WO2021027213A1 - 检测方法、装置、电子设备和计算机可读介质 - Google Patents
检测方法、装置、电子设备和计算机可读介质 Download PDFInfo
- Publication number
- WO2021027213A1 WO2021027213A1 PCT/CN2019/126217 CN2019126217W WO2021027213A1 WO 2021027213 A1 WO2021027213 A1 WO 2021027213A1 CN 2019126217 W CN2019126217 W CN 2019126217W WO 2021027213 A1 WO2021027213 A1 WO 2021027213A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- health
- detection
- data
- historical
- equipment
- Prior art date
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 182
- 230000036541 health Effects 0.000 claims abstract description 196
- 238000000034 method Methods 0.000 claims abstract description 46
- 238000012549 training Methods 0.000 claims abstract description 33
- 238000012423 maintenance Methods 0.000 claims abstract description 27
- 238000007689 inspection Methods 0.000 claims abstract description 26
- 238000010801 machine learning Methods 0.000 claims abstract description 25
- 238000004519 manufacturing process Methods 0.000 claims abstract description 20
- 238000004590 computer program Methods 0.000 claims description 8
- 238000001228 spectrum Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 description 11
- 238000012545 processing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000012360 testing method Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000003129 oil well Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B47/00—Survey of boreholes or wells
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
Definitions
- the embodiments of the present application relate to the field of data processing, and specifically relate to detection methods, devices, electronic equipment, and computer-readable media.
- the oil field moving equipment refers to the rotating equipment (that is, the equipment that consumes energy) driven by the driving machine, such as pumps, compressors, fans, etc., suitable for oil fields.
- the energy source can be electric power, pneumatic power, steam power, etc.
- the fault diagnosis of oilfield equipment is usually based on the fluctuation of sensor signals to artificially judge the health of oilfield equipment, and then perform predictive maintenance on oilfield equipment. Since health depends on human experience, and experience cannot be quantified, the accuracy of health detection of oilfield equipment is low, which leads to the inability to accurately perform predictive maintenance on oilfield equipment.
- the embodiments of the present application propose detection methods, devices, electronic equipment, and computer-readable media to solve the technical problem of low accuracy in detecting the health of oilfield equipment in the prior art.
- an embodiment of the present application provides a detection method.
- the method includes: obtaining operating data to be tested during production or self-inspection of oilfield equipment; and inputting the operating data to be tested into a pre-built health detection model to obtain The health detection results of the oilfield equipment.
- the health detection model uses the historical operation data and historical detection data of the oilfield equipment as training samples, and is trained using machine learning methods to detect the health of the oilfield equipment. Degree detection model.
- an embodiment of the present application provides a detection device, which includes: a first acquisition unit configured to acquire operating data to be tested during production or self-inspection of oilfield equipment; and the detection unit is configured to The operation data to be tested is input into the pre-built health detection model to obtain the health detection results of the oilfield equipment.
- the health detection model uses the historical operation data and historical detection data of the oilfield equipment as training samples and uses machine learning methods. A trained health detection model used to detect the health of oilfield equipment.
- an embodiment of the present application provides a device, including: one or more processors; a storage device, on which one or more programs are stored, when one or more programs are executed by one or more processors , Enabling one or more processors to implement the method described in the first aspect.
- an embodiment of the present application provides a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the method described in the first aspect is implemented.
- the health detection model is a health detection model that uses historical operating data and historical detection data of oilfield equipment as training samples, and is trained using machine learning methods to detect the health of oilfield equipment. Therefore, the health of oilfield equipment can be detected through the health detection model, which can improve the accuracy of the health detection of oilfield equipment and provide strong technical support for predictive maintenance.
- Fig. 1 is a flowchart of an embodiment of the detection method according to the present application.
- FIG. 2 is a flowchart of another embodiment of the detection method according to the present application.
- Fig. 3 is a schematic structural diagram of an embodiment of a detection device according to the present application.
- Fig. 4 is a schematic structural diagram of a computer system suitable for implementing the device of the embodiment of the present application.
- FIG. 1 shows a process 100 of an embodiment of the detection method according to the present application.
- the detection method includes the following steps:
- Step 101 Obtain the to-be-tested operating data of the oilfield equipment during production or self-test.
- the execution body of the detection method can obtain the to-be-detected operating data of the oilfield equipment during production or self-inspection.
- the oil field moving equipment refers to the rotating equipment (that is, the equipment that consumes energy) driven by a driving machine, such as pumps, compressors, fans, etc., suitable for oil fields.
- the energy source can be electric power, pneumatic power, steam power, etc.
- multiple sensors are installed on the above-mentioned oilfield equipment.
- the aforementioned multiple sensors may include but are not limited to: temperature sensors, pressure sensors, ammeters, voltmeters, sound signal collectors, and so on.
- the execution body can periodically or in real time collect the to-be-detected operating data of the oil field moving equipment during production or self-inspection through the above-mentioned multiple sensors.
- the above-mentioned execution subject may be the sensors in the above-mentioned oilfield equipment through RTU (Remote Terminal Unit, remote terminal control system), DCS (Distributed Control System, distributed control system), or PLC (Programmable Logic Controller, programmable The logic controller) and the like are in communication connection with the above-mentioned execution subject.
- RTU Remote Terminal Unit
- DCS Distributed Control System
- PLC Programmable Logic Controller, programmable The logic controller
- the foregoing operating data to be detected may include, but is not limited to, at least one of the following: spectrum, audio, current, voltage, flow, temperature, and pictures.
- the above-mentioned execution subject may also save the acquired operating data to be tested in a database, so as to provide data support for subsequent failure analysis.
- Step 102 Input the running data to be detected into a pre-built health detection model to obtain the health detection result of the oilfield equipment.
- the above-mentioned running data to be tested is input into a pre-built health detection model to obtain the health detection result of the above-mentioned oilfield equipment.
- the above-mentioned health detection model is a health detection model that uses the historical operation data and historical detection data of the above-mentioned oilfield equipment as training samples and is trained by a machine learning method to detect the health of the above-mentioned oilfield equipment.
- the historical operation data may be the data generated during the historical operation or self-inspection of the aforementioned oilfield equipment.
- the specific items (such as frequency spectrum, audio frequency, voltage, current, temperature, etc.) of the foregoing historical operating data and the foregoing operating data to be detected may be the same, and will not be repeated here.
- the historical detection data may be data generated during periodic inspections of the equipment and failures. This data can be manually recorded by the inspector.
- the historical detection data may include a fault identifier used to indicate whether the oilfield equipment is malfunctioning, the type of failure, the amount of alarm, and the service life or months of the oilfield equipment.
- the oil field moving equipment can be inspected at regular intervals (for example, one day, one week, etc.) to determine whether it is malfunctioning, and the inspection data is recorded for each inspection.
- the health degree detection result may include the value of the health degree.
