The remote radio frequency receiver 90 and the transceiver 93 are connected to the switch 96 such that they are able to communicate with each other through / via the switch 96. The transceiver 93 and the switch 96 are preferably part of a mining communication backbone. The remote radio frequency receiver 90, associated antenna (not shown), transceiver 93, transceiver antenna 94,and switch 96 function as a machine mounted sensor assembly reading station 97 of the monitoring system 30. The monitoring system 30 can include multiple machine mounted sensor assembly reading stations 97. For example, the monitoring system 30 can include multiple machine mounted sensor assembly reading stations 97, with each station 97 being mounted on a respective machine 33.

[0081] In an alternative embodiment of the present disclosure, the radio frequency receiver 90 may be configured with onboard computer processing capability (such as, for example, the embedded personal computer 160 shown in the drawings) such that it can directly process sensor data received wirelessly from the sensor component 51, transmitted to the remote radio frequency receiver 90 via the metal disc antenna 54.

[0082] In some examples, the monitoring system 30 may include a monitoring station 130 that includes a Wi-Fi transceiver 131, an antenna 132, and a server 133. The antenna 132 is connected to the transceiver 131 so that the transceiver 131 is able to communicate with the other transceivers 93 and therefore the readers 90 via network 95. For example, the transceiver 131 is able to receive from the transceivers 93 via the network 95 the sensor data which the readers 90 read from the sensor assemblies 31. The transceiver 131 is connected to the server 133 so that they are able to communicate with each other. The transceiver 131 is able to transmit the data (e.g. sensor data) that it receives from the transceivers 93 via the network 95 to the server 133 so that the server 133 can then process the data.

[0083] Server 133 includes a processor 134, memory 135, and a database 136. Software which is stored on the memory 135 is run on the processor 134 of the server 133, which is a central server. The server 133 communicates with the readers 90 via the wireless network 95 and stores all data in the database 136.

[0084] The server 133 is able to generate alarm messages / issue alerts which can be communicated to users via a number of different methods, and the system in general or the server 133 in particular interfaces to existing mine management software usinga data communication link. For example, if the system 30 via the server 133 detects that a tooth 70 to which a sensor assembly 31 is secured has fallen off the machine 33, this will generate an alarm message which will then be communicated to a user (e.g. the operator of the machine 33) by a suitable method (e.g. by radio) so that the user or someone else can take appropriate action to prevent the tooth 70 from finding its way into the crusher. The server 133 can be a standalone physical machine, or a virtual server, for example, provided by the mine operator to utilise their existing infrastructure.

[0085] Referring to Figure 8, in an alternative, the machine mounted sensor assembly reading station 97 can be a stand-alone, rugged, embedded personal computer 160 that is mounted in a cab of the machine 33. Computer 160 is connected to the sensor assembly reader 90 via a data communication link 161 such that the computer 160 is able to communicate with the reader 90. The computer 160 functions in a similar manner to the server 133 in that it is able to process all of the information / data provided by the reader 90. However, unlike the server 133, the computer 160 is located locally with the reader 90. The computer 160 is able to alert / issue an alert to the operator of the machine 33 via local alarms / buzzers should the reader 90 detect the loss of a wear component 32 from the machine 33 or, in some examples, stop the machine or a component of the machine. In this way, the station 97 is able to act as an independent or self-contained monitoring system which does not need to communicate with the monitoring station 130 and therefore does not necessarily require the transceiver 93, antenna 94, and switch 96. However, the station 97 may still include the transceiver 93, antenna 94, and switch 96 so that the reader 90 is able to communicate with the server 133 via the computer 160.

[0086] This embedded computer option can provide a detection system for mines which do not have reliable Wi-Fi infrastructure to transmit the reader data across, or for mines that may want local processing of alarms on the machine 33 and also on the backbone server 133 to provide site-wide monitoring of multiple machines 33.

