阅读说明：本技术 一种液压系统异常监测方法、设备、系统及存储介质 (Hydraulic system abnormity monitoring method, equipment, system and storage medium ) 是由 刘士运 周光志 于 2021-08-04 设计创作，主要内容包括：本发明公开了一种液压系统异常监测方法、设备、系统及存储介质,该方法包括以下步骤：获取多路被监测液压部件工作状态的声音信息,多路所述声音信息由阵列式声音采集器采集；对多路所述声音信息进行处理,并将其合并为一路目标声音信息；将所述目标声音信息输入已训练完成的神经网络模型中,提取得到所述目标声音信息的频率信息；将所述频率信息输入故障识别模型中进行比对,识别被监测液压部件健康状态。其可实现液压系统被监测部件的健康状态识别,识别异常故障,避免出现故障才进行提示造成较大经济损失。(The invention discloses a method, equipment, a system and a storage medium for monitoring the abnormality of a hydraulic system, wherein the method comprises the following steps: acquiring sound information of the working states of a plurality of paths of monitored hydraulic parts, wherein the sound information of the plurality of paths is acquired by an array type sound collector; processing the multiple paths of sound information and combining the multiple paths of sound information into one path of target sound information; inputting the target sound information into a trained neural network model, and extracting frequency information of the target sound information; and inputting the frequency information into a fault identification model for comparison, and identifying the health state of the monitored hydraulic component. The health state identification of the monitored part of the hydraulic system can be realized, abnormal faults can be identified, and great economic loss caused by prompting when faults occur is avoided.)

1. The hydraulic system abnormity monitoring method is characterized by comprising the following steps:

acquiring sound information of the working states of a plurality of paths of monitored hydraulic parts, wherein the sound information of the plurality of paths is acquired by an array type sound collector;

processing the multiple paths of sound information and combining the multiple paths of sound information into one path of target sound information;

inputting the target sound information into a trained neural network model, and extracting frequency information of the target sound information;

and inputting the frequency information into a fault identification model for comparison, and identifying the health state of the monitored hydraulic component.

2. The hydraulic system abnormality monitoring method according to claim 1, wherein the processing and merging the plurality of paths of sound information into one path of target sound information includes:

enhancing the multiple paths of sound signals to obtain multiple paths of enhanced signals;

weighting and adding the multiple paths of enhanced signals to form a pickup beam in the direction of a target signal and attenuate emitted sound in other directions except the direction of the target signal;

and combining all the signals into one path of target sound information.

3. The hydraulic system abnormality monitoring method according to claim 2, wherein the enhancing the plurality of sound signals to obtain a plurality of enhanced signals includes:

and enhancing the multi-path sound signals by adopting a normalized multi-dimensional curve mapping method.

4. The hydraulic system abnormality monitoring method according to claim 2, further comprising, after the step of enhancing the plurality of sound signals to obtain a plurality of enhanced signals:

and removing strong unsteady signals in the multipath enhanced signals by adopting a spatial filtering algorithm.

5. The hydraulic system abnormity monitoring equipment is characterized by comprising a signal receiving unit, a target sound information extraction unit, a multi-dimensional feature extraction unit and a fault identification unit which are sequentially in signal connection,

the signal receiving unit is used for acquiring sound information of the working states of a plurality of paths of monitored hydraulic parts, and the sound information of the plurality of paths is collected by the array type sound collector;

the target sound information extraction unit is used for processing the multiple paths of sound information and combining the multiple paths of sound information into one path of target sound information;

the multi-dimensional feature extraction unit is used for inputting the target sound information into a trained neural network model and extracting frequency information of the target sound information;

and the fault identification unit is used for inputting the frequency information into a fault identification model for comparison and identifying the health state of the monitored hydraulic component.

6. The utility model provides a hydraulic system anomaly monitoring equipment which characterized in that, includes memory and the controller of communication connection in proper order, the memory stores computer program, its characterized in that: the processor is used for reading the computer program and executing the hydraulic system abnormity monitoring method of any one of claims 1-4.

7. A hydraulic system abnormity monitoring system is characterized by comprising a plurality of signal acquisition units and the hydraulic system abnormity monitoring equipment as claimed in claim 5 or 6, wherein the signal acquisition units are arranged in an array.

8. A computer-readable storage medium having instructions stored thereon, characterized in that: performing the hydraulic system anomaly monitoring method of any one of claims 1-4 when said instructions are run on a computer.

Technical Field

The invention belongs to the technical field of hydraulic system abnormity monitoring, and particularly relates to a hydraulic system abnormity monitoring method, equipment, a system and a storage medium.

