Injection molding equipment abnormal vibration diagnosis method and device based on decision network

Abstract

The invention provides an injection molding equipment abnormal vibration diagnosis method and device based on a decision network, and the method comprises the steps: determining deviation characterization based on the expected vibration characteristics and multi-channel vibration signals of injection molding equipment at a current operation stage; whether the normal fluctuation boundary is exceeded or not is judged based on the deviation representation and the normal vibration allowable range of the injection molding equipment in the current operation stage, and a vibration deviation judgment result is obtained; if the vibration deviation judgment result indicates that a normal fluctuation boundary is exceeded, determining an abnormal vibration propagation path based on each fault sensitive area associated with the current operation stage and a local vibration coupling relationship corresponding to each area in combination with a signal distribution mode of the multi-channel vibration signal in the fault sensitive area; and abnormal vibration diagnosis is carried out based on the path structure of the abnormal vibration propagation path and the action direction of the load borne by the injection molding equipment in the current operation stage, and the mechanical fault physical cause is obtained. The accuracy of abnormal vibration diagnosis of the injection molding equipment is improved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method and apparatus for diagnosing abnormal vibrations in injection molding equipment based on a decision network. Background Technology

[0002] Abnormal vibration is an important indicator of the health status of injection molding equipment during operation. In existing technology, a common method for diagnosing abnormal vibration is a threshold comparison method based on a single sensor signal: an accelerometer is installed at a key part of the injection molding equipment to collect vibration signals in real time, and the signal amplitude is compared with a preset fixed threshold; if the threshold is exceeded, it is considered abnormal.

[0003] While this method is simple to implement and responds quickly, it cannot distinguish between normal vibration fluctuations caused by changes in process parameters and genuine abnormal vibrations caused by mechanical faults. For example, when changing molds or adjusting injection speeds, the vibration characteristics of the equipment will change significantly, but such changes fall within the scope of normal operating conditions. However, the threshold comparison method often misjudges these changes as faults, resulting in a high false alarm rate and poor diagnostic reliability. Therefore, there is an urgent need for a diagnostic method that can accurately identify genuine abnormal vibrations. Summary of the Invention

[0004] This invention provides a method and apparatus for diagnosing abnormal vibrations in injection molding equipment based on a decision network, thereby improving the accuracy of abnormal vibration diagnosis in injection molding equipment.

[0005] In a first aspect, the present invention provides a method for diagnosing abnormal vibrations in injection molding equipment based on a decision network, comprising: Based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor, the deviation between the actual vibration behavior and the expected vibration characteristics is determined. Based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment in the current operating stage, it is determined whether the normal fluctuation boundary is exceeded, and the vibration deviation judgment result is obtained. If the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, then based on each fault-sensitive area associated with the current operation stage and the local vibration coupling relationship corresponding to each area, combined with the signal distribution pattern of the multi-channel vibration signal in the fault-sensitive area, the abnormal vibration propagation path is determined. Based on the path structure of the abnormal vibration propagation path and the direction of the load acting on the injection molding equipment during the current operating stage, abnormal vibration diagnosis is performed to obtain the physical cause of mechanical failure.

[0006] In a second aspect, the present invention also provides a decision network-based abnormal vibration diagnosis device for injection molding equipment, applied to the decision network-based abnormal vibration diagnosis method for injection molding equipment as described in the first aspect; the decision network-based abnormal vibration diagnosis device for injection molding equipment includes: The vibration deviation calculation module is used to determine the deviation between the actual vibration behavior and the expected vibration characteristics based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor. The vibration deviation judgment module is used to determine whether the normal fluctuation boundary is exceeded based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment in the current operating stage, and to obtain the vibration deviation judgment result. The abnormal vibration monitoring module is used to determine the abnormal vibration propagation path based on each fault-sensitive area associated with the current operating phase and the local vibration coupling relationship corresponding to each area, combined with the signal distribution pattern of the multi-channel vibration signal in the fault-sensitive area, if the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary. The abnormal vibration diagnosis module is used to diagnose abnormal vibrations based on the path structure of the abnormal vibration propagation path and the direction of the load acting on the injection molding equipment during the current operating stage, so as to obtain the physical cause of the mechanical failure.

[0007] Thirdly, the present invention also provides an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing the abnormal vibration diagnosis method for injection molding equipment based on decision networks as described above.

[0008] Fourthly, the present invention also provides a non-transitory computer-readable storage medium storing a computer software program, which, when executed by a processor, implements the above-described method for diagnosing abnormal vibrations of injection molding equipment based on a decision network.

[0009] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for diagnosing abnormal vibrations in injection molding equipment based on a decision network.

[0010] The abnormal vibration diagnosis method for injection molding equipment based on a decision network provided in this invention determines the deviation between the actual vibration behavior and the expected vibration characteristics based on the expected vibration characteristics of the injection molding equipment in the current operating stage and multi-channel vibration signals. Therefore, the deviation characterization can accurately capture the difference between the current vibration and the normal vibration benchmark of the corresponding operating stage, avoiding the risk of misjudgment caused by fixed thresholds deviating from the operating stage. Based on the deviation characterization and the normal vibration tolerance range of the current operating stage, a vibration deviation judgment result is obtained, realizing the dynamic definition of the normal vibration boundary in combination with the operating stage, replacing the fixed threshold comparison, which can effectively distinguish between normal vibration fluctuations caused by changes in process parameters and abnormal vibrations caused by mechanical faults, initially reducing the risk of false alarms. If the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, based on the fault-sensitive areas associated with the current operating stage and their local vibration coupling relationship, combined with the signal distribution pattern of multi-channel vibration signals in each fault-sensitive area, the abnormal vibration propagation path is determined, thereby capturing the spatial distribution and transmission law of abnormal vibration through the abnormal vibration propagation path, avoiding misjudging vibration changes under normal operating conditions as faults. Based on the path structure of the abnormal vibration propagation path and the direction of the load on the equipment during the current operation phase, abnormal vibration diagnosis is completed and the physical cause of mechanical failure is obtained. This enables accurate tracing from vibration deviation to the root cause of the failure, rather than simply judging by threshold exceeding limits. It effectively solves the problems of being unable to distinguish between normal vibration fluctuations and real abnormal vibrations, high false alarm rate, and poor diagnostic reliability, thus improving the accuracy of abnormal vibration diagnosis for injection molding equipment. Attached Figure Description

[0011] Figure 1 This is a flowchart illustrating the abnormal vibration diagnosis method for injection molding equipment based on a decision network provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the abnormal vibration diagnosis device for injection molding equipment based on decision network provided in an embodiment of the present invention; Figure 3 An embodiment diagram of the electronic device provided in this invention; Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0014] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0015] Optionally, see Figure 1 , Figure 1 This is a flowchart illustrating the abnormal vibration diagnosis method for injection molding equipment based on a decision network provided by the present invention. In this embodiment, the execution entity of the abnormal vibration diagnosis method for injection molding equipment based on a decision network is an abnormal diagnosis device. Therefore, the abnormal vibration diagnosis method for injection molding equipment based on a decision network includes: Step 10: Based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor, determine the deviation between the actual vibration behavior and the expected vibration characteristics.

[0016] Optionally, the abnormality diagnosis device acquires the expected vibration characteristics of the injection molding equipment in the current operating stage. The current operating stage refers to the specific working link of the injection molding equipment, including but not limited to the mold closing stage, injection stage, pressure holding stage, cooling stage, and mold opening stage. The working state of the injection molding equipment is different in each operating stage, and the corresponding expected vibration characteristics are also different.

[0017] The expected vibration characteristics refer to the set of characteristic parameters that the vibration signal of the injection molding equipment should have when it is in normal working condition during the current operation stage. The set of characteristic parameters is determined and stored in advance by the abnormal diagnosis device based on the design parameters of the injection molding equipment, historical normal operation data, and normal operation benchmark data of the same type of equipment, after statistical analysis and experience summary. The characteristic parameters include, but are not limited to, vibration amplitude, vibration frequency, vibration duration, and vibration waveform pattern.

[0018] Furthermore, the abnormality diagnosis device acquires multi-channel vibration signals collected by vibration sensors. Vibration sensors are detection elements installed at different key locations on the injection molding equipment to collect vibration data of the equipment in real time. Multi-channel vibration signals refer to the collection of vibration data from multiple vibration sensors at different locations on the injection molding equipment. Each channel's vibration signal includes parameters such as vibration amplitude, vibration frequency, vibration duration, and vibration waveform pattern.

[0019] Furthermore, the anomaly diagnosis device preprocesses the acquired multi-channel vibration signals. The preprocessing operations include, but are not limited to, signal denoising, signal filtering, and signal normalization. Signal denoising refers to removing irrelevant noise such as environmental interference signals and sensor error signals mixed in with the vibration signals to ensure the authenticity and accuracy of the vibration signals. Signal filtering refers to filtering out signals within the effective frequency range related to the vibration of the injection molding equipment and eliminating interference signals that exceed the effective frequency range. Signal normalization refers to converting the vibration signals from different channels to the same numerical range.

[0020] Furthermore, the anomaly diagnosis device extracts the actual vibration characteristic parameters from the multi-channel vibration signals. The extracted characteristic parameters are completely consistent with the parameter types included in the expected vibration characteristics, ensuring the relevance and rationality of the comparison. Subsequently, the anomaly diagnosis device compares the extracted actual vibration characteristic parameters with the pre-stored expected vibration characteristic parameters under the current operating stage, calculates the difference between the actual value and the expected value of each characteristic parameter, and determines the deviation characterization between the actual vibration behavior and the expected vibration characteristics.

[0021] Deviation characterization refers to a set of characterization parameters used to quantify the degree and type of difference between actual vibration characteristics and expected vibration characteristics. The specific form of deviation characterization is determined according to the type of characteristic parameters being compared, including but not limited to amplitude deviation, frequency deviation, waveform deviation, and duration deviation. Among them, amplitude deviation refers to the difference between the actual vibration amplitude and the expected vibration amplitude, or the proportion of the difference to the expected amplitude; frequency deviation refers to the difference between the actual vibration frequency and the expected vibration frequency, or the proportion of the difference to the expected frequency; waveform deviation refers to the degree of matching between the actual vibration waveform and the expected vibration waveform, determined by comparing parameters such as the peak position, valley position, and waveform period; duration deviation refers to the difference between the actual vibration duration and the expected vibration duration, or the proportion of the difference to the expected duration.

[0022] Furthermore, the anomaly diagnosis device integrates the calculated various deviation parameters to obtain a deviation characterization, ensuring that the deviation characterization can comprehensively and accurately reflect the difference between the actual vibration behavior and the expected vibration characteristics. The deviation characterization is integrated by arranging various deviation parameters in a preset order and clearly labeling the characteristic parameter type, actual value, expected value and deviation calculation result corresponding to each deviation parameter.

[0023] In one embodiment, the current operating stage of the injection molding equipment is the injection stage. The abnormality diagnosis device pre-stores the expected vibration characteristics of this stage as follows: the expected vibration amplitude range is 5 to 8 mm, the expected vibration frequency range is 10 to 15 Hz, the expected vibration duration is 10 to 12 seconds, the expected vibration waveform is a periodic sine wave, and the peak position of the waveform is spaced 2 seconds apart, and the valley position is spaced 2 seconds apart.