- the health level may be a value within a preset value interval (for example, [0, 10]). The greater the value of health, the better the operating condition of the oilfield equipment.
- the aforementioned health detection model may be pre-trained and generated. Before training the health detection model, you can first use the acquired historical detection data to label the historical operating data. Therefore, the historical operating data can be used as input, and the input historical operating data can be used as output, and a machine learning method (such as a supervised learning method) can be used to train to obtain a health detection model.
- a machine learning method such as a supervised learning method
- the health detection model can be trained using various commonly used models for processing big data. For example, models such as Convolutional Neural Networks (CNN).
- CNN Convolutional Neural Networks
- various preset labeling rules can be used to label historical operating data. As an example, after the detection of oilfield equipment during a certain period of time, if the detection result indicates that the oilfield equipment has failed, the historical operating data obtained during the period can be marked as 1; if the detection result indicates the oilfield equipment If no failure occurs, the historical operating data obtained during the time period can be marked as 0.
- each historical operating data can be input one by one, or multiple historical operating data can be input each time in batches.
- the model training can be determined as the health detection model.
- the health of oilfield equipment can be detected in real time through the health detection model.
- the above-mentioned execution subject may further execute the following steps:
- the above-mentioned first correspondence table is used to represent the correspondence between the health level and the health degree range.
- different health levels for example, A, B, C
- the health level corresponding to level A is (8, 10]
- the health level corresponding to level B is (6, 8)
- the health level corresponding to level C is [0, 6].
- the above-mentioned second correspondence table is used to characterize the correspondence between health levels and maintenance methods.
- the maintenance method corresponding to level A is to record real-time operating data
- the maintenance method corresponding to level B is manual inspection
- the maintenance method corresponding to level C is to suspend the work of oilfield equipment and conduct manual inspection.
- the execution subject may select a maintenance method based on the health level corresponding to the detected health. Furthermore, the selected maintenance method is used to maintain the above-mentioned oilfield equipment. Since the maintenance method is selected according to the health level, rather than determined by the experience of the maintenance personnel, the accuracy of the predictive maintenance can be improved.
- the method provided in the above-mentioned embodiment of the present application obtains the operating data to be tested during production or self-inspection of the oilfield equipment, and then inputs the above-mentioned operating data to the pre-built health detection model to obtain the performance of the oilfield equipment.
- Health test results are a health detection model for detecting the health of the above-mentioned oilfield equipment, which is obtained by using machine learning methods to train historical operation data and historical detection data of the oilfield equipment as training samples. Therefore, the health of oilfield equipment can be detected through the health detection model, which can improve the accuracy of the health detection of oilfield equipment and provide strong technical support for predictive maintenance.
- FIG. 2 shows the flow 200 of another embodiment of the detection method.
- the process 200 of the detection method includes the following steps:
- Step 201 Obtain the to-be-tested operating data of the oilfield equipment during production or self-test.
- the above-mentioned oilfield equipment may be equipped with multiple sensors.
- the above-mentioned execution subject may periodically or in real time collect the to-be-detected operating data of the oilfield equipment during production or self-inspection through the above-mentioned multiple sensors.
- the above-mentioned operating data may include, but is not limited to: spectrum, audio, current, voltage, flow, temperature, and pictures.
- Step 202 Input the running data to be detected into a pre-built health detection model to obtain the health detection result of the oilfield equipment.
- the above-mentioned running data to be detected by the execution subject is input into a pre-built health detection model to obtain the health detection result of the above-mentioned oilfield equipment.
- the above-mentioned health detection model can be obtained by training in the following steps:
- the first step is to obtain historical operation data and historical inspection data of the above-mentioned oilfield equipment.
- the second step is to set the historical health of the aforementioned oilfield equipment based on the aforementioned historical detection data.
- the historical health degree may be a value located in a preset value interval (for example, [0, 10]). The greater the value of health, the better the operating condition of the oilfield equipment.
- the historical health degree can be determined by combining the fault identification, fault type, and alarm amount in the historical detection data. See the table below for details:
- the fault indicator is 1, it means that the oil field equipment is malfunctioning; if the fault indicator is 0, it means the oil field equipment is not malfunctioning.
- the second column in the table Take the second column in the table as an example.
- the third step is to establish the correspondence between each historical operating data and historical health.
- each historical operating data For example, if the device is inspected once in each inspection cycle (for example, 24 hours), the historical health level corresponding to the historical detection data recorded after each inspection can be obtained. Furthermore, it is possible to establish a corresponding relationship between the historical operation data collected in each inspection period and the historical health degree corresponding to the historical period.
- the fourth step is to use each historical operating data as the input of the pre-established initial model, and use the historical health corresponding to the input historical operating data as the output of the above-mentioned initial model, and use the machine learning method to train to obtain the above-mentioned oil field A health detection model for the health of mobile equipment.
- Step 203 When the oilfield equipment fails, obtain the operating data of the oilfield equipment during the failure, and obtain the detection data of the oilfield equipment after the failure.
- the above-mentioned executive body when the above-mentioned oil field power equipment fails, can obtain the operating data of the above-mentioned oil field power equipment during the failure period, and obtain the detection data of the above-mentioned oil field power equipment after the failure.
- the above-mentioned executive body may also save the above-mentioned operating data in a database, so as to provide data support for subsequent failure analysis.
- Step 204 Use the running data and the detection data as training samples, and use a machine learning method to train the health detection model to update the health detection model.
- the execution subject may use the operation data and the detection data as training samples, and use a machine learning method to train the health detection model to update the health detection model.
- the model can be updated according to the following steps:
- the target health of the oil field equipment during the failure period is set, and the corresponding relationship between the operating data and the target health is established.
- the setting method of the target health level can refer to the setting method of the historical health level in step 202, which will not be repeated here.
- the above-mentioned operating data is used as the input of the above-mentioned health detection model, and the above-mentioned target health is used as the output of the above-mentioned health detection model.
- Machine learning methods are used to train the above-mentioned health detection model to update the above-mentioned health detection model. parameter.
- the flow 200 of the detection method in this embodiment involves the step of retraining the model by using the operating data of the oilfield equipment when a fault occurs. Therefore, the solution described in this embodiment can optimize the model in real time, and improve the accuracy of the model.
- this application provides an embodiment of a detection device.
- the device embodiment corresponds to the method embodiment shown in Fig. 1, and the device can be specifically applied Used in various electronic devices.
- the detection device 300 described in this embodiment includes: a first acquisition unit 301 configured to acquire operating data to be tested during production or self-inspection of oilfield equipment; and the detection unit 302 is configured to The above-mentioned operational data to be tested is input into the pre-built health detection model to obtain the health detection results of the above-mentioned oilfield equipment, wherein the above-mentioned health detection model uses the historical operation data and historical detection data of the above-mentioned oilfield equipment as training samples, A health detection model trained by a machine learning method for detecting the health of the above-mentioned oilfield equipment.