Background

At present, millions of engineering machines are kept in China. China is the largest capital construction market all over the world, and the world of engineering machinery is the first world in quantity. The gear type parts such as a hydraulic motor, a speed reducer and the like of the engineering machinery are gradually replaced from import to China, and the main maintenance scheme is to replace the whole machine and replace part of the parts. Because the hydraulic motor and the speed reducer belong to rotary machinery and are closed, most faults can be reflected in the operation process, the monitoring is realized by adopting sensor hardware equipment to acquire information, and the alarm or prompt is given only when the equipment fails. When the equipment fails, the alarm is prompted, the machine is stopped, the maintenance cost of the equipment is increased, and the equipment is stopped after the equipment is damaged, so that the economic loss is large.

Disclosure of Invention

The invention provides a method, equipment, a system and a storage medium for monitoring the abnormality of a hydraulic system, which can realize the identification of the health state of a monitored part of the hydraulic system, identify an abnormal fault and avoid great economic loss caused by prompting when the fault occurs.

The invention is realized by the following technical scheme:

the invention provides a hydraulic system abnormity monitoring method, which comprises the following steps:

acquiring sound information of the working states of a plurality of paths of monitored hydraulic parts, wherein the sound information of the plurality of paths is acquired by an array type sound collector;

processing the multiple paths of sound information and combining the multiple paths of sound information into one path of target sound information;

inputting the target sound information into a trained neural network model, and extracting frequency information of the target sound information;

and inputting the frequency information into a fault identification model for comparison, and identifying the health state of the monitored hydraulic component.

This scheme is based on handling sound information and feature extraction, discerns the health status who is monitored hydraulic pressure part to discerning by monitoring hydraulic pressure part unusually, reducing and maintaining the expense, improving and equiping the integrity, just the suggestion leads to the fact great economic loss to the fact in the avoidance of failure.

In a possible design, the processing multiple paths of the sound information and merging the multiple paths of the sound information into one path of target sound information includes:

enhancing the multiple paths of sound signals to obtain multiple paths of enhanced signals;

weighting and adding the multiple paths of enhanced signals to form a pickup beam in the direction of a target signal and attenuate emitted sound in other directions except the direction of the target signal;

and combining all the signals into one path of target sound information.

The scheme realizes target signal extraction based on information enhancement and de-mixing technologies, and can improve the accuracy of target information extraction.

In one possible design, the enhancing the multiple sound signals to obtain multiple enhanced signals includes:

and enhancing the multi-path sound signals by adopting a normalized multi-dimensional curve mapping method.

In high-intensity environmental noise, a traditional noise reduction algorithm cannot easily extract relatively weak fault sound signals. According to the scheme, a normalized multi-dimensional curve mapping method is adopted to enhance the multi-path sound signals, weak abnormal sound signals can be accurately enhanced, complex environmental noise is secondarily and accurately suppressed, and the accuracy of extracting weak fault sound signals is improved.

In one possible design, the enhancing the multiple sound signals to obtain multiple enhanced signals further includes:

and removing strong unsteady signals in the multipath enhanced signals by adopting a spatial filtering algorithm.

The scheme adopts a spatial filtering algorithm, can effectively remove strong unsteady signals in the multipath enhanced signals, and can realize high-intensity accurate suppression on noise and background noise in a complex sound field environment.

The invention provides a hydraulic system abnormity monitoring device, which comprises a signal receiving unit, a target sound information extraction unit, a multi-dimensional characteristic extraction unit and a fault identification unit which are connected in sequence by signals,

the signal receiving unit is used for acquiring sound information of the working states of a plurality of paths of monitored hydraulic parts, and the sound information of the plurality of paths is collected by the array type sound collector;

the target sound information extraction unit is used for processing the multiple paths of sound information and combining the multiple paths of sound information into one path of target sound information;

the multi-dimensional feature extraction unit is used for inputting the target sound information into a trained neural network model and extracting frequency information of the target sound information;

and the fault identification unit is used for inputting the frequency information into a fault identification model for comparison and identifying the health state of the monitored hydraulic component.

A third aspect of the present invention provides a hydraulic system abnormality monitoring apparatus, including a memory and a controller, which are sequentially connected in communication, where the memory stores a computer program, and the processor is configured to read the computer program and execute the hydraulic system abnormality monitoring method according to the first aspect and any one of the possibilities thereof.

The invention provides a hydraulic system abnormity monitoring system, which comprises a plurality of signal acquisition units and hydraulic system abnormity monitoring equipment in the second aspect or the third aspect, wherein the signal acquisition units are arranged in an array.

A fifth aspect of the present invention provides a computer readable storage medium having stored thereon instructions that, when executed on a computer, perform the hydraulic system anomaly monitoring method of the first aspect and any one of its possibilities.