[0024] The vibration sensor has three channels, installed at the injection cylinder, screw, and clamping mechanism of the injection molding equipment, respectively. After preprocessing, the actual vibration characteristic parameters of the collected multi-channel vibration signals are as follows: Channel 1 (injection cylinder): actual vibration amplitude is 10 mm, actual vibration frequency is 18 Hz, actual vibration duration is 15 seconds, and the actual vibration waveform is a non-periodic waveform with irregular peak position intervals; Channel 2 (screw): actual vibration amplitude is 9 mm, actual vibration frequency is 17 Hz, actual vibration duration is 14 seconds, and the actual vibration waveform is a non-periodic waveform with irregular peak position intervals; Channel 3 (clamping mechanism): actual vibration amplitude is 8.5 mm, actual vibration frequency is 16 Hz, actual vibration duration is 13 seconds, and the actual vibration waveform is a non-periodic waveform with irregular peak position intervals.

[0025] The anomaly diagnosis device compares the actual vibration characteristic parameters of each channel with the expected vibration characteristic parameters parameter by parameter, and calculates the deviation characteristics as follows: the amplitude deviation of channel 1 is 2 mm (10 mm - 8 mm), and the amplitude deviation ratio is 25% (2 mm / 8 mm * 100%); the frequency deviation is 3 Hz (18 Hz - 15 Hz), and the frequency deviation ratio is 20% (3 Hz / 15 Hz * 100%); the duration deviation is 3 seconds (15 seconds - 12 seconds), and the duration deviation ratio is 25% (3 seconds / 12 seconds * 100%); the waveform deviation is completely mismatched, with no fixed peak and valley intervals.

[0026] Channel 2 has an amplitude deviation of 1 mm (9 mm - 8 mm), with an amplitude deviation ratio of 12.5% ​​(1 mm / 8 mm * 100%); a frequency deviation of 2 Hz (17 Hz - 15 Hz), with a frequency deviation ratio of 13.3% (2 Hz / 15 Hz * 100%); a duration deviation of 2 seconds (14 seconds - 12 seconds), with a duration deviation ratio of 16.7% (2 seconds / 12 seconds * 100%); and a waveform deviation of no match, with no fixed peak and valley intervals.

[0027] The amplitude deviation of channel 3 is 0.5 mm (8.5 mm - 8 mm), with an amplitude deviation ratio of 6.25% (0.5 mm / 8 mm * 100%); the frequency deviation is 1 Hz (16 Hz - 15 Hz), with a frequency deviation ratio of 6.7% (1 Hz / 15 Hz * 100%); the duration deviation is 1 second (13 seconds - 12 seconds), with a duration deviation ratio of 8.3% (1 second / 12 seconds * 100%); the waveform deviation is completely mismatched, with no fixed peak and valley intervals.

[0028] Step 20: Based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment under the current operating stage, determine whether it exceeds the normal fluctuation boundary, and obtain the vibration deviation judgment result.

[0029] Optionally, the anomaly diagnosis device obtains the normal vibration tolerance range of the injection molding equipment during the current operating phase. The normal vibration tolerance range refers to the maximum allowable deviation between the actual vibration characteristics and the expected vibration characteristics of the injection molding equipment when it is in normal operating condition during the current operating phase. This range is predetermined and stored by the anomaly diagnosis device after scientific calculation and statistical analysis based on the design standards, safety operation requirements, historical fault data, and operating experience of the same model of equipment. The normal vibration tolerance range sets clear upper and lower limits for the deviation of each vibration characteristic parameter. For parameters that are not allowed to have negative deviations (such as vibration amplitude, vibration frequency, and vibration duration), only the upper limit is set to ensure that the deviation is within the range that does not affect the normal operation of the equipment or cause equipment failure.

[0030] The setting of the normal vibration tolerance range needs to be combined with the characteristics of the current operating stage of the injection molding equipment. The normal vibration tolerance range is different in different operating stages. For example, during the injection stage, the equipment bears a larger load and the vibration amplitude is relatively large, so its normal vibration tolerance range can be appropriately relaxed; during the cooling stage, the equipment operates relatively smoothly and the vibration amplitude is smaller, so its normal vibration tolerance range needs to be strictly controlled. At the same time, the normal vibration tolerance range needs to be set separately for each channel of the multi-channel vibration signal, because different channels correspond to different positions of the injection molding equipment, and their vibration sensitivity and normal vibration range are different, ensuring the specificity and accuracy of the judgment.

[0031] Furthermore, the anomaly diagnosis device compares each deviation parameter in the deviation characterization with the normal vibration tolerance range of the corresponding channel and corresponding characteristic parameter under the current operating stage to determine whether each deviation parameter exceeds the corresponding tolerance range. The comparison process is performed parameter-by-parameter and channel-by-channel. The specific comparison rules are as follows: for deviation parameters with upper and lower limits, if the actual value of the deviation parameter is greater than the upper limit or less than the lower limit, it is determined that the deviation parameter exceeds the normal vibration tolerance range; for deviation parameters with only an upper limit, if the actual value of the deviation parameter is greater than the upper limit, it is determined that the deviation parameter exceeds the normal vibration tolerance range; if the actual value of the deviation parameter is within the normal vibration tolerance range, it is determined that the deviation parameter does not exceed the normal vibration tolerance range.

[0032] Furthermore, after comparing all deviation parameters, the anomaly diagnosis device determines the vibration deviation judgment result based on the comparison results. The vibration deviation judgment result includes only two situations: exceeding the normal fluctuation boundary and not exceeding the normal fluctuation boundary. Exceeding the normal fluctuation boundary means that the deviation parameter of at least one channel and at least one vibration characteristic parameter exceeds the normal vibration tolerance range corresponding to the current operating stage; not exceeding the normal fluctuation boundary means that the deviation parameters of all channels and all vibration characteristic parameters are within the normal vibration tolerance range corresponding to the current operating stage.

[0033] Continuing with the above embodiment, the pre-stored allowable normal vibration ranges for each channel in this stage are as follows: Channel 1 (injection cylinder): amplitude deviation upper limit is 1.5 mm (deviation ratio upper limit is 18.75%), frequency deviation upper limit is 2.5 Hz (deviation ratio upper limit is 16.7%), duration deviation upper limit is 2 seconds (deviation ratio upper limit is 16.7%), waveform deviation allowable range is basically consistent (slight irregularity is allowed, complete non-consistency is not allowed); Channel 2 (screw): amplitude deviation upper limit is 1.2 mm (deviation ratio... For example, the upper limit for amplitude deviation is 15%, the upper limit for frequency deviation is 2.2 Hz (the upper limit for deviation ratio is 14.7%), the upper limit for duration deviation is 1.8 seconds (the upper limit for deviation ratio is 15%), and the permissible range for waveform deviation is basically consistent; Channel 3 (mold clamping mechanism): the upper limit for amplitude deviation is 1 mm (the upper limit for deviation ratio is 12.5%), the upper limit for frequency deviation is 1.5 Hz (the upper limit for deviation ratio is 10%), the upper limit for duration deviation is 1.5 seconds (the upper limit for deviation ratio is 12.5%), and the permissible range for waveform deviation is basically consistent.

[0034] The abnormality diagnostic device acquires the deviation characteristics determined in step 10, and compares each channel and each deviation parameter with the above-mentioned normal vibration tolerance range parameter by parameter: Channel 1 amplitude deviation is 2 mm, exceeding the upper limit of 1.5 mm; frequency deviation is 3 Hz, exceeding the upper limit of 2.5 Hz; duration deviation is 3 seconds, exceeding the upper limit of 2 seconds; waveform deviation is completely mismatched, exceeding the tolerance range of basic match; Channel 2 amplitude deviation is 1 mm, not exceeding the upper limit of 1.2 mm; frequency deviation is 2 Hz, not exceeding the upper limit of 2.2 Hz; duration deviation is 2 seconds, exceeding the upper limit of 1.8 seconds; waveform deviation is completely mismatched, exceeding the tolerance range of basic match; Channel 3 amplitude deviation is 0.5 mm, not exceeding the upper limit of 1 mm; frequency deviation is 1 Hz, not exceeding the upper limit of 1.5 mm; duration deviation is 1 second, not exceeding the upper limit of 1.5 seconds; waveform deviation is completely mismatched, exceeding the tolerance range of basic match.

[0035] Since four deviation parameters of channel 1 exceed the allowable range of normal vibration, two deviation parameters of channel 2 exceed the allowable range of normal vibration, and one deviation parameter of channel 3 exceeds the allowable range of normal vibration, the condition of "at least one channel and at least one deviation parameter exceeding the allowable range of normal vibration" is met. Therefore, the abnormality diagnosis device determines the vibration deviation judgment result as: exceeding the normal fluctuation boundary.

[0036] In one embodiment, if, after comparison, all deviation parameters of each channel are within the above-mentioned allowable range of normal vibration, for example, if the amplitude deviation of channel 1 is 1.2 mm, the frequency deviation is 2 Hz, and the duration deviation is 1.5 seconds, and the waveforms are basically consistent; if the amplitude deviation of channel 2 is 1 mm, the frequency deviation is 2 Hz, and the duration deviation is 1.6 seconds, and the waveforms are basically consistent; and if the amplitude deviation of channel 3 is 0.8 mm, the frequency deviation is 1.2 Hz, and the duration deviation is 1.2 seconds, and the waveforms are basically consistent, then the abnormal diagnosis device determines the vibration deviation judgment result as: not exceeding the normal fluctuation boundary.

[0037] Step 30: If the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, then based on each fault-sensitive area associated with the current operation stage and the corresponding local vibration coupling relationship of each area, combined with the signal distribution pattern of multi-channel vibration signals in the fault-sensitive area, the abnormal vibration propagation path is determined.

[0038] Optionally, if the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, the process of the embodiment of the present invention is executed; if the vibration deviation judgment result indicates that it does not exceed the normal fluctuation boundary, the process of the embodiment of the present invention is not executed. Therefore, if the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, the abnormal diagnosis device obtains each fault-sensitive area associated with the current operating stage of the injection molding equipment. The fault-sensitive area refers to the key area of ​​the injection molding equipment that is prone to mechanical failure and is prone to abnormal vibration after the failure occurs in the current operating stage. The fault-sensitive areas associated with each operating stage are different. The division of fault-sensitive areas is determined based on the structural characteristics, working principle, and historical failure location statistics of the injection molding equipment. For example, the fault-sensitive areas in the injection stage include the injection cylinder area, the screw area, and the injection mechanism connection area. The fault-sensitive areas in the mold closing stage include the mold clamping mechanism area, the guide pillar and guide sleeve area, and the mold closing cylinder area.

[0039] Furthermore, the abnormality diagnosis device acquires the local vibration coupling relationship corresponding to each fault-sensitive area. The local vibration coupling relationship refers to the transmission relationship and mutual influence relationship of vibration signals between different components within the same fault-sensitive area and between different fault-sensitive areas. This relationship is based on the mechanical structure connection method and force transmission path of the injection molding equipment, which are predetermined and stored, reflecting the propagation law of vibration in each fault-sensitive area and between components within the area.

[0040] Optionally, the abnormality diagnosis device acquires the preprocessed multi-channel vibration signal in step 10 and extracts the signal distribution pattern of the multi-channel vibration signal in each fault-sensitive area. The signal distribution pattern refers to the intensity distribution, frequency distribution, and time distribution characteristics of the multi-channel vibration signal in different fault-sensitive areas, reflecting the distribution and intensity differences of abnormal vibration in each fault-sensitive area.