- the above-mentioned oilfield dynamic equipment is equipped with multiple sensors; and the above-mentioned first acquisition unit is further configured to collect the oilfield dynamics periodically or in real time through the above-mentioned multiple sensors. The running data to be tested when the equipment is in production or self-inspection.
- the above-mentioned health detection model is obtained through training in the following steps, including: obtaining historical operation data and historical detection data of the above-mentioned oilfield equipment; and setting the above-mentioned oilfield based on the above-mentioned historical detection data.
- the historical health of equipment establish the correspondence between each historical operating data and historical health; use each historical operating data as the input of the pre-established initial model, and use the historical health corresponding to the input historical operating data as the initial
- the output of the model is trained using a machine learning method to obtain a health detection model for detecting the health of the above-mentioned oilfield equipment.
- the above-mentioned device further includes: a first obtaining unit configured to obtain the operating data of the above-mentioned oilfield equipment during the failure period when the above-mentioned oilfield equipment fails, and obtain The detection data of the above-mentioned oilfield equipment after a failure; the updating unit is configured to use the above-mentioned operating data and the above-mentioned detection data as training samples, and use a machine learning method to train the above-mentioned health detection model to update the health detection model.
- the update unit is further configured to: based on the detection data, set the target health of the oilfield equipment during the failure period, and establish the operating data and the above Corresponding relationship of target health; the above operating data is used as the input of the above health detection model, the above target health is used as the output of the above health detection model, and machine learning methods are used to train the above health detection model to update the above The parameters of the health detection model.
- the above-mentioned device further includes: a first searching unit configured to search for the current health level indicated by the above-mentioned health level detection result from a preset first correspondence table Corresponding target health level, wherein the above-mentioned first correspondence table is used to characterize the correspondence relationship between the health level and the health degree range; the second searching unit is configured to look up the above-mentioned target health from a preset second correspondence table The target maintenance mode corresponding to the level, wherein the above-mentioned second correspondence table is used to characterize the corresponding relationship between the health level and the maintenance mode; the maintenance unit is configured to use the above-mentioned target maintenance mode to maintain the oilfield equipment.
- a first searching unit configured to search for the current health level indicated by the above-mentioned health level detection result from a preset first correspondence table Corresponding target health level, wherein the above-mentioned first correspondence table is used to characterize the correspondence relationship between the health level and the health degree range
- the second searching unit is configured to look
- the foregoing operating data to be detected includes at least one of the following: spectrum, audio, current, voltage, flow, temperature, and pictures.
- the foregoing detection device includes a processor and a memory, and the foregoing first acquisition unit, detection unit, etc. are all stored in the memory as a program unit, and the processor executes the foregoing program unit stored in the memory to implement corresponding functions.
- the processor contains the kernel, which calls the corresponding program unit from the memory.
- the kernel can be set to one or more, and the accuracy of the health detection of oilfield equipment can be improved by adjusting the kernel parameters, which provides strong technical support for predictive maintenance.
- the embodiment of the present invention provides a computer-readable medium on which a program is stored, and the program is executed by a processor to implement the detection method.
- the embodiment of the present invention provides a processor, the processor is configured to run a program, wherein the detection method is executed when the program is running.
- the embodiment of the present invention provides a device that includes at least one processor, and at least one memory and a bus connected to the processor; wherein the processor and the memory communicate with each other through the bus; the processor is used to call Program instructions to execute the above detection method.
- the devices in this article can be servers, PCs, PADs, mobile phones, etc.
- This application also provides a computer program product, which when executed on a data processing device, is suitable for executing a program that initializes the following method steps: Obtain the to-be-tested operating data of the oilfield equipment during production or self-test; The detection operation data is input into the pre-built health detection model to obtain the health detection results of the above-mentioned oilfield equipment.
- the above-mentioned health detection model uses the historical operation data and historical detection data of the above-mentioned oilfield equipment as training samples and uses the machine A health detection model trained by the learning method to detect the health of the above-mentioned oilfield equipment.
- the above-mentioned oilfield dynamic equipment is equipped with multiple sensors; and the above-mentioned obtaining the to-be-tested operating data of the oilfield dynamic equipment during production or self-inspection includes: periodically or in real time collecting the oilfield dynamic equipment through the above-mentioned multiple sensors Running data to be tested during production or self-inspection.
- the above-mentioned health detection model is obtained by training in the following steps, including: obtaining historical operation data and historical detection data of the above-mentioned oilfield equipment; based on the above-mentioned historical detection data, setting the historical health of the above-mentioned oilfield equipment; Correspondence between historical operating data and historical health; each historical operating data is used as the input of the pre-established initial model, and the historical health corresponding to the input historical operating data is used as the output of the above-mentioned initial model, and the machine learning method is used for training , To obtain a health detection model for detecting the health of the above-mentioned oilfield equipment.
- a data processing device when executed on a data processing device, it may also be adapted to further execute a program that initializes the following method steps: when the above-mentioned oilfield equipment fails, obtain the operating data of the above-mentioned oilfield equipment during the failure period, and Obtain the detection data of the above-mentioned oilfield equipment after a failure; use the above-mentioned operating data and the above-mentioned detection data as training samples, and use a machine learning method to train the above-mentioned health detection model to update the health detection model.
- the above-mentioned operation data and the above-mentioned detection data are used as training samples, and the above-mentioned health detection model is trained by a machine learning method to update the health detection model, including: setting the above-mentioned oilfield equipment based on the above-mentioned detection data Target health during the failure period, and establish the corresponding relationship between the operating data and the target health; use the operating data as the input of the health detection model, and use the target health as the output of the health detection model , Use a machine learning method to train the above-mentioned health detection model to update the parameters of the above-mentioned health detection model.
- the health detection result of the above-mentioned oilfield equipment may be further adapted to further execute a program initialized with the following method steps: searching for the health detection result from the preset first correspondence table The indicated target health level corresponding to the current health level, wherein the above-mentioned first correspondence table is used to characterize the corresponding relationship between the health level and the health level range; from the preset second correspondence table, search for the corresponding target health level The target maintenance mode of the above-mentioned oilfield equipment, wherein the above-mentioned second correspondence table is used to characterize the corresponding relationship between the health level and the maintenance mode; the above-mentioned target maintenance mode is used to maintain the oilfield equipment.
- the foregoing operating data to be detected includes at least one of the following: spectrum, audio, current, voltage, flow, temperature, and pictures.
- the device includes one or more processors (CPUs), memory, and buses.
- the device may also include input/output interfaces, network interfaces, and so on.
- the memory may include non-permanent memory in computer-readable media, random access memory (RAM) and/or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM), and the memory includes at least one Memory chip.
- RAM random access memory
- ROM read-only memory
- flash RAM flash memory
- the memory is an example of a computer-readable medium.
- Computer-readable media include permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology.
- the information can be computer-readable instructions, data structures, program modules, or other data.
- Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disc (DVD) or other optical storage, Magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media can be used to store information that can be accessed by computing devices. According to the definition in this article, computer-readable media does not include transitory media, such as modulated data signals and carrier waves.
- this application can be provided as methods, systems, or computer program products. Therefore, this application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
- computer-usable storage media including but not limited to disk storage, CD-ROM, optical storage, etc.
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Geology (AREA)
- Mining & Mineral Resources (AREA)
- Computational Mathematics (AREA)
- Environmental & Geological Engineering (AREA)
- Algebra (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Fluid Mechanics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Geochemistry & Mineralogy (AREA)
- Pure & Applied Mathematics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Geophysics (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
一种检测方法,包括:获取油田动设备在生产或自检时的待检测运行数据(101);将待检测运行数据输入预先构建的健康度检测模型,得到油田动设备的健康度检测结果(102),其中,健康度检测模型为将油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测油田动设备的健康度的健康度检测模型。该方法能够提高对油田动设备健康度检测的准确性,为预测性维护提供技术支撑。还提供一种装置、电子设备和计算机可读介质。
Description
本申请要求在2019年8月13日提交中国专利局、申请号为201910746127.0、发明名称为“检测方法、装置、电子设备和计算机可读介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请实施例涉及数据处理领域,具体涉及检测方法、装置、电子设备和计算机可读介质。
各油田在油井生产过程中,通常需要使用各种油田动设备。其中,油田动设备是指适用于油田的、由驱动机带动的转动设备(亦即有能源消耗的设备),如泵、压缩机、风机等。其能源可以是电动力、气动力、蒸汽动力等。
目前对油田动设备的故障诊断,通常是通过传感器信号的波动情况人为来判断油田动设备的健康度,进而对油田动设备进行预测性维护。由于健康度依人的经验而定,且经验无法量化,因而,对油田动设备健康度检测的准确性较低,进而导致无法准确地对油田动设备进行预测性维护。
发明内容
本申请实施例提出了检测方法、装置、电子设备和计算机可读介质,以解决现有技术中对油田动设备的健康度检测的准确性较低的技术问题。
第一方面,本申请实施例提供了一种检测方法,该方法包括:获取油田动设备在生产或自检时的待检测运行数据;将待检测运行数据输入预先构建的健康度检测模型,得到油田动设备的健康度检测结果,其中,健康度检测模型为将油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测油田动设备的健康度的健康度检测模型。
第二方面,本申请实施例提供了一种检测装置,该装置包括:第一 获取单元,被配置成获取油田动设备在生产或自检时的待检测运行数据;检测单元,被配置成将待检测运行数据输入预先构建的健康度检测模型,得到油田动设备的健康度检测结果,其中,健康度检测模型为将油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测油田动设备的健康度的健康度检测模型。
第三方面,本申请实施例提供了一种设备,包括:一个或多个处理器;存储装置,其上存储有一个或多个程序,当一个或多个程序被一个或多个处理器执行,使得一个或多个处理器实现如上述第一方面中所描述的方法。
第四方面,本申请实施例提供了一种计算机可读介质,其上存储有计算机程序,该程序被处理器执行时实现如上述第一方面中所描述的方法。
在本申请实施例中,通过获取油田动设备在生产或自检时的待检测运行数据,而后将待检测运行数据输入预先构建的健康度检测模型,从而得到油田动设备的健康度检测结果。其中,健康度检测模型为将油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测油田动设备的健康度的健康度检测模型。由此,通过该健康度检测模型进行油田动设备的健康度的检测,能够提高对油田动设备健康度检测的准确性,为预测性维护提供强有力的技术支撑。
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述,本申请的其它特征、目的和优点将会变得更明显:
图1是根据本申请的检测方法的一个实施例的流程图;
图2是根据本申请的检测方法的又一个实施例的流程图;
图3是根据本申请的检测装置的一个实施例的结构示意图;
图4是适于用来实现本申请实施例的设备的计算机系统的结构示意图。
下面结合附图和实施例对本申请作进一步的详细说明。可以理解的是,此处所描述的具体实施例仅仅用于解释相关发明,而非对该发明的 限定。另外还需要说明的是,为了便于描述,附图中仅示出了与有关发明相关的部分。
需要说明的是,在不冲突的情况下,本申请中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本申请。
请参考图1,其示出了根据本申请的检测方法的一个实施例的流程100。该检测方法,包括以下步骤:
步骤101,获取油田动设备在生产或自检时的待检测运行数据。
在本实施例中,检测方法的执行主体(例如服务器、用于训练健康度检测模型的智能模型训练平台等)可以获取油田动设备在生产或自检时的待检测运行数据。其中,油田动设备是指适用于油田的、有驱动机带动的转动设备(亦即有能源消耗的设备),如泵、压缩机、风机等。其能源可以是电动力、气动力、蒸汽动力等。
可选的,上述油田动设备安装有多个传感器。上述多个传感器可以包括但不限于:温度传感器、压力传感器、电流表、电压表、声音信号采集器等。执行主体可以通过上述多个传感器,周期性地或者实时地采集油田动设备在生产或自检时的待检测运行数据。
具体地,上述执行主体可以上述油田动设备中的各传感器可以通过RTU(Remote Terminal Unit,远程终端控制系统)、DCS(Distributed Control System,分布式控制系统)、或者PLC(Programmable Logic Controller,可编程逻辑控制器)等与上述执行主体通信连接。此时,上述执行主体可以通过RTU、DCS、或PLC,获取各传感器所采集的待检测运行数据。
可选的,上述待检测运行数据可以包括但不限于以下至少一项:频谱、音频、电流、电压、流量、温度、图片。
