The present disclosure relates to detecting health issues and predicting failures in operating industrial valves, for preventing safety issues, environmental pollutions, for improving maintenance of the industrial Valves and for preventing or predicting failures.

BACKGROUND OF THE INVENTION

An industrial valve is a critical asset in many different industries. The industrial valve allows operators to control the flows within the industrial pipeline. As a heavy-duty asset, it has a variety of failures and malfunctions. The failures and malfunctions can harm the workflow of the industry and lead to huge losses, safety issues and environmental pollutions.

Methods known in the art of testing an industrial valves periodically examine the valves. The typical average time for examination is every six months. The testing requires an experienced technician for analyzing and testing equipment. The testing may use ultrasonic (Acoustic emission sensors), x-ray, visual examination etc that performs passive sensing. Some other methods make use of statistical algorithms over data that retrieves from valve’s environment sensors. In some cases, the critical industrial valve is frequently replaced for preventing faults. SUMMARY OF THE INVENTION

The term computing device refers herein to a device that includes a processing unit. Examples for such device are a personal computer, a laptop, a server, a tablet a cellular device and IOT (internet of things) device.

The term presenting refers herein to displaying or playing a content on a computing device.

The term health issue refers herein to any change in the condition of the industrial valve that leads to a failure. Such condition may include, for example, corrosion accumulation, erosion accumulation, moving parts friction and torque deviations, cracks, wear of the Gasket and seat of the industrial valve, wear of ball of the industrial valve, wear of the plug of the industrial valve, wear of the gate of the industrial valve, Cavity accumulation, loosening of the bolts and etc.

The term valve and industrial valve are used herein interchangeably.

The term Normal Healthy condition refers herein to a condition in which the industrial valve can adjust the flow of the fluid or gas and can enable or disable the flow of the gas or liquid; by disabling it meant hermetic shut of the industrial valve.

The term failure refers herein to mal functioning of the industrial valve. Examples of such mal functioning are, failure to open the industrial valve, failure to close the industrial valve, failure to adjust the flow of the fluid or gas and internal or external leakages.

Embodiments of the invention provide system and method for monitoring and testing industrial valves. According to some embodiments the system sends acoustic pulses through the industrial valve and collects the data of the waves that are scattered and activated by the acoustic pulses. The system then analyses the data and produces health condition report or measurement reports of the target industrial valve. The report may indicate health issues, or a failure or may indicate that the industrial valve is in normal healthy condition. The report may also provide measurements associated with the industrial valve. In some embodiments, when the system detects health issues or when the system detects a failure the system may generate alerts. In some cases, the system may detect a failure or an incoming failure from the combination of the health issues and the measurement.

The measurements of the industrial valve may include measurements of parameters associated with the status of the industrial valve. Such measurements may include, for example, temperature of the industrial valve, pressure measurement, flow measurements inside the valve and position of internal parts of the valve.

According to some embodiments the system sends an acoustic pulse of one or more frequencies which are associated with a certain health issue. The system then calculates the acoustic signature of the waves that are collected from the emitted, scattered and triggered by the acoustic pulse through the industrial valve.

The acoustic signature is a value that is calculated from the combination of these waves, this value represents a distinguish pattern of the waves, regarding the frequencies, timing and power density of the waves. The system associates a level of the health issue according to results of analysis of the acoustic signature. In some embodiments the system associates a level of health issue according to the deviation of the acoustic signature from the acoustic signature in normal condition.

According to some embodiments, the system associates weight per each health issue and calculates the health condition of the industrial valve by summing the level of all the health issues each multiply by its weight. That is to say, the health condition is calculated from the level of calculated health issues while taking to account the weight of each health issue. If the health condition is below a certain threshold, then the system assumes that the industrial valve is in normal healthy condition. Otherwise, the system may predict a failure if the health condition is above a certain threshold associated with a failure.

The term monitoring zone refers herein to a certain zone in the industrial valve or in the peripheral of the industrial valve which is associated with the tested health issue or with the currently measured parameter associated with the status of the industrial valve.

According to some embodiments the system sends an acoustic pulse of a single frequency or from a range of frequencies. The frequency or the range of the frequencies of the acoustic pulse are associated with a certain measurement (for example with temperature or with pressure of the media inside the valve or the flow of the media inside the industrial valve). The system then calculates the acoustic signature of the waves that are emitted, scattered and triggered by this acoustic pulse through the industrial valve. The system assesses measurements or health status associated with the according analysis of the acoustic signature.

One technical problem dealt with by the present disclosure is how to avoid industrial valve failures. Such failures may cause safety events and environmental pollution and may lead to cumbersome maintenance.

One technical solution is training the system to identify health condition from the waves generated by acoustic pulses and then continuously and or periodically testing the industrial valve by sending acoustic pulses through the industrial valve, analysing the data that is collected from the waves generated by the pulses for detecting health issues in the industrial valve and for predicting or detecting failures and alerting if failure is predicted or detected.