[0041] Furthermore, the abnormality diagnosis device determines the abnormal vibration propagation path based on each fault-sensitive area associated with the current operating phase, the local vibration coupling relationship corresponding to each area, and the signal distribution pattern of multi-channel vibration signals in the fault-sensitive area, as described in steps 301 to 304.

[0042] Step 40: Based on the path structure of the abnormal vibration propagation path and the direction of the load on the injection molding equipment under the current operating stage, abnormal vibration diagnosis is performed to obtain the physical cause of mechanical failure.

[0043] Optionally, the load direction under the current operating stage refers to the direction of the force exerted on each component of the injection molding equipment when it is working in the current operating stage. This direction is predetermined based on the working principle and action process of the injection molding equipment in the current operating stage. The load direction is different in different operating stages. For example, the load direction in the injection stage is mainly axial (along the screw length direction) and radial (perpendicular to the screw length direction), while the load direction in the mold closing stage is mainly transverse (along the direction of movement of the mold clamping mechanism).

[0044] Optionally, the abnormality diagnosis device analyzes the propagation law, starting source, and scope of influence of abnormal vibration based on the path structure of the acquired abnormal vibration propagation path. The path structure refers to the number of fault-sensitive areas contained in the abnormal vibration propagation path, the connection relationship between the areas, and the vibration transmission sequence, reflecting the diffusion characteristics and degree of influence of the abnormal vibration.

[0045] Furthermore, the abnormality diagnosis device combines the load direction under the current operating stage of the injection molding equipment to perform abnormal vibration diagnosis and determine the physical cause of mechanical failure, as shown in steps 401 to 404. The physical cause of mechanical failure is characterized by the existence of a type of structural defect in mechanical components under load and motion conditions caused by the dominant failure mechanism. The dominant failure mechanism refers to the main cause of structural defects in mechanical components, including but not limited to wear, fatigue, deformation, fracture, and loosening. Structural defects refer to the abnormalities in shape, size, and structure of mechanical components caused by the failure mechanism, including but not limited to excessive dimensional wear, surface fatigue cracks, component deformation, component fracture, and loosening of connection parts.

[0046] The embodiments of the present invention enable precise tracing from vibration deviation to the root cause of the fault, rather than simply judging the threshold exceedance, thereby improving the accuracy of abnormal vibration diagnosis of injection molding equipment.

[0047] Optionally, the processes of steps 301 to 304 include: Step 301: Based on each fault-sensitive region and the corresponding local vibration coupling relationship, construct the structural connection topology between fault-sensitive regions.

[0048] Optionally, the anomaly diagnosis device uses each fault-sensitive area as an independent node and constructs a structural connection topology between fault-sensitive areas based on the local vibration coupling relationship.

[0049] Among them, the structural connection topology represents the physical adjacency relationship of each fault-sensitive region on the mechanical structure and the dynamic energy transfer path established through local vibration coupling relationship. It is a visual and structured representation of the vibration transmission law between fault-sensitive regions, and includes two core elements: First, there is the physical adjacency relationship, which refers to the actual connection of each fault-sensitive area in the mechanical structure of the injection molding equipment. This includes direct connection and indirect connection. Direct connection means that the mechanical parts corresponding to two fault-sensitive areas are in direct contact and fixedly connected, which can realize the direct transmission of vibration signals. Indirect connection means that the mechanical parts corresponding to two fault-sensitive areas are not in direct contact, but are indirectly related through other parts, and the vibration signal needs to be transmitted through intermediate parts.

[0050] Second, there is the dynamic energy transfer path, which is determined based on the local vibration coupling relationship. It is the correlation of the direction, path and intensity of vibration energy transfer between various fault-sensitive areas. The dynamic energy transfer path corresponds to the physical adjacency relationship. Directly connected fault-sensitive areas form a direct dynamic energy transfer path, while indirectly connected fault-sensitive areas form an indirect dynamic energy transfer path. The correlation of transfer intensity reflects the difference in the transmission efficiency of vibration energy in different paths.

[0051] Furthermore, when constructing the structural connection topology, the anomaly diagnostic device must clearly label the name of the fault-sensitive area corresponding to each node, specify the connection type between nodes (direct connection, indirect connection), and label the transmission direction and transmission intensity level (strong, medium, weak) of each dynamic energy transfer path. The transmission intensity level is determined based on the transmission efficiency of the vibration signal in the local vibration coupling relationship; the higher the transmission efficiency, the stronger the transmission intensity level. The construction of the structural connection topology must strictly follow the actual mechanical structure and local vibration coupling relationship of the injection molding equipment to ensure that the topology can truly and accurately reflect the physical connection and vibration transmission law between fault-sensitive areas.

[0052] Step 302: Based on the structural connection topology and the signal distribution pattern of multi-channel vibration signals in the fault-sensitive area, determine the regional activity state of each fault-sensitive area.

[0053] Optionally, the anomaly diagnosis device identifies the fault-sensitive area corresponding to each node in the structural connection topology based on the structural connection topology, and accurately maps the signal distribution pattern to each fault-sensitive area to ensure that each fault-sensitive area has corresponding multi-channel vibration signal distribution data.

[0054] Among them, the active state of the region indicates whether the corresponding fault-sensitive region exhibits vibration response characteristics that exceed the normal fluctuation boundary. The active state of the region includes only two situations: abnormal state and normal state.

[0055] An abnormal state refers to the vibration signal distribution pattern corresponding to the fault-sensitive area, which exhibits vibration response characteristics exceeding the normal fluctuation boundary. That is, the vibration signal intensity, frequency, time distribution, and other parameters in this area are significantly different from the signal distribution pattern under normal operation, and the difference exceeds the preset range. A normal state refers to the vibration signal distribution pattern corresponding to the fault-sensitive area, which does not exhibit vibration response characteristics exceeding the normal fluctuation boundary. That is, the vibration signal intensity, frequency, time distribution, and other parameters in this area are basically consistent with the signal distribution pattern under normal operation, and the difference is within the preset range.

[0056] Optionally, the specific process for determining the active state of a region in this embodiment of the invention is as follows: The core parameters of the vibration signal distribution pattern corresponding to each fault-sensitive area are extracted. These core parameters include the maximum vibration signal intensity, the vibration frequency fluctuation range, and the timing of vibration signal occurrence. These parameters can intuitively reflect the vibration response characteristics of the area. Then, the signal distribution reference parameters of the fault-sensitive area during normal operation in the current operating stage are called up. The signal distribution reference parameters are predetermined and stored by the anomaly diagnosis device based on historical normal operation data and are completely consistent with the extracted core parameter types. Next, the core parameters of each fault-sensitive area are compared with the corresponding signal distribution reference parameters to determine whether they exceed the preset difference range. The preset difference range is determined based on the normal operation standards of the injection molding equipment.

[0057] Optionally, the comparison and judgment rule in this embodiment of the invention is as follows: if at least one of the core parameters of a fault-sensitive area exceeds the corresponding preset difference range, the area activity state of the fault-sensitive area is determined to be abnormal; if all the core parameters of the fault-sensitive area are within the corresponding preset difference range, the area activity state of the fault-sensitive area is determined to be normal. The anomaly diagnosis device performs the above comparison and judgment process for each fault-sensitive area, determining the area activity state of each fault-sensitive area one by one, ensuring no omissions and no misjudgments.

[0058] Step 303: Based on the fault-sensitive regions whose active state is in an abnormal state, combined with the structural connection topology, determine the set of candidate regions for anomaly sources.

[0059] Optionally, the anomaly diagnosis device filters out all fault-sensitive regions whose active state is in an abnormal state, forming an initial set of anomaly regions. This initial set of anomaly regions includes all fault-sensitive regions exhibiting abnormal vibration response characteristics. Subsequently, the anomaly diagnosis device invokes the structural connection topology constructed in step 301. Based on the physical adjacency relationships and dynamic energy transfer paths of each fault-sensitive region in the structural connection topology, it performs a correlation analysis on each fault-sensitive region in the initial set of anomaly regions, filtering out regions that may serve as sources of abnormal vibration, thus forming a candidate set of anomaly source regions.

[0060] Optionally, the core logic of the correlation analysis in this embodiment of the invention is as follows: the propagation of abnormal vibration has directionality and temporality. The source region of abnormal vibration (i.e., the abnormal source) is usually the region in the initial abnormal region set where the vibration signal appears earliest, the vibration signal intensity is strongest, and it can transmit vibration energy to other abnormal regions through the dynamic energy transfer path in the structural connection topology. If the vibration signal of a certain abnormal region appears later than other abnormal regions, and its vibration energy can only come from other abnormal regions (it cannot transmit vibration energy to other abnormal regions), then this region does not belong to the candidate region of abnormal source, but is only the propagation region of abnormal vibration, rather than the source region.

[0061] Optionally, the specific analysis process of this embodiment of the invention is as follows: extract the occurrence time sequence and maximum intensity value of the vibration signal of each abnormal region in the initial abnormal region set, sort the time sequence to determine the abnormal region with the earliest time sequence, sort the intensity values ​​to determine the abnormal region with the strongest intensity; then, combine the dynamic energy transfer path of the structural connection topology to determine whether the abnormal region with the earliest time sequence and the strongest intensity can transfer vibration energy to other abnormal regions through the transfer path. If it can transfer, the region is included in the abnormal source candidate region set; if there are multiple abnormal regions with the earliest time sequence and the strongest intensity, and these regions can transfer vibration energy to each other, then all these regions are included in the abnormal source candidate region set; if an abnormal region has a later time sequence and a weaker intensity, and its vibration energy can only come from other abnormal regions and has no path for outward transfer, then it is not included in the abnormal source candidate region set.

[0062] It should be noted that the set of candidate regions for anomaly sources can contain one or more anomaly regions. If it contains only one anomaly region, then that region is the only candidate region for anomaly source; if it contains multiple anomaly regions, then all of these regions are possible anomaly sources.

[0063] Step 304: Based on the set of candidate regions of the abnormal source and the adjacency pointing relationship in the structural connection topology, the propagation path is analyzed to obtain the propagation path of the abnormal vibration.

[0064] Optionally, the adjacency orientation relationship is a core component of the structural connection topology, representing the physical adjacency association between nodes in each fault-sensitive area and the direction of dynamic energy transfer. That is, it clarifies which nodes each node (fault-sensitive area) can transfer vibration energy to and which nodes it can receive vibration energy from, matching the direction and intensity level of the dynamic energy transfer path in the structural connection topology.

[0065] Optionally, propagation path analysis is performed based on the adjacency pointing relationship in the set of candidate regions of the abnormal source and the structural connection topology to obtain the propagation path of the abnormal vibration, as in steps 3041 to 3043.

[0066] Based on fault-sensitive areas, local vibration coupling relationships, and multi-channel vibration signal distribution patterns, this invention progressively completes the construction of structural connection topology, determination of regional active states, screening of candidate regions for abnormal sources, and analysis of abnormal vibration propagation paths. It accurately determines the abnormal vibration propagation path, effectively solving the problem of low accuracy of abnormal vibration propagation paths and easy fault diagnosis deviations due to incorrect path judgment in the diagnosis of abnormal vibrations in injection molding equipment, thus improving the accuracy of abnormal vibration diagnosis in injection molding equipment.