[0087] In an example embodiment where there are multiple sensor assemblies in a reference component, for example, two sensor assemblies each having an IMU, a relationship between acceleration experienced by reference components and wear components takes advantage there being two reference IMUs. To be able to make use of two or more reference IMUs, in this example first and second bucket IMUs, one IMU reference frame is transformed into another IMU reference based on a sample set of acceleration vectors. An example is described below for two reference IMUs, however the concept can be extended to more IMUs. It will also be appreciated that a single IMU can also be used. In this respect, it will be appreciated that two reference IMUs provides a balance between providing some redundancy and additional hardware cost.

[0088] In an example, a bucket calibration method 200 as shown in FIG. 9, involves obtaining 205 a plurality of data points such as a time series of data points for the IMUs of two buckets and then filtering 210 out any data points where either (i) one of the two bucket IMUs did not report or (ii) either bucket acceleration magnitude was not close to an expected value of gravity. In an example, a normalised acceleration is used and the expected value of gravity is 32.

[0089] At step 215, the filtered data is used in a cost function that takes in three parameters (pl, p2 and p3) representing roll, pitch and yaw angles to transform the second IMU signal. In this example, the cost function uses the roll, pitch and yaw angles to generate a 3D rotation matrix and applies it to all of the second IMU acceleration vector samples across a range of time and bucket poses. Each sample of the first IMU and the corresponding second transformed IMU vectors are then differenced and the square root of magnitude of the resultant vector added as an element of the cost output array. In this example, the square root is applied because it makes the cost function less biased towards outliers.>c[i] = VloHi] - R(pl,p2,p ) * a^[i]\

[0090] At step 220, the cost function and rotation angles are used as input parameters of a nonlinear multivariate numerical solver to optimize the angles to minimize the defined cost. In this context, a nonlinear solver refers to any mathematical procedure that takes a system defined by a set of input parameters and a resultant cost that can be calculated based on the input values and finds the optimum values of those parameters to minimize the output of the cost function. Common examples of numerical solvers are Steepest Descent and Newton-Raphson. In an embodiment, a Levenberg-Marquardt numerical solver was used which is a hybrid of steepest descent and Newton-Raphson, however, the skilled person will appreciate that other numerical solvers may be employed.

[0091] Once the transform is determined using the numerical solver, it is applied at step 225 to all of the second IMU accelerations. The transformed second IMU accelerations are then used in conjunction with the first IMU accelerations in later stages. In this respect, it will be appreciated that transforming the second IMU accelerations to the first IMU reference frame is an arbitrary selection and the transformation could alternatively be to the second IMU reference frame.

[0092] FIG. 10 shows one example of two bucket accelerometers being aligned using the process of FIG. 9. The top frame 1010 shows the original set of accelerometer values across the x-axis lOllx , y-axis lOlly and z-axis lOllz of the IMU for the first bucket and the original set of accelerometer values across the x-axis 1012x , y-axis 1012y and z-axis 1012z of the IMU for the second bucket. The bottom frame 1020 shows the same accelerometer values lOllx, lOlly, lOllz for the first bucket and the transformed accelerometer values across the x-axis 1022x , y-axis 1022y and z-axis 1022z for the second bucket after the above alignment algorithm is applied.

[0093] FIG. 11 show a calibration method 300 for determining relative orientations of each wear component accelerometer in the combined bucket frame using a plurality of data points, for example a time series of data points obtained at step 305.In an example, calibration is only needed when a new sensor is added. While not shown in FIG. 11, similarly to FIG. 9, some data points may be filtered out, for example where IMUs did not report or acceleration magnitude was not close to an expected value of gravity.

[0094] In an example, two angles in a 3D rotation transform are sufficient to cover all possible vector orientations since rotation about the vector itself doesn't change its orientation. In this example, at step 310 a cost function is set up using a standard 2-degree of freedom rotation of a unit vector (leave one angle equal to zero) to get candidate unit vectors. Within the cost function, for each candidate unit vector, a dot product is obtained with the corresponding bucket acceleration vector across all samples and the signed differences to the corresponding wear component acceleration are output to the cost array. In this respect, in an example an average bucket acceleration vector is used which is derived from the first bucket acceleration vector and the transformed second bucket acceleration vector.