Compared with the prior art, the invention at least has the following advantages and beneficial effects:

this scheme is based on handling sound information and feature extraction, discerns the health status who is monitored hydraulic pressure part to discerning by monitoring hydraulic pressure part unusually, reducing and maintaining the expense, improving and equiping the integrity, just the suggestion leads to the fact great economic loss to the fact in the avoidance of failure.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.

FIG. 1 is a flow chart of a hydraulic system anomaly monitoring method of the present invention.

Fig. 2 shows an audio message of a running part of a subway 3085.

Fig. 3 shows another sound information of a running part of a subway 3085.

Detailed Description

The invention is further described with reference to the following figures and specific embodiments. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto. Specific structural and functional details disclosed herein are merely illustrative of example embodiments of the invention. This invention may, however, be embodied in many alternate forms and should not be construed as limited to the embodiments set forth herein.

It will be understood that when an element is referred to herein as being "connected," "connected," or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may be present. Conversely, if a unit is referred to herein as being "directly connected" or "directly coupled" to another unit, it is intended that no intervening units are present. In addition, other words used to describe the relationship between elements should be interpreted in a similar manner (e.g., "between … …" versus "directly between … …", "adjacent" versus "directly adjacent", etc.).

It should be understood that specific details are provided in the following description to facilitate a thorough understanding of example embodiments. However, it will be understood by those of ordinary skill in the art that the example embodiments may be practiced without these specific details. For example, systems may be shown in block diagrams in order not to obscure the examples in unnecessary detail. In other instances, well-known processes, structures and techniques may be shown without unnecessary detail in order to avoid obscuring example embodiments.

The scheme is suitable for abnormal monitoring of various hydraulic equipment, such as a speed reducer, a motor and the like, the hydraulic equipment is frequently started, braked or changed in speed, the dynamic load change range during operation is large and frequent, the fatigue speed of mechanical key parts such as a bearing, a gear and the like can be greatly accelerated, namely, stress concentration of corresponding contact parts causes the internal lattice structure of a metal material to change, broadband elastic stress waves are emitted outwards, the metal parts vibrate locally or integrally, surrounding air can be synchronously driven to vibrate, and a large number of abnormal sound signals are generated. When the lattice change reaches a certain degree, metal microcracks can be generated, the microcracks can gradually expand along with the operation of a machine, further faults such as stripping, abrasion, cracks, corrosion, abrasion, cracks, gear cracks, broken teeth, collapse and the like of an inner ring or an outer ring of a bearing are caused, and abnormal sound signals with certain characteristic changes are accompanied. Therefore, the related development conditions of the sound source of the fault point such as the bearing, the gear and the like of the key metal parts can be known through analyzing the accompanying abnormal sound, and the health state of the specific target key parts is reflected.

As shown in fig. 1, the present embodiment discloses a hydraulic system abnormality monitoring method, which may be, but is not limited to, executed by a monitoring device, where the monitoring device may be software, or a combination of software and hardware, and the monitoring device may be integrated in an intelligent device such as an intelligent mobile terminal, a tablet, a computer, and the like. Specifically, the monitoring method includes the following steps S01 to S04.

And step S01, acquiring sound information of the working states of the multiple paths of monitored hydraulic parts, wherein the multiple paths of sound information are collected by the array type sound collector. The monitored hydraulic parts can be bearings, gears and other key parts. The sound information is collected by a plurality of sound collectors which are arranged in an array mode, and the sound information can be arranged in a ring mode, a square mode or other arrangement modes. The number of sound collectors is multiple, such as 6, 8, 10 or 16 paths. The sound collector may be a microphone, acoustic sensor or other pickup that is fixedly mounted on the hydraulic component being monitored.

And step S02, processing the multiple paths of sound information and combining the multiple paths of sound information into one path of target sound information. Specifically, the step includes steps S021 to S023.

And S021, enhancing the multiple paths of sound signals to obtain multiple paths of enhanced signals. The sound signal may be enhanced using a variety of techniques, such as using a normalized multidimensional curve mapping method, or using beamforming techniques. Preferably, aiming at the multi-path sound signals of the scheme, a normalized multi-dimensional curve mapping method is adopted for enhancement.

And then removing strong unsteady signals in the multipath enhanced signals by adopting a spatial filtering algorithm, wherein the strong unsteady signals are signals with strong signals and unsteady states.

Step S022, performing weighted addition on the multipath enhanced signals to form a pickup beam in the direction of the target signal, and simultaneously attenuating the emitted sound in the other directions except the direction of the target signal. After the step, the emitted sound in other directions except the direction of the target signal can be effectively attenuated, the interference is effectively reduced, and the accuracy of subsequent frequency feature extraction is improved.