[0067] Optionally, the processes of steps 3041 to 3043 include: Step 3041: Based on the set of candidate regions of anomaly sources and the adjacency pointing relationship in the structural connection topology, determine the sequence of local propagation directions starting from each candidate region of anomaly source.

[0068] Optionally, the anomaly diagnosis device identifies and marks each anomaly candidate region in the set of anomaly candidate regions one by one, and clarifies the node position in the structural connection topology corresponding to each anomaly candidate region, ensuring that each anomaly candidate region can accurately correspond to a unique node in the structural connection topology, avoiding the situation of region-node correspondence error, and ensuring the accuracy of subsequent analysis.

[0069] Furthermore, for each marked candidate region of an anomaly source, the anomaly diagnosis device extracts all adjacency pointing relationships of the corresponding nodes in the structural connection topology. It then filters out pointing relationships where the node acts as a vibration energy output end, transmitting vibration energy to nodes corresponding to other adjacent fault-sensitive regions, and removes pointing relationships where the node acts as a vibration energy receiver, receiving vibration energy from other nodes. This ensures that all extracted adjacency pointing relationships are related to the transmission of vibration energy from the current candidate region of anomaly source outwards. Adjacent fault-sensitive regions refer to fault-sensitive regions where the nodes corresponding to the current candidate region of anomaly source have a direct physical connection in the structural connection topology and can receive vibration energy from the current candidate region of anomaly source through a dynamic energy transfer path. Adjacent fault-sensitive regions do not include regions that only have indirect connections and cannot directly receive vibration energy from the current candidate region of anomaly source.

[0070] Furthermore, the selected outward propagation adjacency relationships are analyzed. Based on the dynamic energy transfer intensity level in the structural connection topology, the outward propagation relationships corresponding to each anomaly source candidate region are sorted from strongest to weakest, forming a local propagation direction sequence originating from that anomaly source candidate region. This local propagation direction sequence represents the path direction of vibration energy propagation from the anomaly source candidate region along the structural connection topology to adjacent fault-sensitive regions. It is the result of analyzing the outward propagation direction of vibration energy from a single anomaly source candidate region. Each local propagation direction sequence corresponds to a unique anomaly source candidate region, and each item in the sequence represents a specific propagation direction, i.e., from the current anomaly source candidate region to a certain adjacent fault-sensitive region, while also indicating the corresponding dynamic energy transfer intensity level.

[0071] Optionally, if a node corresponding to a candidate region of an anomaly source does not have an adjacency relationship in the structural connection topology that transmits vibration energy outward (i.e., no adjacent fault-sensitive region can receive its vibration energy), then the local propagation direction sequence corresponding to that candidate region of anomaly source is an empty sequence. The anomaly diagnosis device needs to mark this situation separately and subsequently exclude the candidate region of anomaly source from the propagation path analysis scope to avoid invalid analysis. For all candidate regions of anomaly source, the anomaly diagnosis device performs the above extraction, screening, and sorting process to determine the local propagation direction sequence corresponding to each candidate region of anomaly source.

[0072] Step 3042: Based on the temporal phase relationship between the local propagation direction sequence and the multi-channel vibration signal in adjacent fault-sensitive areas, determine the phase consistency judgment result in each propagation direction.

[0073] Optionally, the anomaly diagnosis device analyzes each local propagation direction sequence one by one. If a local propagation direction sequence is empty, the anomaly source candidate region corresponding to the sequence is skipped and no subsequent phase consistency judgment is performed. If a local propagation direction sequence is not empty, the anomaly diagnosis device performs phase consistency judgment for each propagation direction in the sequence and determines the phase consistency judgment result for each propagation direction.

[0074] Furthermore, the specific process for phase consistency judgment in a single propagation direction according to the embodiments of the present invention is as follows: First, identify the two fault-sensitive areas corresponding to the propagation direction, namely the propagation start area (current anomaly source candidate area) and the propagation target area (adjacent fault-sensitive area pointed to by the propagation direction); then, extract the vibration signals corresponding to the propagation start area and the propagation target area from the multi-channel vibration signals to ensure that the extracted vibration signals are signals collected within the same time period, avoid phase judgment errors caused by different collection times, and ensure the accuracy of the judgment results.

[0075] Furthermore, the anomaly diagnostic device analyzes the temporal phase relationship of the extracted vibration signals from the two regions. This temporal phase relationship refers to the phase difference and order of occurrence between the vibration signals from the propagation initiation region and the propagation target region, characterizing the synchronicity and sequence of the two vibration signals. The phase difference refers to the difference in phase values ​​between the two vibration signals at the same moment, and the order of occurrence refers to the sequential timing of the abnormal fluctuations in the vibration signals from the two regions. The anomaly diagnostic device determines the temporal phase relationship between the two vibration signals by comparing their waveform changes, peak occurrence times, and trough occurrence times. This comparison does not involve complex calculations; it clarifies the phase correlation between the two vibration signals solely through a direct comparison of waveform characteristics and time points.

[0076] Furthermore, based on the determined temporal phase relationship and the physical connection sequence corresponding to the propagation direction in the structural connection topology, the anomaly diagnosis device determines whether the vibration signals of the two regions meet the preset phase consistency requirements. The phase consistency judgment result indicates whether the vibration signals appear sequentially according to the physical connection sequence of the structural connection topology, including phase consistency and phase inconsistency. Phase consistency means that the vibration signal in the propagation initiation region exhibits abnormal fluctuations first, followed by the vibration signal in the propagation target region, and the phase difference between the two regions is within a preset reasonable range, conforming to the physical law of vibration energy transmission from the propagation initiation region to the propagation target region in the structural connection topology. Phase inconsistency means that the vibration signal in the propagation target region exhibits abnormal fluctuations first, followed by the vibration signal in the propagation initiation region, or that the vibration signals in both regions exhibit abnormal fluctuations simultaneously but the phase difference exceeds the preset reasonable range, or that the vibration signals in the two regions have no obvious phase correlation, which does not conform to the vibration energy transmission law in the structural connection topology.

[0077] The preset reasonable range is a phase difference range that the anomaly diagnosis device pre-determines and stores based on the mechanical structure characteristics and vibration energy transmission laws of the injection molding equipment, combined with historical normal operation data. This range can accurately reflect the phase correlation characteristics of vibration signals in two adjacent regions during normal vibration energy transmission, ensuring the rationality and rigor of the phase consistency judgment. Therefore, the anomaly diagnosis device executes the above judgment process for each propagation direction, determining the phase consistency judgment result for each propagation direction one by one.

[0078] Step 3043: Based on the phase consistency judgment results, perform propagation path analysis to obtain the abnormal vibration propagation path.

[0079] Optionally, propagation path analysis is performed based on the phase consistency judgment results to obtain the abnormal vibration propagation path, as described in steps 30431 to 30433.

[0080] The abnormal vibration propagation path obtained in this embodiment of the invention ensures that the path closely matches the actual vibration energy transfer law, thereby accurately locking the propagation trajectory of abnormal vibration. This effectively compensates for the low accuracy of propagation paths in abnormal vibration diagnosis of injection molding equipment and improves the accuracy of abnormal vibration diagnosis of injection molding equipment.

[0081] Optionally, the processes of steps 30431 to 30433 include: Step 30431: Based on the propagation direction that satisfies temporal consistency in the phase consistency judgment result and combined with the structural connection topology, determine the abnormal vibration propagation chain.

[0082] Optionally, the anomaly diagnosis device filters all phase consistency judgment results and extracts propagation directions whose phase consistency judgment results are consistent with the phase. Such propagation directions are propagation directions that satisfy temporal consistency. Temporal consistency means that the vibration signal in the propagation starting area first shows abnormal fluctuations, and the vibration signal in the propagation target area shows abnormal fluctuations later. The phase difference between the vibration signals in the two areas is within a preset reasonable range, which conforms to the physical law of vibration energy transfer in the structural connection topology. Propagation directions whose phase consistency judgment results are inconsistent with the phase are eliminated to avoid invalid propagation directions interfering with subsequent analysis.

[0083] Furthermore, the anomaly diagnosis device categorizes and organizes the selected propagation directions that meet the temporal consistency requirements, grouping propagation directions belonging to the same anomaly source candidate region together. This ensures that each group of propagation directions corresponds to a unique anomaly source candidate region, facilitating subsequent analysis of propagation paths according to each anomaly source candidate region. For each group of propagation directions, the anomaly diagnosis device combines the physical adjacency relationships and dynamic energy transfer paths in the structural connection topology to connect and analyze the propagation directions in series, constructing an abnormal vibration propagation chain.

[0084] The abnormal vibration propagation chain is a linear or branching path structure constructed from multiple fault-sensitive regions connected in chronological and structural adjacency order. It clearly represents the initial path of abnormal vibration energy starting from the candidate region of the abnormal source and gradually propagating to other fault-sensitive regions along a propagation direction that satisfies temporal consistency. Each abnormal vibration propagation chain takes an abnormal source candidate region as its starting point. Each subsequent connected fault-sensitive region satisfies the structural adjacency relationship with the previous fault-sensitive region, and the propagation directions corresponding to two adjacent fault-sensitive regions satisfy temporal consistency, ensuring the continuity and rationality of the propagation chain.

[0085] Optionally, the following rules must be followed when constructing abnormal vibration propagation chains in this embodiment of the invention: First, the starting endpoint of each propagation chain must be a candidate region of an abnormal source, and non-candidate regions of abnormal sources must not be used as the starting endpoint of the propagation chain; Second, two adjacent fault-sensitive regions in the propagation chain must have a physical adjacency relationship in the structural connection topology, and the propagation directions between them must satisfy temporal consistency, and fault-sensitive regions that are not structurally adjacent or have inconsistent temporal connections must not be present; Third, if a fault-sensitive region can receive vibration energy from the same candidate region of an abnormal source through multiple propagation directions that satisfy temporal consistency, then a branched propagation chain is constructed, with each branch corresponding to an independent propagation direction that satisfies temporal consistency; Fourth, if the propagation target region corresponding to a certain propagation direction has no other propagation directions that satisfy temporal consistency extending from it, then the propagation target region is used as the termination endpoint of the propagation chain, and a linear propagation chain is constructed.

[0086] Optionally, the anomaly diagnosis device performs the above-mentioned serial sorting process on the propagation direction corresponding to each anomaly source candidate region, and constructs an abnormal vibration propagation chain one by one. If the propagation direction corresponding to a certain anomaly source candidate region that meets the temporal consistency cannot be serialized to form a coherent propagation chain (i.e., there is no correlation or connection between the propagation directions), then the situation is marked and the corresponding abnormal vibration propagation chain is not constructed for the time being.

[0087] Step 30432: Determine the energy attenuation matching degree based on the signal amplitude attenuation trend of each fault-sensitive area in the abnormal vibration propagation chain and the path length in the structural connection topology.