可选的,上述执行主体还可以将所获取的待检测运行数据保存至数据库中,以便于为后续的故障分析提供数据支持。
步骤102,将待检测运行数据输入预先构建的健康度检测模型,得到油田动设备的健康度检测结果。
在本实施例中,上述待检测运行数据输入预先构建的健康度检测模型,得到上述油田动设备的健康度检测结果。其中,上述健康度检测模型为将上述油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测上述油田动设备的健康度的健康度检测模型。
此处,历史运行数据可以是上述油田动设备在历史运行或者自检过程中所产生的数据。上述历史运行数据与上述待检测运行数据的具体项目(如频谱、音频、电压、电流、温度等)可以是相同的,此处不再赘述。
此处,历史检测数据可以是在定期对设备巡检时和故障时生成的数据。该数据可以由检测人员人工记录。例如,历史检测数据可以包括用于指示油田动设备是否故障的故障标识、故障类型、报警量、油田动设备的使用年限或使用月数等。实践中,可以每隔固定时间(例如一天、一周等),对油田动设备进行检测,以确定其是否故障,并针对每次检测记录检测数据。
此处,健康度检测结果可以包括健康度的取值。实践中,健康度可以是位于预设数值区间(例如[0,10])的数值。健康度的取值越大,则油田动设备的运行状况越好。
在本实施例中,上述健康度检测模型可以预先训练生成。在训练健康度检测模型前,可以首先利用所获取的历史检测数据,对历史运行数据进行标注。从而,可以将历史运行数据作为输入,将与所输入的历史运行数据作为输出,利用机器学习方法(如有监督学习方法)训练得到健康度检测模型。实践中,健康度检测模型可使用各种常用的处理大数据的模型进行训练。例如卷积神经网络(Convolutional Neural Networks,CNN)等模型。
需要说明的是,可以利用各种预设的标注规则,进行历史运行数据的标注。作为示例,在某个时间段对油田动设备进行检测后,若检测结果指示油田动设备发生了故障,则可以将该时间段所获取的历史运行数据标注为1;若检测结果指示油田动设备未发生故障,则可以将该时间段所获取的历史运行数据标注为0。
需要指出的是,在训练前,可以预先对所采集的运行数据进行数据处理(如数值转换、特征提取等操作),以便于进行模型训练。在训练健康度检测模型的过程中,可以逐一输入各历史运行数据,也可以分批次地每次输入多个历史运行数据。模型训练完成后(例如,满足训练次数达到预设次数、模型预测的准确性达到预设值等预先设定的训练结束条件后),即可将训练完成后的模型确定为健康度检测模型。由此,通过健康度检测模型,即可对油田动设备的健康度进行实时检测。
在本实施例的一些可选的实现方式中,在得到健康度检测结果后,上述执行主体还可以执行如下步骤:
首先,从预设的第一对应关系表中,查找与上述健康度检测结果所指示的当前健康度对应的目标健康等级。其中,上述第一对应关系表用于表征健康等级与健康度范围的对应关系。其中,不同的健康等级(例如A级、B级、C级)可以对应不同的健康度范围。例如,A级对应的健康度范围为(8,10],B级对应的健康度范围为(6,8],C级对应的健康度范围为[0,6]。
而后,从预设的第二对应关系表中,查找上述目标健康等级对应的目标维护方式。其中,上述第二对应关系表用于表征健康等级与维护方式的对应关系。例如,A级对应的维护方式为记录实时的运行数据,B级对应的维护方式为人工检查,C级对应的维护方式为暂停油田动设备的工作并进行人工检查。
最后,利用上述目标维护方式,对上述油田动设备进行维护。由此,在对上述油田动设备的健康度进行实时检测的过程中,上述执行主体可以基于所检测出的健康度对应的健康等级,选取维护方式。进而,利用所选取的维护方式,对上述油田动设备进行维护。由于维护方式根据健康度等级而选取,而非根据维护人员经验所决定,由此,可提高预测性维护的准确性。
本申请的上述实施例提供的方法,通过获取油田动设备在生产或自检时的待检测运行数据,而后将上述待检测运行数据输入预先构建的健康度检测模型,从而得到上述油田动设备的健康度检测结果。其中,健康度检测模型为将油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测上述油田动设备的健康度的健康度检测模型。由此,通过该健康度检测模型进行油田动设备的健康度的检测,能够提高对油田动设备健康度检测的准确性,为预测性维护提供强有力的技术支撑。
进一步参考图2,其示出了检测方法的又一个实施例的流程200。该检测方法的流程200,包括以下步骤:
步骤201,获取油田动设备在生产或自检时的待检测运行数据。
在本实施例中,上述油田动设备可以安装有多个传感器。上述执行 主体可以通过上述多个传感器,周期性地或者实时地采集油田动设备在生产或自检时的待检测运行数据。其中,上述运行数据可以包括但不限于:频谱、音频、电流、电压、流量、温度、图片。
步骤202,将待检测运行数据输入预先构建的健康度检测模型,得到油田动设备的健康度检测结果。
在本实施例中,上述执行主体上述待检测运行数据输入预先构建的健康度检测模型,得到上述油田动设备的健康度检测结果。其中,上述健康度检测模型可以通过如下步骤训练得到:
第一步,获取上述油田动设备的历史运行数据和历史检测数据。
第二步,基于上述历史检测数据,设定上述油田动设备的历史健康度。此处,历史健康度可以是位于预设数值区间(例如[0,10])的数值。健康度的取值越大,则油田动设备的运行状况越好。
此处,可以利用各种预设的规则计算各历史检测数据对应的历史健康度。作为示例,可以结合历史检测数据中的故障标识、故障类型以及报警量确定历史健康度。具体可参见下表:
如上表所示,故障标识为1,则表示油田动设备发生故障;故障标识为0,则表示油田动设备未发生故障。以表中第二列为例,当油田动设备发生故障、故障类型为断裂性故障且报警量处于第三档位时,则历史健康度可以设置为0。
需要说明的是,设定各历史检测数据对应的历史健康度的规则可以根据需要而进行其他设定,此处不作限定。
第三步,建立各历史运行数据与历史健康度的对应关系。作为示例,每一个巡检周期(例如24小时)对设备进行一次巡检,则可以得到每次巡检后所记录的历史检测数据对应的历史健康度。进而,可以对每个巡检周期所采集的历史运行数据与该历史周期对应的历史健康度建立对应关系。
第四步,将各历史运行数据作为预先建立的初始模型的输入,将与所输入的历史运行数据对应的历史健康度作为上述初始模型的输出,利用机器学习方法训练,得到用于检测上述油田动设备的健康度的健康度检测模型。
步骤203,在油田动设备发生故障时,获取油田动设备在发生故障期间的运行数据,并获取油田动设备在发生故障后的检测数据。
在本实施例中,上述执行主体在上述油田动设备发生故障时,可以获取上述油田动设备在发生故障期间的运行数据,并获取上述油田动设备在发生故障后的检测数据。实践中,上述执行主体还可以将上述运行数据保存至数据库中,以便于为后续的故障分析提供数据支持。
步骤204,将运行数据与检测数据作为训练样本,利用机器学习方法对健康度检测模型进行训练,以更新健康度检测模型。
在本实施例中,上述执行主体可以将上述运行数据与上述检测数据作为训练样本,利用机器学习方法对上述健康度检测模型进行训练,以更新健康度检测模型。
在本实施例的一些可选的实现方式中,可以按照如下步骤更新模型:
首先,基于上述检测数据,设定上述油田动设备在上述故障期间内的目标健康度,并建立上述运行数据与上述目标健康度的对应关系。此处,目标健康度的设定方式可以参见步骤202中的历史健康度的设定方式,此处不再赘述。
而后,将上述运行数据作为上述健康度检测模型的输入,将上述目标健康度作为上述健康度检测模型的输出,利用机器学习方法对上述健康度检测模型进行训练,以更新上述健康度检测模型的参数。
从图2中可以看出,与图1对应的实施例相比,本实施例中的检测方法的流程200涉及了利用油田动设备发生故障时的运行数据对模型进行再次训练的步骤。由此,本实施例描述的方案可以对模型进行实时地优化,提高了模型的准确性。
进一步参考图3,作为对上述各图所示方法的实现,本申请提供了一种检测装置的一个实施例,该装置实施例与图1所示的方法实施例相对应,该装置具体可以应用于各种电子设备中。