[0095] At step 315, a numerical solver, such as a Levenberg-Marquardt solver, is used to find the two variable transform angles that minimize the cost across all samples and they are stored at step 320 as vectors in a data record associated with the respective wear component.

[0096] Once wear component vectors are obtained in this second calibration process, taking a dot product of the respective wear component unit vector with the corresponding bucket acceleration vector yields an expected linear acceleration value for the wear component that can be compared with the measured acceleration for the respective wear component.

[0097] FIG. 12 shows examples of expected accelerations calculated using vectors obtained using the above method being applied to actual data sets alongside. FIG. 12 shows the accelerometer data sets from nine wear component mounted IMUs. Thetop panel 1210 of FIG. 12 shows the alignment between the accelerations between four expected shroud accelerations and four measured shroud accelerations. The bottom panel 1220 of FIG. 12 shows the alignment between the accelerations between five expected tooth accelerations and five measured tooth accelerations. The results show good correlation between expected vs actual acceleration across a range of wear components and machine types, such as backhoes, face shovels, wheel loaders and rope shovels with bucket capacities ranging from 14 to upwards of 100 tonnes. Noise in the wear component accelerometer outputs may be reduced through filtering based on comparison to neighbouring samples in time.

[0098] Once the relative alignments of the wear components to the reference components are determined, as shown in FIG. 13, a method 400 of wear component event detection may be performed. In this example, the method includes obtaining 405 the bucket acceleration and wear component accelerations from the relevant sensors, and calculating 410 the expected wear component acceleration from the measured bucket acceleration using the relationship established above before comparing 415 the measured acceleration of the wear components to the expected acceleration.

[0099] In an embodiment, the bucket acceleration vector is determined by (a) where both bucket acceleration vectors are available, determining an average bucket acceleration vector from the first bucket acceleration vector and the transformed second bucket acceleration vector, or (b) where only one bucket acceleration is available, using the available bucket acceleration vector. If neither bucket acceleration vector is available, the process proceeds to the next data point.

[0100] In another example where there are two (or more) reference IMUs, one of the IMUs may be used as the default from which an acceleration vector is obtained and the other used as a backup.

[0101] In an example, taking a dot product of each wear component unit vector with the bucket acceleration vector yields an expected linear acceleration value. If the expected acceleration value deviates from the actual value recorded by the respective wear component accelerometer by more than a threshold, it can be flagged 420 as an event indicating the potential detachment of a wear component.

[0102] If at step 415, the expected acceleration is within an expected threshold of the measured acceleration, the process reverts to step 405 and processes the next sample bucket acceleration.

[0103] In some examples of the method 400, additional processing may be performed to remove false alarms from the potential errors so as not to trigger false alarms due to impact events while still maintaining good up-time on loss-detection. FIG. 14 illustrates such processing and shows the difference in expected vs actual acceleration against this range (normalized to 32 / lg on both axes) for a four different GET tags 1401-1404.

[0104] An example of further processing is removal of any wear component samples where the acceleration is more than a set threshold above lg. These can be seen as the lighter colour data points in Figure 14.

[0105] Another example of further processing is to omit any samples where the maximum range in values between one sample before and one sample after the current sample is greater than some threshold.

[0106] Most of the recorded data exists in the first band of 0-5 min / max range (x-axis). Accordingly, in one example, loss events are detected by filtering to only keep data within that range and alarming if the difference is more than half of one g (16 on the y-axis). In an example, this enables an alarm to be triggered in approximately 2 / 3 of possible bucket orientations that will be met at some time through a standard dig cycle.

[0107] As the present invention may be embodied in several forms without departing from the essential characteristics of the invention, it should be understood that the above described embodiments should not be considered to limit the present invention but rather should be construed broadly. Various modifications, improvements and equivalent arrangements will be readily apparent to those skilled in the art, and are intended to be included within the spirit and scope of the invention. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Claims

CLAIMS:

1. A method of detecting an event associated with a wear component, the method comprising:a. obtaining a reference acceleration of a reference component of an excavator to which the wear component is connected;b. determining an expected acceleration of the wear component from the reference acceleration based on a relationship between acceleration experienced by the reference component and acceleration experienced by the wear component; andc. comparing the expected acceleration to a corresponding measured acceleration of the wear component in order to determine whether an event associated with the wear component has occurred.