[0088] Optionally, the anomaly diagnosis device analyzes each abnormal vibration propagation chain individually. If an abnormal vibration propagation chain only contains the starting endpoint (candidate region of the anomaly source) and has no other connected fault-sensitive regions, then the propagation chain cannot be analyzed for energy attenuation. The anomaly diagnosis device marks it separately and excludes it from subsequent analysis to avoid invalid analysis. For abnormal vibration propagation chains containing two or more fault-sensitive regions, the anomaly diagnosis device first extracts the vibration signal amplitude corresponding to each fault-sensitive region in the propagation chain. The extracted vibration signal amplitude is the maximum amplitude of the vibration signal in the fault-sensitive region during the occurrence of abnormal vibration, ensuring that the extracted amplitude can accurately reflect the intensity of abnormal vibration in the region. During the extraction process, it is necessary to ensure that the vibration signal amplitude of each fault-sensitive region comes from the same time period to avoid amplitude deviation caused by different acquisition times and ensure the accuracy of subsequent energy attenuation analysis.

[0089] Furthermore, the extracted vibration signal amplitudes are analyzed to determine the signal amplitude attenuation trend in each fault-sensitive area of ​​the abnormal vibration propagation chain. The signal amplitude attenuation trend refers to the variation of the maximum amplitude of the vibration signal as the abnormal vibration energy propagates along the propagation chain. Specifically, it manifests as a gradual decrease in the maximum amplitude of the vibration signal from the starting point (candidate region of the anomaly source) to the ending point of the propagation chain. If the amplitude increases or shows no significant change, it is considered to lack a normal attenuation trend. The anomaly diagnostic device compares the maximum amplitude of the vibration signals in two adjacent fault-sensitive areas within the propagation chain to determine the direction and degree of amplitude change, thus identifying the signal amplitude attenuation trend throughout the entire propagation chain.

[0090] Furthermore, the anomaly diagnosis device, in conjunction with the structural connection topology, determines the path length in the abnormal vibration propagation chain. The path length refers to the actual physical distance from the starting endpoint (candidate region of the anomaly source) to each intermediate fault-sensitive region and the ending endpoint in the abnormal vibration propagation chain. This distance is predetermined based on the actual installation position dimensions of the mechanical components corresponding to each fault-sensitive region in the structural connection topology. The anomaly diagnosis device directly calls the physical distance data between each fault-sensitive region stored in the structural connection topology to sort out the path length distribution of the entire propagation chain, clarify the physical distance between each adjacent fault-sensitive region, and the cumulative path length from the starting endpoint to each region.

[0091] Furthermore, the anomaly diagnosis device determines the energy attenuation matching degree based on the determined signal amplitude attenuation trend and path length. The energy attenuation matching degree is used to determine whether the amplitude change of the actual vibration signal along the abnormal vibration propagation chain conforms to the physical law of vibration energy attenuation with propagation distance in a mechanical structure. The energy attenuation matching degree is a quantitative representation of the degree of fit between the actual amplitude attenuation trend and the theoretical vibration energy attenuation law. The specific determination process is as follows: A preset vibration energy attenuation law is invoked. This law is based on the mechanical structural characteristics of the injection molding equipment, the principle of vibration energy transmission, and is pre-determined and stored in conjunction with historical normal operation data, clarifying the theoretical attenuation trend of vibration energy with propagation distance (i.e., the longer the path length, the more obvious the vibration signal amplitude attenuation). Subsequently, the actual signal amplitude attenuation trend is compared with the preset vibration energy attenuation law, and the degree of fit between the two is determined by considering the distribution of path length. The higher the degree of fit, the higher the energy attenuation matching degree, and vice versa.

[0092] Step 30433: Based on the energy attenuation matching degree, the abnormal vibration propagation chain that satisfies the preset physical attenuation law is determined as the abnormal vibration propagation path.

[0093] Optionally, the energy attenuation matching degree corresponding to each abnormal vibration propagation chain is judged one by one. Abnormal vibration propagation chains whose energy attenuation matching degree meets the preset physical attenuation law are selected, and abnormal vibration propagation chains whose energy attenuation matching degree does not meet the preset physical attenuation law are eliminated. For the selected abnormal vibration propagation chains that meet the preset physical attenuation law, if there are multiple propagation chains that meet the requirements, and these propagation chains belong to the same abnormal source candidate region and their propagation paths overlap, then the overlapping parts are merged, and different branches are retained to form a complete abnormal vibration propagation path; if there are multiple propagation chains that meet the requirements, and they belong to different abnormal source candidate regions and their propagation paths are independent of each other, then all of them are included in the results as valid abnormal vibration propagation paths; if there is only one propagation chain that meets the requirements, then that propagation chain is the abnormal vibration propagation path.

[0094] Furthermore, the abnormality diagnosis device identifies the abnormal vibration propagation chain that meets the preset physical attenuation law as the abnormal vibration propagation path. The abnormal vibration propagation path must clearly mark the starting source area (abnormal source candidate area), the intermediate propagation area, the ending area, as well as the physical distance between each adjacent area, the dynamic energy transfer intensity level, and the signal amplitude attenuation, clearly characterizing the complete process and law of abnormal vibration energy gradually propagating from the source along the mechanical structure.

[0095] The embodiments of the present invention use a propagation chain that satisfies a preset physical attenuation law as the propagation path of abnormal vibration, ensuring the accuracy and rationality of the path, eliminating invalid propagation chains that do not conform to physical laws, and accurately locking the propagation trajectory of abnormal vibration, thereby improving the accuracy of abnormal vibration diagnosis of injection molding equipment.

[0096] Optionally, the processes of steps 401 to 404 include: Step 401: Determine the spatial relative relationship based on the arrangement order of each fault-sensitive region in the path structure and the direction of load action.

[0097] Optionally, the abnormal diagnosis device analyzes the path structure of the abnormal vibration propagation path and extracts the arrangement order of each fault-sensitive area in the path structure. This arrangement order is the transmission order of abnormal vibration energy along the propagation path, clarifying the specific location of each fault-sensitive area in the propagation path and the connection relationship of adjacent areas, ensuring that the sequential arrangement pattern of all fault-sensitive areas can be accurately identified.

[0098] Subsequently, the abnormality diagnosis device clarifies the specific direction of the load acting on the injection molding equipment during the current operating stage. This direction is preset based on the working characteristics of the injection molding equipment during the current operating stage. For example, the load direction during the injection stage is mainly axial (along the screw length direction) and radial (perpendicular to the screw length direction), while the load direction during the mold closing stage is mainly transverse (along the direction of movement of the clamping mechanism). The preset load direction data is directly called up without recalculation.

[0099] Furthermore, for each fault-sensitive region in the path structure, the anomaly diagnosis device analyzes the spatial relative relationship between the fault-sensitive region and the load direction. The spatial relative relationship characterizes the spatial orientation of each fault-sensitive region along the abnormal vibration propagation path relative to the load direction, specifically including three types: parallel relationship, perpendicular relationship, and inclined relationship.

[0100] Parallel relationship means that the main extension direction of the fault-sensitive area is completely consistent with the direction of the load, and there is no angle between the direction of vibration energy transmission and the direction of the load. Perpendicular relationship means that the main extension direction of the fault-sensitive area is at a 90-degree angle to the direction of the load, and the direction of vibration energy transmission is perpendicular to the direction of the load, with no same or opposite components. Inclined relationship means that the main extension direction of the fault-sensitive area is neither completely consistent with the direction of the load nor at a 90-degree angle, but the angle is between 0 and 90 degrees (excluding 0 and 90 degrees), and there are certain same or opposite components between the direction of vibration energy transmission and the direction of the load.

[0101] Optionally, the specific process for determining the spatial relative relationship in this embodiment of the invention is as follows: the main extension direction of each fault-sensitive area is determined in advance based on the structural characteristics of the mechanical components corresponding to the fault-sensitive area. For example, the main extension direction of the injection cylinder area is axial, and the main extension direction of the mold clamping mechanism area is transverse. Subsequently, the main extension direction of each fault-sensitive area is compared with the direction of the load action. Based on the included angle and the consistency of the directions, the spatial relative relationship between the fault-sensitive area and the direction of the load action is determined.

[0102] Step 402: Based on the continuous fault-sensitive regions that are not perpendicular to the load direction in the spatial relative relationship, determine the set of load-sensitive propagation sections.

[0103] Optionally, the anomaly diagnosis device first filters the spatial relative relationships of all fault-sensitive areas, extracting fault-sensitive areas that are not perpendicular to the load direction. "Non-perpendicular" refers to situations where the spatial relative relationship is parallel or inclined. Fault-sensitive areas perpendicular to the load direction are then eliminated. The reason for eliminating perpendicularly related fault-sensitive areas is that their main extension direction is perpendicular to the load direction, and their vibration energy transmission direction has no component in the same or opposite direction as the load direction. The load has minimal impact on the vibration energy transmission in these areas and therefore they are not load-sensitive areas. Conversely, fault-sensitive areas that are parallel or inclined have vibration energy transmission directions in the same or opposite direction as the load direction, and the load directly affects the vibration energy transmission in these areas, thus they are load-sensitive areas.

[0104] Furthermore, the anomaly diagnosis device, based on the arrangement order of various fault-sensitive regions in the path structure, performs a continuous analysis of the selected non-perpendicularly oriented fault-sensitive regions to determine the load-sensitive propagation segment. The load-sensitive propagation segment refers to a section in the abnormal vibration propagation path composed of multiple continuous fault-sensitive regions that are not perpendicular to the load direction. The fault-sensitive regions within this segment are continuously arranged in the path structure and are all load-sensitive regions, with no perpendicularly related fault-sensitive regions interspersed within them. If two non-perpendicularly oriented fault-sensitive regions have a perpendicularly related fault-sensitive region, then these two non-perpendicularly oriented fault-sensitive regions belong to different load-sensitive propagation segments.

[0105] Optionally, when sorting out load-sensitive propagation sections, the anomaly diagnosis device needs to follow the arrangement order of fault-sensitive areas in the path structure. Starting from the starting source area of ​​the propagation path, it should check the adjacent areas of each non-vertical fault-sensitive area one by one, and determine whether the adjacent areas are also non-vertical fault-sensitive areas. If the adjacent areas are non-vertical, they are classified into the same load-sensitive propagation section, and the sorting is extended sequentially until a vertically related fault-sensitive area or the end point of the propagation path is encountered. If the adjacent areas of a certain non-vertical fault-sensitive area are all vertically related fault-sensitive areas, then the fault-sensitive area is treated as a separate load-sensitive propagation section.

[0106] Step 403: Based on the positional order of each fault-sensitive region in the load-sensitive propagation segment set in the abnormal vibration propagation path and the vector direction of the load action, determine the load-driven propagation trend.

[0107] Optionally, for each load-sensitive propagation segment, the anomaly diagnosis device first extracts the positional order of each fault-sensitive area within that segment in the abnormal vibration propagation path. This positional order represents the transmission order of abnormal vibration energy within that segment, clearly defining the sequential position of each fault-sensitive area within the segment and from which area the vibration energy is transmitted, ensuring accurate control of the vibration energy propagation direction within the segment. Subsequently, the anomaly diagnosis device determines the vector direction of the load application. The vector direction refers to the specific directional direction of the load application (e.g., forward or backward axially, left or right laterally). Unlike simple direction types, the vector direction accurately characterizes the specific location of the load application, directly calling preset load application direction vector direction data without the need for re-determination.

[0108] Furthermore, the anomaly diagnostic device compares and analyzes the positional order (i.e., the direction of vibration energy transmission) of each fault-sensitive area within the load-sensitive propagation section with the vector pointing in the direction of load action. Combined with the spatial relative relationships of each fault-sensitive area within the section, it determines the load-driven propagation trend of the load-sensitive propagation section. This load-driven propagation trend indicates whether vibration energy is transmitted forward along the load direction, reflected backward, or exhibits deflection and diffusion.