One other technical problem is how to measure parameters associated with the status of the valve. One technical solution is training the system to measure the parameters associated with status of the valve and then continuously and or periodically sending acoustic pulses through the industrial valve, analysing the data that is collected from the waves that are generated by the pulses in accordance with the training and estimating measurements of such parameters.

Such measurement obviates the need for equipment such as manometer or thermometer at the site of the industrial valve.

The system may predict or detect a failure by analysing a combination of several measurements with several health issues.

Such a system enables to continuously and automatically without intervention of a human being, monitor health issues and failures in the industrial valve. Such continuous monitoring inspects failures that are not inspected by periodic tests of human being and thus prevents risks of unexpected failures. Such a system enables to monitor the industrial valve without shutting down the operation of the valve.

Embodiments of the invention disclose an apparatus, the apparatus comprises: an acoustic sensor mounted or embedded on an industrial valve; a transmitter mounted or embedded on the industrial valve; the transmitter is configured for issuing an acoustic pulse; the acoustic pulse being associated with a health issue of the industrial valve or with a measurement of a parameter associated with a status of the industrial valve; a computing device, the computing device being configured for commanding the transmitter for the issuing the at acoustic pulse and for collecting data received from the acoustic sensor; the data being generated from sensing, by the sensor, a wave generated from the acoustic pulse in an inner or in a peripheral part of the industrial valve; wherein the computing device is further configured for detecting the health condition or for measuring the measurement from the data. According to some embodiments at least one transmitter being one member selected from a group consisting of an electro-mechanic transducer, an electrooptic transducer, a speaker and a piezo-electric ceramic. According to some embodiments data being collected- only from sensors that sense the monitoring zone. According to some embodiments the at least one acoustic sensor being one member selected from a group consisted of a sonic sensor, an ultrasonic sensor and a vibration sensor^ strain gauge sensor. According to some embodiments the waveform and the frequency of the acoustic pulse being associated with the health issue or with the measurement.

Embodiments of the invention further disclose a method, the method comprises: triggering a transmitter mounted or embedded on an industrial valve to issue an acoustic pulse, the acoustic pulse being associated with a health issue of the industrial valve or with a measurement of a parameter associated with a status of the industrial valve; collecting data received from the acoustic sensor, the data being generated from sensing by a sensor mounted on the industrial valve, a wave generated from the acoustic pulse in an inner or a peripheral part of the industrial valve; and detecting the health condition or measuring the measurement from the data. According to some embodiments the detecting being by a machine learning model; and further comprising tagging a signature generated from the acoustic pulse with the health issue or measurement and utilizing the tagging for the generating the machine learning model.

According to some embodiments wherein the data collection is being only from sensors that sense the monitoring zone. According to some embodiments the method further comprises predicting a failure in the industrial valve or further comprises detecting a failure in the industrial valve as a result of the detecting the health issues or the measuring the measurement. Embodiments of the invention may be implemented as a computer process (method), a computing system, or as an article of manufacture, such as a computer program product or a non-transitory computer- readable media. The computer program product may be a computer storage media readable by a computer system and encoding a computer program of instructions for executing a computer process on the computer and network devices. The computer program product may also be a propagated signal on a carrier readable by a computing system and encoding a computer program of instructions for executing a computer process.

THE BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE

DRAWINGS

The present disclosed subject matter will be understood and appreciated more fully from the following detailed description taken in conjunction with the drawings in which corresponding or like numerals or characters indicate corresponding or like components. Unless indicated otherwise, the drawings provide exemplary embodiments or aspects of the disclosure and do not limit the scope of the disclosure. In the drawings:

Fig. 1 shows a block diagram of a system for monitoring and testing industrial valves, in accordance with some exemplary embodiments of the subject matter;

Fig. 2 shows a flowchart diagram of a scenario for monitoring and testing industrial valves, in accordance with some exemplary embodiments of the disclosed subject matter;

Fig. 3 shows a flowchart diagram of the training process, in accordance with some exemplary embodiments of the disclosed subject matter;

Fig. 4 shows a flowchart diagram of the analyzing process; in accordance with some exemplary embodiments of the disclosed subject matter;

Fig. 5 shows a spectrogram of the region of interest, in accordance with some exemplary embodiments of the disclosed subject matter; and

Fig. 6 shows a flowchart diagram of an anomality detection analyzing process for predicting health issues, in accordance with some exemplary embodiments of the disclosed subject matter. DETAILED DESCRIPTION

Fig. 1 shows a block diagram of a system for monitoring and testing industrial valves, in accordance with some exemplary embodiments of the subject matter.