Step S023, the existing signals are merged into one path of target sound information.

And step S03, inputting the target sound information into the trained neural network model, and extracting the frequency information of the target sound information. The neural network model needs to be trained in advance, and the neural network model is trained in advance by a training set. The neural network model can adopt a convolutional neural network, and specifically, a LeNet network, an AlexNet network, a VGGNet network, a NiN network, a GooLeNet network, a ResNet network or a DenseNet network can be selected. Specifically and preferably, the YOLO model is used. When the YOLO model is used for detecting the target to be identified, the method has the advantages of high detection speed, low background false detection rate and strong universality, and can greatly improve the accuracy and efficiency of character target detection.

And step S04, inputting the frequency information into a fault identification model for comparison, and identifying the health state of the monitored hydraulic component.

Taking the case that a surface of a certain element of the bearing is damaged as an example, in the loaded operation process, a damaged point is required to impact the surface of other elements to generate impact pulse force, and the vibration of the system can be excited by the impact generated by the damage. Since the point of damage strikes other components periodically, a periodic pulse force is generated, producing a series of high frequency natural damping vibrations. When the comparison is carried out, the health state identification is realized directly according to the frequency information and the monitored hydraulic component. In particular, in relation to the bearing,

the outer ring fault frequency is:

the inner ring failure frequency is:

the ball failure frequency is:

the retainer touches the inner ring:

the retainer touches the outer ring:

wherein D is the pitch circle diameter, D is the ball diameter, α is the contact angle, z is the number of balls, and R is the rotational frequency of the shaft.

Specifically, as shown in fig. 2 and 3, the collected sound information of a running gear bearing of a subway 3085 vehicle is judged to be suspected of peeling and to have a later-stage severity by the above method.

By adopting the method, the sound information is processed and the characteristics are extracted, and the health state of the monitored hydraulic component is identified, so that the abnormity of the monitored hydraulic component is identified, the maintenance cost is reduced, the integrity of equipment is improved, and the phenomenon that the alarm is given out only when a fault occurs to cause great economic loss is avoided.

The invention provides a hydraulic system abnormity monitoring device, which comprises a signal receiving unit, a target sound information extraction unit, a multi-dimensional characteristic extraction unit and a fault identification unit which are connected in sequence by signals,

the signal receiving unit is used for acquiring sound information of the working states of a plurality of paths of monitored hydraulic parts, and the sound information of the plurality of paths is collected by the array type sound collector; the signal receiving unit may be an interface unit, or other receiving unit.

The target sound information extraction unit is used for processing the multiple paths of sound information and combining the multiple paths of sound information into one path of target sound information.

And the multi-dimensional feature extraction unit is used for inputting the target sound information into the trained neural network model and extracting the frequency information of the target sound information.

And the fault identification unit is used for inputting the frequency information into a fault identification model for comparison and identifying the health state of the monitored hydraulic component.

A third aspect of the present invention provides a hydraulic system abnormality monitoring apparatus, including a memory and a controller, which are sequentially connected in communication, where the memory stores a computer program, and the processor is configured to read the computer program and execute the hydraulic system abnormality monitoring method according to the first aspect and any one of the possibilities thereof. For example, the Memory may include, but is not limited to, a Random-Access Memory (RAM), a Read-Only Memory (ROM), a Flash Memory (Flash Memory), a First-in First-out (FIFO), a First-in First-out (First Input Last Output, FILO), and/or a First-in Last-out (First Input Last Output, FILO), etc.; the processor may not be limited to the use of a microprocessor of the model number STM32F105 family. Furthermore, the computer device may also include, but is not limited to, a power supply unit, a display screen, and other necessary components.

The invention relates to a hydraulic system abnormity monitoring system in a fourth aspect, which comprises a plurality of signal acquisition units and hydraulic system abnormity monitoring equipment in the second aspect or the third aspect, wherein the signal acquisition units are arranged in an array.

The operation principle of the devices and systems in the second to fourth aspects of the present invention is described in detail in the first aspect, and is not described herein again.

A fifth aspect of the present invention is a computer-readable storage medium having stored thereon instructions that, when executed on a computer, perform the hydraulic system anomaly monitoring method of the first aspect and any one of its possibilities. The computer-readable storage medium refers to a carrier for storing data, and may include, but is not limited to, floppy disks, optical disks, hard disks, flash memories, flash disks and/or Memory sticks (Memory sticks), etc., and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: modifications may be made to the embodiments described above, or equivalents may be substituted for some of the features described. And such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.