[0109] Forward transmission means that the direction of vibration energy transmission within a section is consistent with the vector direction of the load, and the vibration energy is gradually transmitted along the load direction without any reverse or deflection phenomenon; reverse reflection means that the direction of vibration energy transmission within a section is completely opposite to the vector direction of the load, and the vibration energy is reflected and propagated in the opposite direction of the load direction after encountering an obstacle.

[0110] Deflection diffusion refers to the phenomenon where the direction of vibration energy transmission within a section forms a certain angle with the vector pointing towards the direction of the load. The vibration energy is not transmitted in the forward or reverse direction of the load, but rather deflects in direction, and the vibration energy diffuses during transmission, with the propagation range gradually expanding.

[0111] Optionally, when determining the load-driven propagation trend, a comprehensive judgment should be made in conjunction with the spatial relative relationship: for sections with a parallel spatial relative relationship, the consistency between the direction of vibration energy transmission and the direction of load action vector should be compared to determine whether it is forward transmission or reverse reflection; for sections with an inclined spatial relative relationship, the angle between the direction of vibration energy transmission and the direction of load action vector should be analyzed, and the deflection and diffusion phenomenon should be determined in conjunction with the propagation range of vibration energy.

[0112] Step 404: Based on the sections exhibiting reverse reflection or deflection diffusion in the load-driven propagation trend, abnormal vibration diagnosis is performed to obtain the physical cause of mechanical failure.

[0113] Optionally, the anomaly diagnosis device filters the load-driven propagation trend of all load-sensitive propagation sections, extracts the sections that show reverse reflection or deflection diffusion in the load-driven propagation trend, and removes the sections that show positive transmission in the load-driven propagation trend.

[0114] Optionally, the reason for excluding the positive transmission section is that when vibration energy is transmitted positively along the load direction, its propagation trend conforms to the normal law of vibration energy transmission in mechanical structures. It is a normal vibration energy transmission phenomenon with no abnormalities and no need for abnormal vibration diagnosis.

[0115] The section exhibiting reverse reflection or deflection diffusion shows a vibration energy transmission trend that does not conform to normal patterns, indicating the presence of factors hindering the normal transmission of vibration energy within that section. These factors are structural defects related to mechanical faults and are the core object of abnormal vibration diagnosis.

[0116] Furthermore, based on the selected sections exhibiting reverse reflection or deflection diffusion, abnormal vibration diagnosis is performed to obtain the physical cause of the mechanical failure, as detailed in steps 4041 to 4043.

[0117] The physical causes of mechanical failures identified in this invention accurately pinpoint the structural defects and corresponding mechanical components that cause abnormal vibrations, solving the problems of being unable to distinguish between normal vibration fluctuations and real abnormal vibrations, high false alarm rates, and poor diagnostic reliability, thereby improving the accuracy of abnormal vibration diagnosis in injection molding equipment.

[0118] Optionally, the processes of steps 4041 to 4043 include: Step 4041: Based on the segments exhibiting reverse reflection or deflection diffusion in the load-driven propagation trend, and combined with the branching or converging nodes in the path structure, determine the potential stress concentration nodes.

[0119] Optionally, the anomaly diagnosis device first analyzes each anomaly propagation segment one by one, clarifying the specific location of each anomaly propagation segment in the abnormal vibration propagation path, the fault-sensitive areas it contains, and the direction of vibration energy transmission and abnormal manifestations (reverse reflection or deflection diffusion) within the segment, ensuring accurate understanding of the vibration characteristics and spatial location of each anomaly propagation segment. Subsequently, the anomaly diagnosis device extracts branch nodes and converging nodes in the path structure. A branch node is a node in the abnormal vibration propagation path where multiple branch paths branch off from a main path. This node corresponds to a fault-sensitive area, from which vibration energy is transmitted to multiple branch paths in different directions. A converging node is a node in the abnormal vibration propagation path where multiple branch paths converge to a main path. This node corresponds to a fault-sensitive area, where vibration energy from multiple branch paths in different directions converges and continues to propagate along the main path. Both branch nodes and converging nodes are key nodes in the abnormal vibration propagation path where the direction of vibration energy transmission changes or where energy converges or disperses. They are also areas in the mechanical structure prone to stress concentration.

[0120] Furthermore, each abnormal propagation segment is associated and matched with the branch nodes and intersection nodes in the path structure to determine whether each abnormal propagation segment, as well as its starting and ending points, contains branch nodes or intersection nodes. Specifically, the starting point of an abnormal propagation segment refers to the initial fault-sensitive area where vibration energy enters the segment within the abnormal vibration propagation path; the ending point refers to the final fault-sensitive area where vibration energy leaves the segment within the abnormal vibration propagation path.

[0121] Optionally, the core logic of the matching judgment in this embodiment of the invention is as follows: when vibration energy is reflected or deflected in the abnormal propagation section, the essence is that the transmission of vibration energy is hindered. As key locations where the direction of vibration energy transmission changes or energy converges and disperses, the branch nodes and intersection nodes in the path structure are prone to unbalanced loads on their corresponding mechanical components, which in turn cause stress concentration. Stress concentration will hinder the normal transmission of vibration energy, resulting in the phenomenon of reverse reflection or deflection of vibration energy. Therefore, such nodes are potential stress concentration nodes.

[0122] Optionally, the specific matching process in this embodiment of the invention is as follows: if a certain abnormal propagation section contains a branch node or a junction node, then the node is included in the range of potential stress concentration nodes; if the starting or ending point of the abnormal propagation section is a branch node or a junction node, then the node is also included in the range of potential stress concentration nodes; if neither the abnormal propagation section nor its two ends contain branch nodes or junction nodes, then the location of the abnormal vibration energy transmission in the section is analyzed, and the fault-sensitive area corresponding to the starting position of the reverse reflection or deflection diffusion of vibration energy is regarded as a potential stress concentration node (although this area is not a branch or junction node, the vibration energy transmission is abnormal due to local stress concentration).

[0123] Step 4042: Based on the type of mechanical components connected to each potential stress concentration node and the working state of the kinematic pairs of the injection molding equipment in the current operating stage, determine the node functional failure mode.

[0124] Optionally, the anomaly diagnosis device analyzes each potential stress concentration node one by one to identify the mechanical component corresponding to each potential stress concentration node, that is, the specific mechanical component associated with the node that is used to realize vibration energy transmission and mechanical movement. Each potential stress concentration node corresponds to one or more mutually cooperating mechanical components, and the working state of these components directly affects the vibration transmission function of the node.

[0125] Furthermore, the anomaly diagnosis device extracts the type of mechanical component connected to each potential stress concentration node. The type of mechanical component refers to the specific type of mechanical component associated with the node, including but not limited to bearings, gears, bolts, connecting rods, guide pillars, guide sleeves, etc. Different types of mechanical components have different structural characteristics, working principles and failure modes. The anomaly diagnosis device directly calls the preset association data between nodes and mechanical component types to clarify all types of mechanical components connected to each node.

[0126] Optionally, the anomaly diagnosis device retrieves the working status of the kinematic pairs of the injection molding equipment during the current operating stage. The working status of the kinematic pairs refers to the actual working condition of the kinematic pairs composed of mechanical components connected to potential stress concentration nodes in the injection molding equipment. A kinematic pair refers to a combination of two or more mechanical components that achieve relative movement through contact and cooperation, including but not limited to revolute pairs, prismatic pairs, and rolling pairs. The working status of the kinematic pairs specifically includes normal working status and abnormal working status. Abnormal working status includes motion jamming, abnormal fit clearance, uneven relative movement, etc. This working status is based on the pre-monitoring and storage of the working parameters of the injection molding equipment during the current operating stage.

[0127] Furthermore, the anomaly diagnosis device, based on the type of mechanical component connected to each potential stress concentration node and the corresponding working state of the kinematic pair, determines the node functional failure mode of each potential stress concentration node one by one. The node functional failure mode refers to the specific manifestation of the failure of the vibration energy transmission function and mechanical coordination function of the potential stress concentration node due to anomalies in the connected mechanical components. Specifically, it includes five types: loosening, wear, cracks, abnormal gaps, and preload loss.

[0128] Loosening refers to the slackness of the connection between mechanical components at nodes, making it impossible to achieve a tight fit, resulting in discontinuous vibration energy transmission and thus causing abnormal vibration. Wear refers to the material loss on the surface of mechanical components at nodes due to long-term relative motion, causing changes in the size and shape of the components, affecting the fit accuracy and vibration energy transmission. Cracks refer to the micro-cracks that appear on the surface or inside of mechanical components at nodes due to long-term load and stress concentration. Cracks will lead to a decrease in the strength of the components and hinder the normal transmission of vibration energy. Abnormal clearance refers to the fit clearance between mechanical components at nodes exceeding the preset reasonable range. Both excessively large and excessively small fit clearances will lead to abnormal vibration energy transmission, resulting in reverse reflection or deflection diffusion. Preload loss refers to the reduction of the preload of mechanical components at nodes (such as bolted connections), which cannot meet the preload requirements for normal operation, leading to loose component connections and abnormal vibration transmission.

[0129] Furthermore, the specific process by which the abnormal diagnostic device determines the functional failure mode of a node is as follows: combining the structural characteristics and working principle of the mechanical component type, analyzing the possible functional failure modes of this type of component under the current working state of the kinematic pair; combining the abnormal vibration manifestations (reverse reflection or deflection diffusion) of the abnormal propagation section corresponding to the potential stress concentration node, verifying the rationality of the failure mode, and finally determining the unique node functional failure mode of each potential stress concentration node (if there are multiple possible failure modes, the most fitting one is selected by combining the abnormal vibration manifestations).

[0130] Step 4043: Based on the vibration response characteristics corresponding to each node functional failure mode and the signal propagation characteristics of the corresponding segment in the abnormal vibration propagation path, abnormal vibration diagnosis is performed to obtain the physical cause of mechanical failure.

[0131] Optionally, the vibration response characteristics corresponding to each node functional failure mode refer to the unique characteristics of the vibration signal generated when the mechanical component experiences this type of functional failure, including the variation law of parameters such as vibration amplitude, frequency, and waveform. Based on historical fault data and experimental data, these characteristics are predetermined. The vibration response characteristics corresponding to different node functional failure modes are significantly different and can serve as the core basis for fault diagnosis. The signal propagation characteristics of the corresponding section in the abnormal vibration propagation path refer to the vibration signal transmission characteristics of the abnormal propagation section where the potential stress concentration node is located, including the amplitude attenuation trend, frequency fluctuation range, and waveform variation law of the vibration signal, which are directly related to the abnormal vibration performance (reverse reflection or deflection diffusion) in this area.

[0132] Optionally, the abnormal diagnosis device performs abnormal vibration diagnosis based on the vibration response characteristics corresponding to each node functional failure mode and the signal propagation characteristics of the corresponding segment in the abnormal vibration propagation path, thereby obtaining the physical cause of the mechanical failure, as specifically in steps 40431 to 40433.

[0133] The physical causes of mechanical failures identified in this invention accurately pinpoint the structural defects and corresponding mechanical components that cause abnormal vibrations, solving the problems of being unable to distinguish between normal vibration fluctuations and real abnormal vibrations, high false alarm rates, and poor diagnostic reliability, thereby improving the accuracy of abnormal vibration diagnosis in injection molding equipment.