System 100 includes one or more transmitters shown in the figure as 109 and 110, one or more sensors shown in the figure as 101 , 102, 103, 104, 105, 106, 107 and 108 and a computing device 111.

The one or more transmitters shown in the figure as 109 and 110 may be electro-mechanic or electrooptic transducer or speaker or piezoelectric ceramic and are mounted over the industrial valve 200 peripheral. In some embodiments the one or more transmitters are embedded in the industrial valve. The mounting or embedding may be for example by gluing, screwed, magnetic mounting etc.

The one or more transmitters 109 and 110 are configured for sending acoustic pulses through the industrial valve for testing the condition of the industrial valve 200 or for measuring parameters associated with the status of the valve. The acoustic pulse may be generated from a variety of waveform and frequencies. The acoustic pulse is transmitted to the inner parts or in a peripheral part of the industrial valve. In some embodiments the acoustic pulse is associated with a certain health issue in some other embodiments the acoustic pulse is associated with a certain measurement of the industrial valve status. The status may include temperature of the valve, pressure of the liquid inside the valve position of internal parts (e.g ball) of the industrial valve and etc.

The acoustic pulses may be transmitted periodically or as a result of a request of the user.

The one or more transmitters 109 and 110 are in communication with the computing device 111. The communication may be via wired or wireless protocol such as the WIFI, ZigBee, Bluetooth etc. The sensors shown in the figure as 101 , 102, 103, 104, 105, 106, 107 and 108 are acoustic sensors such as sonic sensors or ultrasonic sensors or vibration sensors and are mounted over the industrial valve 200 peripheral. In some embodiments the one or more sensors are embedded or mounded on the industrial valve. The mounting or embedding may be for example by gluing, screwing and by magnetic means.

The sensors 101 , 102, 103, 104, 105, 106, 107 and 108 are configured for sensing the waves that are generated by the acoustic pulses in the inner and outer part of the industrial valve.

The sensors 101 , 102, 103, 104, 105, 106, 107 and 108 are in communication with the computing device 111. The communication may be via wired or wireless protocol.

The computing device 111 is configured for continuously and periodically commanding the one or more transmitters 109 and 110 to send the acoustic pulse. The computing device 111 is also configured for receiving from the sensors 101 , 102, 103, 104, 105, 106, 107 and 108 the data collected from sensing the waves.

The computing device 111 is configured for analysing the data that is received from the sensors by utilizing a pre-trained machine learning software. In some embodiments the computing device 111 is configured for analysing data only from sensors that sense the monitoring zone 1 12.

It should be noted monitoring zone 112 of the figure is an example. The computing device 111 is configured for reporting the measurements of parameters associated with status of the valve and the health-condition of the valve and for alerting in case of predicting or detecting failures.

The computing device 111 may communicate with the sensors 101 , 102, 103, 104, 105, 106, 107 and 108 and the one or more transmitters 109 and 110 via wired or wireless communication. In some embodiments the computing device 1 11 may be located nearby the industrial valve 200 such that it can be wired to the one or more transmitters 109 and 110. In some other embodiments the computing device 111 may be connected to a cloud service; In such a case the computing device 111 may issue the pulse and collects the data from the sensors 101 , 102, 103, 104, 105, 106, 107 and 108 while the analysis or portions of the analysis may be performed in a remote server (not shown in the figure).

The industrial valve 200 are used in many industries. In some cases the industrial valve is a control valve. Examples for industrial valve 200 may be double ported globe valve, globe valve, gate valve, diaphragm valve, butterfly valve, ball valve, disk valve, isolation valve, safety valve, check valve, control valve, on/off valve, pressure relief valve, manual or automatic or actuated valve or for any other industrial valve. It should be noted that though the figure depicts a globe valve the depicted globe valve is exemplary in nature and may be replaced by any other type of industrial valve.

Fig. 2 shows a flowchart diagram of a scenario for monitoring and testing industrial valves in accordance with some exemplary embodiments of the disclosed subject matter.

At block 210 the system initiates a periodic test by commanding the transmitter to send acoustic pulse through the industrial valve. As a result, the pulse waves are resonating through the industrial valve.

In some embodiments the acoustic pulses are associated with a certain health issue. In some other embodiment the acoustic pulses are associated with a certain measurement.

At block 220 the sensors sense the waves that resonate in the industrial valve due to the acoustic pulse.

At block 230 the sensors transmit the data collected from the waves to the computing device. At block 235 the computing device performs a pre process in which the data is prepared for the analysis. In some embodiments the preparation may include cropping, filtering (noise etc), Digital Signal Processing such as Fourier transform (FT) and the like.

At block 240 the system analyzes the data by applying Al (artificial intelligence) algorithm. Such Algorithm may utilize machine learning trained models. The analysis is for detecting health issues or for measuring parameters associate with the status of the industrial valve. The system predicts failures or detects failures from the detected health issue or from a combination of detected health issues and/or measurements. The process of analyzing is expressed in greater details in fig. 4. It should be noted that the algorithm may be implemented partially by a remote server.