如图3所示,本实施例所述的检测装置300包括:第一获取单元301,被配置成获取油田动设备在生产或自检时的待检测运行数据;检测单元302,被配置成将上述待检测运行数据输入预先构建的健康度检测模型,得到上述油田动设备的健康度检测结果,其中,上述健康度检测模型为将上述油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测上述油田动设备的健康度的健康度检测模型。
在本实施例的一些可选的实现方式中,上述油田动设备安装有多个传感器;以及上述第一获取单元,进一步被配置成:通过上述多个传感器,周期性地或者实时地采集油田动设备在生产或自检时的待检测运行数据。
在本实施例的一些可选的实现方式中,上述健康度检测模型通过如下步骤训练得到,包括:获取上述油田动设备的历史运行数据和历史检测数据;基于上述历史检测数据,设定上述油田动设备的历史健康度;建立各历史运行数据与历史健康度的对应关系;将各历史运行数据作为预先建立的初始模型的输入,将与所输入的历史运行数据对应的历史健康度作为上述初始模型的输出,利用机器学习方法训练,得到用于检测上述油田动设备的健康度的健康度检测模型。
在本实施例的一些可选的实现方式中,上述装置还包括:第一获取单元,被配置成在上述油田动设备发生故障时,获取上述油田动设备在发生故障期间的运行数据,并获取上述油田动设备在发生故障后的检测数据;更新单元,被配置成将上述运行数据与上述检测数据作为训练样本,利用机器学习方法对上述健康度检测模型进行训练,以更新健康度检测模型。
在本实施例的一些可选的实现方式中,上述更新单元,进一步被配置成:基于上述检测数据,设定上述油田动设备在上述故障期间内的目标健康度,并建立上述运行数据与上述目标健康度的对应关系;将上述运行数据作为上述健康度检测模型的输入,将上述目标健康度作为上述 健康度检测模型的输出,利用机器学习方法对上述健康度检测模型进行训练,以更新上述健康度检测模型的参数。
在本实施例的一些可选的实现方式中,上述装置还包括:第一查找单元,被配置成从预设的第一对应关系表中,查找与上述健康度检测结果所指示的当前健康度对应的目标健康等级,其中,上述第一对应关系表用于表征健康等级与健康度范围的对应关系;第二查找单元,被配置成从预设的第二对应关系表中,查找上述目标健康等级对应的目标维护方式,其中,上述第二对应关系表用于表征健康等级与维护方式的对应关系;维护单元,被配置成利用上述目标维护方式,对上述油田动设备进行维护。
在本实施例的一些可选的实现方式中,上述待检测运行数据包括以下至少一项:频谱、音频、电流、电压、流量、温度、图片。
上述检测装置包括处理器和存储器,上述第一获取单元、检测单元等均作为程序单元存储在存储器中,由处理器执行存储在存储器中的上述程序单元来实现相应的功能。
处理器中包含内核,由内核去存储器中调取相应的程序单元。内核可以设置一个或以上,通过调整内核参数来提高对油田动设备健康度检测的准确性,为预测性维护提供强有力的技术支撑。
本发明实施例提供了一种计算机可读介质,其上存储有程序,该程序被处理器执行时实现所述检测方法。
本发明实施例提供了一种处理器,所述处理器用于运行程序,其中,所述程序运行时执行所述检测方法。
本发明实施例提供了一种设备,设备包括至少一个处理器、以及与处理器连接的至少一个存储器、总线;其中,处理器、存储器通过总线完成相互间的通信;处理器用于调用存储器中的程序指令,以执行上述的检测方法。本文中的设备可以是服务器、PC、PAD、手机等。
本申请还提供了一种计算机程序产品,当在数据处理设备上执行时,适于执行初始化有如下方法步骤的程序:获取油田动设备在生产或自检时的待检测运行数据;将上述待检测运行数据输入预先构建的健康度检测模型,得到上述油田动设备的健康度检测结果,其中,上述健康度检测模型为将上述油田动设备的历史运行数据和历史检测数据作为训练样 本,利用机器学习方法训练得到的用于检测上述油田动设备的健康度的健康度检测模型。
可选的,上述油田动设备安装有多个传感器;以及上述获取油田动设备在生产或自检时的待检测运行数据,包括:通过上述多个传感器,周期性地或者实时地采集油田动设备在生产或自检时的待检测运行数据。
可选的,上述健康度检测模型通过如下步骤训练得到,包括:获取上述油田动设备的历史运行数据和历史检测数据;基于上述历史检测数据,设定上述油田动设备的历史健康度;建立各历史运行数据与历史健康度的对应关系;将各历史运行数据作为预先建立的初始模型的输入,将与所输入的历史运行数据对应的历史健康度作为上述初始模型的输出,利用机器学习方法训练,得到用于检测上述油田动设备的健康度的健康度检测模型。
可选的,当在数据处理设备上执行时,还可适于进一步执行初始化有如下方法步骤的程序:在上述油田动设备发生故障时,获取上述油田动设备在发生故障期间的运行数据,并获取上述油田动设备在发生故障后的检测数据;将上述运行数据与上述检测数据作为训练样本,利用机器学习方法对上述健康度检测模型进行训练,以更新健康度检测模型。
可选的,上述将上述运行数据与上述检测数据作为训练样本,利用机器学习方法对上述健康度检测模型进行训练,以更新健康度检测模型,包括:基于上述检测数据,设定上述油田动设备在上述故障期间内的目标健康度,并建立上述运行数据与上述目标健康度的对应关系;将上述运行数据作为上述健康度检测模型的输入,将上述目标健康度作为上述健康度检测模型的输出,利用机器学习方法对上述健康度检测模型进行训练,以更新上述健康度检测模型的参数。
可选的,在上述得到上述油田动设备的健康度检测结果之后,还可适于进一步执行初始化有如下方法步骤的程序:从预设的第一对应关系表中,查找与上述健康度检测结果所指示的当前健康度对应的目标健康等级,其中,上述第一对应关系表用于表征健康等级与健康度范围的对应关系;从预设的第二对应关系表中,查找上述目标健康等级对应的目标维护方式,其中,上述第二对应关系表用于表征健康等级与维护方式的对应关系;利用上述目标维护方式,对上述油田动设备进行维护。
可选的,上述待检测运行数据包括以下至少一项:频谱、音频、电流、电压、流量、温度、图片。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
在一个典型的配置中,设备包括一个或多个处理器(CPU)、存储器和总线。设备还可以包括输入/输出接口、网络接口等。
存储器可能包括计算机可读介质中的非永久性存储器,随机存取存储器(RAM)和/或非易失性内存等形式,如只读存储器(ROM)或闪存(flash RAM),存储器包括至少一个存储芯片。存储器是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带,磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
还需要说明的是,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、商品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、商品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括要素的过程、方法、商品或者设备中还存在另外的相同要素。
本领域技术人员应明白,本申请的实施例可提供为方法、系统或计算机程序产品。因此,本申请可采用完全硬件实施例、完全软件实施例或结合软件和硬件方面的实施例的形式。而且,本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
以上仅为本申请的实施例而已,并不用于限制本申请。对于本领域技术人员来说,本申请可以有各种更改和变化。凡在本申请的精神和原理之内所作的任何修改、等同替换、改进等,均应包含在本申请的权利要求范围之内。
Claims (10)
- 一种检测方法,其特征在于,所述方法包括:获取油田动设备在生产或自检时的待检测运行数据;将所述待检测运行数据输入预先构建的健康度检测模型,得到所述油田动设备的健康度检测结果,其中,所述健康度检测模型为将所述油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测所述油田动设备的健康度的健康度检测模型。
- 根据权利要求1所述的检测方法,其特征在于,所述油田动设备安装有多个传感器;以及所述获取油田动设备在生产或自检时的待检测运行数据,包括:通过所述多个传感器,周期性地或者实时地采集油田动设备在生产或自检时的待检测运行数据。