[0134] Optionally, the processes of steps 40431 to 40433 include: Step 40431: Determine the characteristic matching result based on the vibration response characteristics corresponding to each node functional failure mode and the signal propagation characteristics of the corresponding segment in the abnormal vibration propagation path.

[0135] Optionally, the anomaly diagnosis device first analyzes each potential stress concentration node one by one to identify the node functional failure mode corresponding to each node and the signal propagation characteristics of the anomaly propagation segment where the node is located, ensuring that the failure mode of each potential stress concentration node and the signal propagation characteristics of the corresponding segment can accurately correspond. Subsequently, for each potential stress concentration node, the anomaly diagnosis device extracts the vibration response characteristics corresponding to the node functional failure mode, and identifies the core parameters included in the vibration response characteristics. The core parameters include the vibration amplitude variation range, vibration frequency fluctuation range, vibration waveform characteristics, and phase change law. These parameters can accurately characterize the unique performance of the vibration signal under this failure mode and are the core basis for characteristic matching.

[0136] Optionally, the anomaly diagnosis device extracts the core parameters of the signal propagation characteristics of the abnormal propagation section where the potential stress concentration node is located. The core parameters are completely consistent with the core parameters of the vibration response characteristics, including the amplitude attenuation trend range, frequency fluctuation range, waveform change characteristics, and phase distortion of the vibration signal in the abnormal section. This ensures that the two can be compared and matched in a targeted manner, avoiding matching errors caused by inconsistent parameter types.

[0137] Furthermore, the anomaly diagnostic device compares the core parameters of the vibration response characteristics corresponding to the extracted node functional failure modes with the core parameters of the signal propagation characteristics of the corresponding abnormal propagation sections, parameter by parameter, to determine the characteristic matching result. The characteristic matching result indicates whether the failure mode can explain the observed vibration energy reflection, attenuation anomalies, or phase distortion phenomena in the path, including two situations: qualified matching and unqualified matching. Specifically, abnormal vibration energy reflection refers to the phenomenon of reverse reflection of vibration energy within the abnormal propagation section, opposite to the normal propagation direction; abnormal vibration energy attenuation refers to the attenuation amplitude of vibration energy within the abnormal propagation section exceeding the preset normal range, or the phenomenon of no attenuation but rather enhancement; phase distortion refers to the abnormal phase change pattern of the vibration signal, inconsistent with the phase change pattern of the normal vibration signal, exhibiting phase shifts, phase abrupt changes, etc.

[0138] Optionally, the specific comparison and matching process in this embodiment of the invention is as follows: The corresponding core parameters of the vibration response characteristics and signal propagation characteristics are compared one by one to determine whether the reflection anomalies, attenuation anomalies, or phase distortion phenomena in the signal propagation characteristics can be explained by the vibration response characteristics corresponding to the functional failure mode of that node. If all core parameters correspond to each other, and all types of abnormal phenomena in the signal propagation characteristics can be reasonably explained by the vibration response characteristics, and the parameter deviation is within a preset reasonable range, then the characteristic matching result is determined to be a qualified match; if at least one core parameter does not correspond, or a certain type of abnormal phenomenon in the signal propagation characteristics cannot be reasonably explained by the vibration response characteristics, or the parameter deviation exceeds a preset reasonable range, then the characteristic matching result is determined to be a failed match.

[0139] Step 40432: Based on the node functional failure modes that meet the physical consistency condition in the characteristic matching results, and combined with the load direction's influence mechanism on the mechanical component's stress, determine the dominant failure mechanism.

[0140] Optionally, the anomaly diagnosis device filters all characteristic matching results, extracts node function failure modes with qualified characteristic matching results, and removes node function failure modes with unqualified characteristic matching results. This is because unqualified failure modes cannot explain the abnormal phenomena in signal propagation characteristics, are unrelated to the current abnormal vibration, and do not require further analysis, thus avoiding interference from invalid analysis with the diagnostic results.

[0141] Subsequently, the anomaly diagnosis device clarifies the direction of the load on the injection molding equipment during the current operating stage, as well as the mechanism by which the load direction affects the mechanical components. The mechanism by which the load direction affects the mechanical components refers to the specific stress effect and stress distribution law of the force applied in the direction of the load on the mechanical components connected to potential stress concentration nodes. This mechanism is predetermined and stored based on the mechanical structural characteristics and load transmission principle of the injection molding equipment, clarifying the stress location, stress concentration location, and stress change law of various mechanical components under different load directions.

[0142] Furthermore, the anomaly diagnosis device analyzes each of the selected matching failure modes of the nodes, and, in conjunction with the force influence mechanism of the load direction on the mechanical components, determines the dominant failure mechanism corresponding to each matching failure mode. The dominant failure mechanism characterizes the physical degradation process that leads to the formation of abnormal vibration propagation paths under the load direction, specifically including two types: fatigue crack propagation caused by alternating loads and loosening of connectors caused by impact loads.

[0143] Fatigue crack propagation caused by alternating loads refers to mechanical components with potential stress concentration nodes subjected to periodically changing alternating loads during the current operating phase. Long-term repeated alternating loads cause micro-cracks to form on the surface or inside the component, and the cracks gradually propagate with repeated load application, eventually leading to crack-like functional failure modes, hindering the normal transmission of vibration energy, and forming abnormal vibration propagation paths. Loosening of connectors caused by impact loads refers to mechanical components with potential stress concentration nodes subjected to instantaneous impact loads during the current operating phase. Impact loads cause a decrease in the preload of the connection between components and a loosening of the connection relationship, leading to loosening or preload loss-like functional failure modes, hindering the normal transmission of vibration energy, and forming abnormal vibration propagation paths.

[0144] Optionally, the specific determination process in this embodiment of the invention is as follows: Analyze the specific type of the qualified node function failure mode. If the failure mode is a crack, then, in combination with the load direction's influence mechanism on the mechanical component's stress, determine whether the crack is caused by periodic alternating loads. If so, determine that the dominant failure mechanism is fatigue crack propagation caused by alternating loads. If the failure mode is loosening or preload loss, then, in combination with the load direction's influence mechanism on the mechanical component's stress, determine whether the loosening or preload loss is caused by instantaneous impact loads. If so, determine that the dominant failure mechanism is loosening of the connector caused by impact loads. If the same failure mode can correspond to two dominant failure mechanisms, then, in combination with the abnormal phenomena in the signal propagation characteristics, select the dominant failure mechanism that best matches the abnormal phenomena.

[0145] Step 40433: Based on the dominant failure mechanism and the mechanical structure configuration of the injection molding equipment in the current operating stage, determine the physical cause of the mechanical failure.

[0146] Optionally, the mechanical structure configuration refers to the installation position, connection method, fit relationship, dimensional parameters and material properties of each mechanical component of the injection molding equipment in the current operating stage. This configuration is based on the design standards and structural drawings of the injection molding equipment, which are predetermined and stored. It clarifies the specific structural information of the mechanical components connected to each potential stress concentration node and is the core basis for determining the physical cause of mechanical failure.

[0147] Optionally, the anomaly diagnosis device analyzes each dominant failure mechanism one by one, identifies the node functional failure mode corresponding to each dominant failure mechanism, and the potential stress concentration node corresponding to the failure mode. It further clarifies the type, installation location, and matching relationship of the mechanical components connected to the potential stress concentration node, ensuring that the specific fault-related mechanical components can be accurately located, and avoiding ambiguity in the location of the faulty components.

[0148] Subsequently, the anomaly diagnosis device extracts structural information related to the mechanical components connected to the potential stress concentration node in the mechanical structure configuration of the injection molding equipment during the current operating stage. This includes the material properties, dimensional parameters, connection methods, fit clearance requirements, and preload requirements of the mechanical components. This structural information directly affects the effect of the dominant failure mechanism and determines what type of structural defects the mechanical components will produce under the action of the dominant failure mechanism.

[0149] Furthermore, based on the determined dominant failure mechanism, the anomaly diagnosis device, combined with the structural information of the mechanical components connected to the potential stress concentration node and the corresponding node functional failure mode, determines the physical cause of the mechanical failure corresponding to each potential stress concentration node. Optionally, the specific determination process in this embodiment of the invention is as follows: clarifying the physical degradation process of the dominant failure mechanism, analyzing what type of structural defects the degradation process will cause in the mechanical components under the current mechanical structure configuration; subsequently, combining the specific manifestations of the node functional failure mode, verifying whether the structural defect can lead to the generation of the failure mode and whether it can explain the abnormal phenomena in the signal propagation characteristics; finally, clarifying the mechanical component, defect location, and specific manifestations of the structural defect, forming a complete physical cause of the mechanical failure, ensuring that the physical cause of the mechanical failure can accurately characterize the structural defects generated by the dominant failure mechanism under load and motion conditions, and can comprehensively explain the generation and propagation of abnormal vibrations.

[0150] Optionally, if multiple potential stress concentration nodes correspond to the same dominant failure mechanism and the connected mechanical components have the same type and structural information, then these nodes are determined to have the same type of mechanical failure physical cause and are integrated to form a unified mechanical failure physical cause; if multiple potential stress concentration nodes correspond to different dominant failure mechanisms or the connected mechanical components have different types and structural information, then the mechanical failure physical cause corresponding to each node is determined separately to form multiple independent mechanical failure physical causes.

[0151] The physical causes of mechanical failures identified in this invention accurately pinpoint the structural defects and corresponding mechanical components that cause abnormal vibrations, solving the problems of being unable to distinguish between normal vibration fluctuations and real abnormal vibrations, high false alarm rates, and poor diagnostic reliability, thereby improving the accuracy of abnormal vibration diagnosis in injection molding equipment.

[0152] Furthermore, the abnormal vibration diagnosis device for injection molding equipment based on decision network provided by the present invention will be described below. The abnormal vibration diagnosis device for injection molding equipment based on decision network described below can be referred to in correspondence with the abnormal vibration diagnosis method for injection molding equipment based on decision network described above.

[0153] Optional, refer to Figure 2 , Figure 2This is a schematic diagram of the abnormal vibration diagnosis device for injection molding equipment based on a decision network provided by the present invention. The abnormal vibration diagnosis device for injection molding equipment based on a decision network includes: Vibration deviation calculation module 210 is used to determine the deviation between the actual vibration behavior and the expected vibration characteristics based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor. The vibration deviation judgment module 220 is used to determine whether the normal fluctuation boundary is exceeded based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment in the current operating stage, and to obtain the vibration deviation judgment result. The abnormal vibration monitoring module 230 is used to determine the abnormal vibration propagation path based on each fault-sensitive area associated with the current operating phase and the local vibration coupling relationship corresponding to each area, combined with the signal distribution pattern of the multi-channel vibration signal in the fault-sensitive area if the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary. The abnormal vibration diagnosis module 240 is used to diagnose abnormal vibration based on the path structure of the abnormal vibration propagation path and the direction of the load acting on the injection molding equipment in the current operating stage, so as to obtain the physical cause of mechanical failure.

[0154] The embodiments of the present invention achieve accurate tracing from vibration deviation to the root cause of the fault, rather than simply judging the threshold exceedance. This effectively solves the problems of being unable to distinguish between normal vibration fluctuations and real abnormal vibrations, high false alarm rate, and poor diagnostic reliability, and improves the accuracy of abnormal vibration diagnosis of injection molding equipment.