At block 245 the system reports the predicted or detected health issues. In some other embodiments the system reports measurement results.

Fig. 3 shows a block diagram of the training process, in accordance with some exemplary embodiments of the disclosed subject matter.

The system is trained to detect the deficiencies in the industrial valve in various health issues and measurements.

The training is done by sending acoustic pulses to the industrial valve for various health issue of the valve and for various measurements, tagging the signature with the associated health issue or measurement and by generating a machine learning model.

Referring now to the drawing:

Blocks 300-325 describe the process of preparing a data set for a certain health issue or for a certain measurement and for a certain industrial valve. Such a process can be performed for any of the health issue or measurement and for any industrial valve. At block 300 the system selects the heath issue or the measurement for which the training is performed for a certain industrial valve.

At block 305 the system issues an acoustic pulse of one or more frequencies which are associated with the selected heath issue or with the selected measurement.

At block 310 the system records the acoustic pulse that are sensed by the sensors. In some embodiments the system records data only from the monitoring zone associated with the currently trained health issue or from the currently trained measurement.

At block 315 the system generates a signature from the acoustic pulse. The system tags the signature with the real value. The real value is corresponding to the trained health issue or the trained measurement.

At block 325 the system prepares the data set in accordance with the tagging by cleaning, performing DSP (digital signal processing) such as Fourier transform, filtering etc.

At block 330 the system utilizes the data set to train an Al (Artificial Intelligence) machine learning model. The output of the training is a trained machine learning model. The training may be performed, for example, by any combination of models such as prediction, anomaly detection, classification, regression and clustering. The models are selected according to the tested health issues or the desired measurement.

In one example the anomaly detection is applied based on deviations from the resonance frequency and from the acoustic signature of the normal healthy mode. In some embodiments the deviation is calculated per each health issue. In the case of measuring health condition, the system may apply a score and weight for every detected health issue and calculates a total score from the score and weight of each health issue. The system may predict a probability for failure according to the total score and may send an alert to the operators if the total score accedes a threshold. In some embodiments the system predicts probability of a failure to occur.

Every rigid physical structure has normal mode. In normal mode there is a set of unique frequencies that the structure can vibrate. The value of those frequencies depends on parameters like, material shape, dense, inner structure. A defected industrial valve may change its shape, dense and inner structure causing a change in the acoustic signature.

Fig. 4 shows a flowchart diagram of the analyzing process; in accordance with some exemplary embodiments of the disclosed subject matter.

At block 410 the system selects the current test that is to be performed. Such a test may be, for example, a test of a certain health issue or of a certain measurement.

At block 415 the system issues the acoustic pulse that is associated with this test, the system associates one or more acoustic pulse per each measurement.

At block 420 the system collects the data of the waves that are scattered and activated by the acoustic pulses. Example of such data is depicted in figure 5.

At block 425 the system applies the machine learning model that was generated for this test by the training process. The system operates the machine learning process that was generated for the selected heath issue or for the selected measurement on the collected data.

At block 435 the system analyses the output of the machine learning process.

At block 440 which is performed if system tests measurement the system presents the assessed measurement to the user. For example, if the position of the ball is measured the system presents the current estimated position of the ball. At block 445 which is performed if the system tests health issue the system checks if the health issue is above a threshold if so, the system predicts a coming fault or detects an existence fault according to the threshold. If a fault is detected or if an incoming fault is detected, the system presents the fault to the user.

In some embodiments the system analyses a combination plurality of results of a plurality of health issues and/or measurements for detecting or predicting a fault.

The process described in this figure is repeated periodically for all the health issues and measurements associated with a certain industrial valve.

Fig. 5 shows a spectrogram of the region of interest, in accordance with some exemplary embodiments of the disclosed subject matter.

Fig. 6 shows a flowchart diagram of an anomality detection analyzing process for predicting health issues; in accordance with some exemplary embodiments of the disclosed subject matter.

At block 610 the system collects the data that is received from the sensors as a result of issuing various acoustic pulses each associated with a certain health issue.

At block 615 the system determines, from the collected data, tested acoustic signature per each tested health issue.

At block 620 the system calculates the deviation of the tested acoustic signature of each tested health issue from the acoustic signature of this health issue in normal condition.

At block 625 the system associates a level of the failure for each health issue according to this deviation.

At block 630 the system calculates the health condition of the industrial valve by summing the level of all the health issues each multiply by its associated weight. At block 635 if the health condition exceeds a certain value the system predicts a failure wherein the level of prediction may be determined according to the value of health condition.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

It should be noted that, in some alternative implementations, the functions noted in the block of a figure may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.