- 根据权利要求1所述的检测方法,其特征在于,所述健康度检测模型通过如下步骤训练得到,包括:获取所述油田动设备的历史运行数据和历史检测数据;基于所述历史检测数据,设定所述油田动设备的历史健康度;建立各历史运行数据与历史健康度的对应关系;将各历史运行数据作为预先建立的初始模型的输入,将与所输入的历史运行数据对应的历史健康度作为所述初始模型的输出,利用机器学习方法训练,得到用于检测所述油田动设备的健康度的健康度检测模型。
- 根据权利要求1所述的检测方法,其特征在于,所述方法还包括:在所述油田动设备发生故障时,获取所述油田动设备在发生故障期间的运行数据,并获取所述油田动设备在发生故障后的检测数据;将所述运行数据与所述检测数据作为训练样本,利用机器学习方法对所述健康度检测模型进行训练,以更新健康度检测模型。
- 根据权利要求4所述的检测方法,其特征在于,所述将所述运行数据与所述检测数据作为训练样本,利用机器学习方法对所述健康度检 测模型进行训练,以更新健康度检测模型,包括:基于所述检测数据,设定所述油田动设备在所述故障期间内的目标健康度,并建立所述运行数据与所述目标健康度的对应关系;将所述运行数据作为所述健康度检测模型的输入,将所述目标健康度作为所述健康度检测模型的输出,利用机器学习方法对所述健康度检测模型进行训练,以更新所述健康度检测模型的参数。
- 根据权利要求1所述的检测方法,其特征在于,在所述得到所述油田动设备的健康度检测结果之后,所述方法还包括:从预设的第一对应关系表中,查找与所述健康度检测结果所指示的当前健康度对应的目标健康等级,其中,所述第一对应关系表用于表征健康等级与健康度范围的对应关系;从预设的第二对应关系表中,查找所述目标健康等级对应的目标维护方式,其中,所述第二对应关系表用于表征健康等级与维护方式的对应关系;利用所述目标维护方式,对所述油田动设备进行维护。
- 根据权利要求1所述的检测方法,其特征在于,所述待检测运行数据包括以下至少一项:频谱、音频、电流、电压、流量、温度、图片。
- 一种检测装置,其特征在于,所述装置包括:第一获取单元,被配置成获取油田动设备在生产或自检时的待检测运行数据;检测单元,被配置成将所述待检测运行数据输入预先构建的健康度检测模型,得到所述油田动设备的健康度检测结果,其中,所述健康度检测模型为将所述油田动设备的历史运行数据和历史检测数据作为训练样本,利用机器学习方法训练得到的用于检测所述油田动设备的健康度的健康度检测模型。
- 一种设备,其特征在于,包括:一个或多个处理器;存储装置,其上存储有一个或多个程序,当所述一个或多个程序被所述一个或多个处理器执行,使得所述一个或多个处理器实现如权利要求1-7中任一所述的方法。
- 一种计算机可读介质,其上存储有计算机程序,其特征在于,该程序被处理器执行时实现如权利要求1-7中任一所述的方法。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910746127.0 | 2019-08-13 | ||
CN201910746127.0A CN112393931B (zh) | 2019-08-13 | 2019-08-13 | 检测方法、装置、电子设备和计算机可读介质 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021027213A1 true WO2021027213A1 (zh) | 2021-02-18 |
Family
ID=74569681
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/126217 WO2021027213A1 (zh) | 2019-08-13 | 2019-12-18 | 检测方法、装置、电子设备和计算机可读介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN112393931B (zh) |
WO (1) | WO2021027213A1 (zh) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113404742A (zh) * | 2021-07-09 | 2021-09-17 | 中国人民解放军火箭军工程大学 | 一种基于测试数据的电液伺服机构健康评估方法及系统 |
CN113586961A (zh) * | 2021-07-22 | 2021-11-02 | 北京山太公务机维修技术有限公司 | 一种飞机维修充氧方法、系统、终端及存储介质 |
CN113847305A (zh) * | 2021-09-06 | 2021-12-28 | 盛景智能科技(嘉兴)有限公司 | 作业机械液压系统的预警方法、预警系统及作业机械 |
CN115361254A (zh) * | 2022-08-03 | 2022-11-18 | 杭州大杰智能传动科技有限公司 | 一种用于塔机三大机构的智能主站通讯方法及控制系统 |
WO2023045032A1 (zh) * | 2021-09-23 | 2023-03-30 | 烟台杰瑞石油装备技术有限公司 | 井场设备健康状态的评估方法、评估装置及存储介质 |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113221455A (zh) * | 2021-05-10 | 2021-08-06 | 北京和利时系统工程有限公司 | 一种设备的健康状态检测方法和装置 |
CN114664061A (zh) * | 2022-02-24 | 2022-06-24 | 深圳有象智联科技有限公司 | 安全监控方法、装置、终端及计算机可读存储介质 |
CN115186924A (zh) * | 2022-07-28 | 2022-10-14 | 网思科技股份有限公司 | 基于人工智能的设备健康状态的评估方法和装置 |
CN116699390B (zh) * | 2023-04-20 | 2024-04-26 | 上海宇佑船舶科技有限公司 | 柴油机组故障检测方法和系统 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106022352A (zh) * | 2016-05-05 | 2016-10-12 | 哈尔滨理工大学 | 基于支持向量机的潜油柱塞泵故障诊断方法 |
US20160348479A1 (en) * | 2012-11-16 | 2016-12-01 | Us Well Services Llc | Wireline power supply during electric powered fracturing operations |
CN108460515A (zh) * | 2018-01-16 | 2018-08-28 | 中国海洋石油集团有限公司 | 一种电潜泵井机组故障风险预测方法 |
WO2019023255A1 (en) * | 2017-07-24 | 2019-01-31 | Bj Services, Llc | DEVELOPMENT OF OIL FIELD MODELS USING COGNITIVE CALCULATION |
CN109460561A (zh) * | 2018-07-25 | 2019-03-12 | 北京国双科技有限公司 | 采油设备故障诊断方法及装置 |
CN109854304A (zh) * | 2019-03-11 | 2019-06-07 | 天地(常州)自动化股份有限公司 | 煤矿安全监控系统分级分区报警方法及安全监控系统 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2614653C2 (ru) * | 2011-07-08 | 2017-03-28 | Шлюмбергер Текнолоджи Б.В. | Система и способ определения исправности бурового оборудования |
US9611705B2 (en) * | 2013-10-11 | 2017-04-04 | Antelope Oil Tool & Mfg. Co. | Centralizer preconditioning and testing apparatus and method |