[0155] Please see Figure 3 , Figure 3 An embodiment diagram of an electronic device provided in accordance with the present invention. For example... Figure 3 As shown, this embodiment of the invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, it implements the processes of steps 10 to 40: Based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor, the deviation between the actual vibration behavior and the expected vibration characteristics is determined. Based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment under the current operating stage, it is determined whether the vibration deviation exceeds the normal fluctuation boundary, and the vibration deviation judgment result is obtained. If the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, then based on each fault-sensitive area associated with the current operation stage and the local vibration coupling relationship corresponding to each area, combined with the signal distribution pattern of multi-channel vibration signals in the fault-sensitive area, the abnormal vibration propagation path is determined. Abnormal vibration is diagnosed based on the path structure of the abnormal vibration propagation path and the direction of the load on the injection molding equipment during the current operation stage, so as to obtain the physical cause of mechanical failure.

[0156] Please see Figure 4 , Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with an embodiment of the present invention is shown. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, it implements the processes of steps 10 to 40. Based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor, the deviation between the actual vibration behavior and the expected vibration characteristics is determined. Based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment under the current operating stage, it is determined whether the vibration deviation exceeds the normal fluctuation boundary, and the vibration deviation judgment result is obtained. If the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, then based on each fault-sensitive area associated with the current operation stage and the local vibration coupling relationship corresponding to each area, combined with the signal distribution pattern of multi-channel vibration signals in the fault-sensitive area, the abnormal vibration propagation path is determined. Abnormal vibration is diagnosed based on the path structure of the abnormal vibration propagation path and the direction of the load on the injection molding equipment during the current operation stage, so as to obtain the physical cause of mechanical failure.

[0157] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the decision network-based abnormal vibration diagnosis method for injection molding equipment provided by the above methods, which includes steps 10 to 40: Based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor, the deviation between the actual vibration behavior and the expected vibration characteristics is determined. Based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment under the current operating stage, it is determined whether the vibration deviation exceeds the normal fluctuation boundary, and the vibration deviation judgment result is obtained. If the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, then based on each fault-sensitive area associated with the current operation stage and the local vibration coupling relationship corresponding to each area, combined with the signal distribution pattern of multi-channel vibration signals in the fault-sensitive area, the abnormal vibration propagation path is determined. Abnormal vibration is diagnosed based on the path structure of the abnormal vibration propagation path and the direction of the load on the injection molding equipment during the current operation stage, so as to obtain the physical cause of mechanical failure.

[0158] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0159] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for diagnosing abnormal vibration in injection molding equipment based on a decision network, characterized in that, include: Based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor, the deviation between the actual vibration behavior and the expected vibration characteristics is determined. Based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment in the current operating stage, it is determined whether the normal fluctuation boundary is exceeded, and the vibration deviation judgment result is obtained. If the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary, then based on each fault-sensitive area associated with the current operation stage and the local vibration coupling relationship corresponding to each area, combined with the signal distribution pattern of the multi-channel vibration signal in the fault-sensitive area, the abnormal vibration propagation path is determined. Based on the path structure of the abnormal vibration propagation path and the direction of the load acting on the injection molding equipment during the current operating stage, abnormal vibration diagnosis is performed to obtain the physical cause of mechanical failure.

2. The abnormal vibration diagnosis method for injection molding equipment based on decision networks according to claim 1, characterized in that, The steps for determining the physical cause of the mechanical failure include: Based on the arrangement order of each fault-sensitive region in the path structure and the direction of the load, the spatial relative relationship is determined; the spatial relative relationship characterizes the spatial orientation characteristics of each fault-sensitive region along the abnormal vibration propagation path relative to the direction of the load; including parallel relationship, perpendicular relationship and inclined relationship; Based on the continuous fault-sensitive regions in the spatial relative relationship that are not perpendicular to the direction of the load action, a set of load-sensitive propagation segments is determined. Based on the positional order of each fault-sensitive region in the load-sensitive propagation segment set in the abnormal vibration propagation path and the vector direction of the load action, the load-driven propagation trend is determined; the load-driven propagation trend indicates that the vibration energy is transmitted in the forward direction, reflected in the reverse direction, or there is a deflection and diffusion phenomenon along the load action direction. Abnormal vibration diagnosis is performed on the sections exhibiting reverse reflection or deflection diffusion in the load-driven propagation trend to obtain the physical cause of the mechanical fault.

3. The abnormal vibration diagnosis method for injection molding equipment based on decision networks according to claim 2, characterized in that, The abnormal vibration diagnosis based on the sections exhibiting reverse reflection or deflection diffusion in the load-driven propagation trend, to obtain the physical causes of the mechanical fault, includes: Based on the segments exhibiting reverse reflection or deflection diffusion in the load-driven propagation trend, and combined with the branching or converging nodes in the path structure, potential stress concentration nodes are identified. Based on the type of mechanical components connected to each potential stress concentration node and the working state of the kinematic pairs of the injection molding equipment in the current operating stage, the functional failure modes of the nodes are determined; the functional failure modes of the nodes include loosening, wear, cracks, abnormal gaps, and loss of preload. Abnormal vibration diagnosis is performed based on the vibration response characteristics corresponding to each node functional failure mode and the signal propagation characteristics of the corresponding segment in the abnormal vibration propagation path to obtain the physical cause of the mechanical failure.

4. The abnormal vibration diagnosis method for injection molding equipment based on decision networks according to claim 3, characterized in that, The abnormal vibration diagnosis, based on the vibration response characteristics corresponding to each node functional failure mode and the signal propagation characteristics of the corresponding segment in the abnormal vibration propagation path, yields the physical cause of the mechanical fault, including: Based on the vibration response characteristics corresponding to each node functional failure mode and the signal propagation characteristics of the corresponding segment in the abnormal vibration propagation path, the characteristic matching result is determined; the characteristic matching result indicates whether the failure mode can explain the observed vibration energy reflection, attenuation anomaly or phase distortion phenomenon in the path. Based on the node functional failure modes that meet the physical consistency condition in the characteristic matching results, and combined with the load direction's influence mechanism on the mechanical component's stress, the dominant failure mechanism is determined. The dominant failure mechanism characterizes the physical degradation process that leads to the formation of abnormal vibration propagation paths under the load direction; including fatigue crack propagation caused by alternating loads and loosening of connectors caused by impact loads. Based on the dominant failure mechanism and the mechanical structure configuration of the injection molding equipment in the current operating stage, the physical causes of the mechanical failure are determined; the physical causes of mechanical failure indicate the existence of a type of structural defect in mechanical components caused by the dominant failure mechanism under load and motion conditions.

5. The method for diagnosing abnormal vibration of injection molding equipment based on decision networks according to claim 1, characterized in that, The steps for determining the abnormal vibration propagation path include: Based on each fault-sensitive region and the corresponding local vibration coupling relationship, a structural connection topology between fault-sensitive regions is constructed. The structural connection topology represents the physical adjacency relationship of each fault-sensitive region on the mechanical structure and the dynamic energy transfer path established through the local vibration coupling relationship. Based on the structural connection topology and the signal distribution pattern of the multi-channel vibration signal in the fault-sensitive area, the regional activity state of each fault-sensitive area is determined; the regional activity state indicates whether the corresponding fault-sensitive area exhibits vibration response characteristics that exceed the normal fluctuation boundary. Based on the fault-sensitive regions whose active states are in abnormal states, and combined with the structural connection topology, a set of candidate regions for anomaly sources is determined. The propagation path of the abnormal vibration is obtained by analyzing the adjacency pointing relationship between the candidate region set of the abnormal source and the structural connection topology.

6. The abnormal vibration diagnosis method for injection molding equipment based on decision networks according to claim 5, characterized in that, The propagation path analysis based on the adjacency pointing relationship between the candidate region set of anomaly sources and the structural connection topology is used to obtain the propagation path of the abnormal vibration, including: Based on the adjacency pointing relationship between the set of candidate regions of anomalies and the structural connection topology, a sequence of local propagation directions starting from each candidate region of anomalies is determined; the sequence of local propagation directions represents the path direction of vibration energy transmission from the candidate region of anomalies along the structural connection topology to the adjacent fault-sensitive region. Based on the temporal phase relationship between the local propagation direction sequence and the multi-channel vibration signal in adjacent fault-sensitive areas, the phase consistency judgment result in each propagation direction is determined; the phase consistency judgment result indicates whether the vibration signal appears sequentially according to the physical connection order of the structural connection topology. Based on the phase consistency judgment results, the propagation path is analyzed to obtain the propagation path of abnormal vibration.

7. The abnormal vibration diagnosis method for injection molding equipment based on decision networks according to claim 6, characterized in that, The propagation path analysis based on the phase consistency judgment result yields the abnormal vibration propagation path, including: Based on the propagation direction that satisfies temporal consistency in the phase consistency judgment result and combined with the structural connection topology, the abnormal vibration propagation chain is determined; the abnormal vibration propagation chain is a linear or branched path structure constructed by multiple fault-sensitive regions connected in chronological order and structural adjacency order. Based on the signal amplitude attenuation trend of each fault-sensitive region in the abnormal vibration propagation chain and the path length in the structural connection topology, the energy attenuation matching degree is determined; the energy attenuation matching degree is used to determine whether the amplitude change of the actual vibration signal along the abnormal vibration propagation chain conforms to the physical law of vibration energy attenuation with propagation distance in mechanical structures. Based on the energy attenuation matching degree, the abnormal vibration propagation chain that satisfies the preset physical attenuation law is determined as the abnormal vibration propagation path.

8. A diagnostic device for abnormal vibration in injection molding equipment based on a decision network, characterized in that, The method for diagnosing abnormal vibrations of injection molding equipment based on decision networks, as described in any one of claims 1 to 7, is applied; the device for diagnosing abnormal vibrations of injection molding equipment based on decision networks comprises: The vibration deviation calculation module is used to determine the deviation between the actual vibration behavior and the expected vibration characteristics based on the expected vibration characteristics of the injection molding equipment in the current operating stage and the multi-channel vibration signals collected by the vibration sensor. The vibration deviation judgment module is used to determine whether the normal fluctuation boundary is exceeded based on the deviation characterization and the normal vibration tolerance range of the injection molding equipment in the current operating stage, and to obtain the vibration deviation judgment result. The abnormal vibration monitoring module is used to determine the abnormal vibration propagation path based on each fault-sensitive area associated with the current operating phase and the local vibration coupling relationship corresponding to each area, combined with the signal distribution pattern of the multi-channel vibration signal in the fault-sensitive area, if the vibration deviation judgment result indicates that it exceeds the normal fluctuation boundary. The abnormal vibration diagnosis module is used to diagnose abnormal vibrations based on the path structure of the abnormal vibration propagation path and the direction of the load acting on the injection molding equipment during the current operating stage, so as to obtain the physical cause of the mechanical failure.

9. An electronic device, comprising: Memory, used to store computer software programs; A processor for reading and executing the computer software program, characterized in that, when the processor executes the computer software program, it implements the abnormal vibration diagnosis method for injection molding equipment based on a decision network as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, wherein a computer software program is stored therein, characterized in that, When the computer software program is executed by the processor, it implements the abnormal vibration diagnosis method for injection molding equipment based on decision networks as described in any one of claims 1 to 7.

Images