2. The method of claim 1, wherein the event is detachment or partial detachment of the wear component.

3. The method of claim 1 or claim 2, wherein obtaining a reference acceleration comprises obtaining a reference acceleration vector.

4. The method of claim 3, wherein determining an expected acceleration of the wear component from the reference acceleration based on a relationship between acceleration experienced by the reference component and acceleration experienced by the wear component, comprises calculating a dot product of a wear component vector and the reference acceleration vector.

5. The method of any one of claims 1 to 4, wherein obtaining a reference acceleration comprises obtaining a reference acceleration from at least one accelerometer mounted to the reference component.

6. The method of claim 5, wherein obtaining a reference acceleration from at leastone accelerometer comprises, obtaining reference accelerations from two accelerometers mounted to the reference component.

7. The method of claim 6, further comprising transforming a first of the reference accelerations to be in a same reference frame as a second of the reference accelerations.

8. The method of claim 7, further comprising averaging the first and second accelerations and using the averaged acceleration to determine the expected acceleration.

9. The method of any one of claims 1 to 6, further comprising determining the relationship between wear component acceleration and reference component acceleration by:setting up a cost function comprising candidate wear component unit vectors fora plurality of data points, and differences between (a) an acceleration calculated from the candidate unit vector and a corresponding bucket acceleration; and (b) a corresponding measured wear component acceleration; andusing a numerical solver to find a wear component vector that minimises cost across the plurality of data points.

10. A method of taking an action in respect of a wear component, comprising detecting an event using a method as claimed in any one of claims 1 to 9, and initiating an action in response to the event, wherein the action comprises at least one of generating an alarm, generating a warning message, and stopping at least part of a machine.

11. A monitoring system for detecting an event associated with a wear component, the monitoring system comprising:a wear component sensor for obtaining a measured wear component acceleration;a reference sensor for obtaining a reference acceleration of a reference component of an excavator to which the wear component is connected; andat least one processor for:determining an expected acceleration of the wear component from the reference acceleration based on a relationship between acceleration experienced by the reference component and acceleration experienced by the wear component; andcomparing the expected acceleration to a corresponding measured acceleration of the wear component in order to determine whether an event associated with the wear component has occurred.

12. The monitoring system of claim 11, wherein the event is detachment or partial detachment of the wear component.

13. The monitoring system of claim 11 or claim 12, wherein the processor uses a reference acceleration vector obtained from the reference sensor.

14. The monitoring system of any one of claims 11 to 13, wherein the processor determines the expected acceleration of the wear component by calculating a dot product of a wear component vector and the reference acceleration vector.

15. The monitoring system of any one of claims 11 to 14, comprising another reference sensor mounted to the reference component for obtaining another reference acceleration.

16. The monitoring system of claim 15, wherein the processor transforms a first of the reference accelerations to be in a same reference frame as a second of the reference accelerations.

17. The monitoring system of claim 16, wherein the processor averages the first and second accelerations and using the averaged acceleration to determine theexpected acceleration of the wear component.

18. The monitoring system of any one of claims 11 to 17, wherein the processor determines the relationship between wear component acceleration and reference component acceleration by:using a cost function comprising candidate wear component unit vectors for a plurality of data points, and differences between (a) an acceleration calculated from the candidate unit vector and a corresponding bucket acceleration; and (b) a corresponding measured wear component acceleration; andusing a numerical solver to find a wear component vector that minimises cost across the plurality of data points.

19. The monitoring system of any one of claims 11 to 18, comprising a plurality of wear component sensors corresponding to respective wear components, and wherein the processor determines an expected acceleration for each wear component and compares, for each wear component, a corresponding measured acceleration of the wear component in order to determine whether an event associated with the respective wear component has occurred20. The monitoring system as claimed in any one of claims 11 to 19, wherein the processor initiates an action in response to the event, wherein the action comprises at least one of generating an alarm, generating a warning message, and stopping at least part of a machine.

Images