CN109800139A (zh) * | 2018-12-18 | 2019-05-24 | 东软集团股份有限公司 | 服务器健康度分析方法,装置,存储介质及电子设备 |
CN109657982B (zh) * | 2018-12-20 | 2022-02-11 | 三一重能有限公司 | 一种故障预警方法及装置 |
CN109632355B (zh) * | 2018-12-20 | 2021-03-16 | 广州航天海特系统工程有限公司 | 基于机电设备状态数据漂移的故障预测方法及系统 |
-
2019
- 2019-08-13 CN CN201910746127.0A patent/CN112393931B/zh active Active
- 2019-12-18 WO PCT/CN2019/126217 patent/WO2021027213A1/zh active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160348479A1 (en) * | 2012-11-16 | 2016-12-01 | Us Well Services Llc | Wireline power supply during electric powered fracturing operations |
CN106022352A (zh) * | 2016-05-05 | 2016-10-12 | 哈尔滨理工大学 | 基于支持向量机的潜油柱塞泵故障诊断方法 |
WO2019023255A1 (en) * | 2017-07-24 | 2019-01-31 | Bj Services, Llc | DEVELOPMENT OF OIL FIELD MODELS USING COGNITIVE CALCULATION |
CN108460515A (zh) * | 2018-01-16 | 2018-08-28 | 中国海洋石油集团有限公司 | 一种电潜泵井机组故障风险预测方法 |
CN109460561A (zh) * | 2018-07-25 | 2019-03-12 | 北京国双科技有限公司 | 采油设备故障诊断方法及装置 |
CN109854304A (zh) * | 2019-03-11 | 2019-06-07 | 天地(常州)自动化股份有限公司 | 煤矿安全监控系统分级分区报警方法及安全监控系统 |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113404742A (zh) * | 2021-07-09 | 2021-09-17 | 中国人民解放军火箭军工程大学 | 一种基于测试数据的电液伺服机构健康评估方法及系统 |
CN113404742B (zh) * | 2021-07-09 | 2024-01-26 | 中国人民解放军火箭军工程大学 | 一种基于测试数据的电液伺服机构健康评估方法及系统 |
CN113586961A (zh) * | 2021-07-22 | 2021-11-02 | 北京山太公务机维修技术有限公司 | 一种飞机维修充氧方法、系统、终端及存储介质 |
CN113847305A (zh) * | 2021-09-06 | 2021-12-28 | 盛景智能科技(嘉兴)有限公司 | 作业机械液压系统的预警方法、预警系统及作业机械 |
WO2023045032A1 (zh) * | 2021-09-23 | 2023-03-30 | 烟台杰瑞石油装备技术有限公司 | 井场设备健康状态的评估方法、评估装置及存储介质 |
US11867594B2 (en) | 2021-09-23 | 2024-01-09 | Yantai Jereh Petroleum Equipment & Technologies Co., Ltd. | Evaluation method and evaluation device of health state of well site equipment, and storage medium |
CN115361254A (zh) * | 2022-08-03 | 2022-11-18 | 杭州大杰智能传动科技有限公司 | 一种用于塔机三大机构的智能主站通讯方法及控制系统 |
CN115361254B (zh) * | 2022-08-03 | 2024-01-19 | 杭州大杰智能传动科技有限公司 | 一种用于塔机三大机构的智能主站通讯方法及控制系统 |
Also Published As
Publication number | Publication date |
---|---|
CN112393931B (zh) | 2022-04-19 |
CN112393931A (zh) | 2021-02-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021027213A1 (zh) | 检测方法、装置、电子设备和计算机可读介质 | |
US11162837B2 (en) | Detecting faults in rotor driven equipment | |
WO2020019681A1 (zh) | 采油设备故障诊断方法及装置 | |
CN112508105B (zh) | 一种采油机故障检测与检索方法 | |
US20190360942A1 (en) | Information processing method, information processing apparatus, and program | |
US11853047B2 (en) | Sensor-agnostic mechanical machine fault identification | |
US12007745B2 (en) | Apparatus for predicting equipment damage | |
CN110057583A (zh) | 一种轴承故障识别方法、装置及计算机设备 | |
CN111174370A (zh) | 故障检测方法及装置、存储介质、电子装置 | |
CN117270514A (zh) | 基于工业物联网的生产过程全流程故障检测方法 | |
WO2023093919A2 (zh) | 一种洗衣机异音检测方法、装置、电子设备及存储介质 | |
CN113931258A (zh) | 自我诊断方法和无负压叠压供水设备 | |
CN112529036A (zh) | 一种故障预警方法、装置、设备及存储介质 | |
CN115146530A (zh) | 构建焊接质量检测模型的方法、设备、介质及程序产品 | |
CN113762344A (zh) | 机床主轴的故障识别方法、故障识别模型训练方法及装置 | |
CN114492629A (zh) | 异常检测方法、装置、电子设备及存储介质 | |
CN114002332A (zh) | 一种结构损伤监测预警方法及结构完整性数字孪生系统 | |
CN112926636A (zh) | 牵引变流器柜体温度异常检测方法和装置 | |
CN112577664A (zh) | 传感器故障检测方法、装置及相关产品 | |
CN112882898B (zh) | 基于大数据日志分析的异常检测方法、系统、设备及介质 | |
CN110780198B (zh) | 一种伺服焊接系统运动的波形识别方法及设备 | |
CN116482460A (zh) | 电网设备故障诊断方法及相关设备 | |
CN114165430B (zh) | 基于边缘计算的机泵健康监测方法、系统、设备及介质 | |
CN115659271A (zh) | 传感器异常检测方法、模型训练方法、系统、设备及介质 | |
CN112885049B (zh) | 基于运行数据的智能电缆预警系统、方法及装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19941551 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19941551 Country of ref document: EP Kind code of ref document: A1 |