31 results about "Indicator diagram" patented technology

The invention discloses an oil well intelligent monitoring and remote control system, which belongs to the technical field of oil well monitoring and control, and is characterized by comprising the following steps: acquiring oil well production data, and obtaining a real-time indicator diagram and indicator diagram characteristics according to a preset electrical parameter inverse demonstration indicator diagram model; sending a preset security perturbation instruction sequence and obtaining response data corresponding to the security perturbation instruction sequence; obtaining core mechanism parameters of the oil pumping unit system based on the response data and a preset safety perturbation instruction sequence; obtaining a digital shadow model according to the core mechanism parameters and the indicator diagram characteristics; according to the method, core mechanism parameters are identified through anti-demonstration indicator diagram characteristics of an electric parameter anti-demonstration indicator diagram model and a safety perturbation instruction, and closed-loop intelligent control from oil well state monitoring to liquid production flow regulation and control is formed through the digital shadow model. And the accuracy of oil well flow control, the safety of system operation and the energy efficiency management level are improved.

The invention provides an oil well load indicator diagram prediction method and system, and the method comprises the steps: synchronously collecting an initial electrical parameter data set, a load data set, an accelerometer data set and a barometer data set of an oil well, the initial electrical parameter data set comprising an active power data unit, and then obtaining an active power data segment and a load data segment; obtaining a time lag based on the two, and further obtaining a final electrical parameter data segment; acquiring a reference displacement sequence through the accelerometer data set and the barometer data set; obtaining a dynamic stroke length based on the reference displacement sequence and the load data segment, and further obtaining a final displacement sequence; and constructing a prediction neural network, and obtaining an oil well load indicator diagram based on the prediction neural network and the real-time electrical parameter data segment. According to the technical scheme, the indicator diagram is obtained through continuous and high-frequency real-time electric parameter data segments, motion period information loss caused by low-frequency data collection is avoided, accumulative errors are avoided, and the accuracy of the indicator diagram is ensured.

The present application relates to a frequency and voltage coordination optimization method for beam pumping system based on dynamometer card, which converts the traditional dynamometer card into a P-T dynamometer card of load and time, and obtains the mapping relationship between the P-T dynamometer card and the electric diagram by comparing the P-T dynamometer card and the electric diagram in the same time domain and the same period, relying on the mathematical relationship between the equivalent load torque of the polished rod and the equivalent driving torque of the motor. The method converts the electric diagram features into P-T dynamometer card features, takes the mapping feature points of the P-T dynamometer card as input, relies on deep learning to output the optimal P-T dynamometer card load, and finally obtains the corresponding optimal frequency and optimal voltage according to the mapping model processing data, to establish a flexible intelligent coordination optimization model for the frequency and voltage of the beam pumping system based on the dynamometer card. The present application avoids the application of electric diagram acquisition equipment, improves the energy saving effect and the anti-theft performance, and can ensure the real-time flexible intelligent coordination optimization of the frequency and voltage of the beam pumping system.

This invention relates to the field of computer data processing and data visualization technology, specifically a personalized report configuration platform system, comprising: a data virtualization module: establishing physical data source mapping and constructing atomic data pools; an indicator graph construction module: generating atomic indicator dependency graphs, storing logical operation relationships rather than calculation results; a context-aware module: capturing environmental parameters such as tenant roles and time windows; a dynamic compilation engine: injecting environmental parameters into the graph and instantiating and parsing logical nodes to generate structured query instructions; and a rendering and compositing module: executing instructions to obtain result data and mapping it to view components. This invention achieves dynamic instantiation of logic defined once in multiple places, avoiding hard-coding development and improving the maintainability and response speed of the system.

The invention relates to a structure-semantic double-graph collaborative SAR image quality evaluation method, which comprises the following steps of separating an SAR image, and constructing a double-channel quality measurement set for independent modeling; indexes are obtained, correlation calculation is carried out, an index map is constructed, and each node corresponds to one index; on the basis of the index map, the dependency relationship between indexes is calculated from the structure dimension and the semantic dimension through self-supervised learning, the dependency relationship of the two dimensions is integrated, and an image level score based on the integration result is output; and constructing a measurement index shared with system analysis and a measurement index special for information retrieval, and carrying out quality measurement evaluation: after extracting a structure sub-graph, integrating an IR special index which is not contained in SAR prior into a global graph. The method is superior to a reference model in the aspects of recognition accuracy, recall rate, precision and efficiency, and a solution with high universality and good interpretability is provided for cross-modal image quality evaluation.

The invention provides a real-time classification method and system for working condition data, a storage medium and a computer. The method comprises the following steps: preprocessing sampling data obtained by continuously sampling an oil well to obtain an operation cycle of the oil well; the sampling frequency is adjusted according to the operation cycle of the oil well, secondary sampling is conducted on the oil well through the adjusted sampling frequency, and obtained secondary sampling data is compressed; performing indicator diagram classification coding based on the past indicator diagram classification data to obtain a coding table, and training and simulating a neural network by using the past indicator diagram classification data to obtain a neural network model; and carrying out data processing on the secondary sampling data by utilizing a neural network model so as to realize real-time classification of the working condition data of the oil well. The machine learning based on the neural network algorithm can be used for analyzing and identifying the wellhead working condition, data acquisition is quantized, a simplified neural network is constructed, and a foundation is laid for improvement of an oil extraction working condition identification technology.

The invention relates to the technical field of oil and gas exploitation, in particular to a rod-pumped well working condition diagnosis and identification method and device based on edge intelligence. At the edge end, a deep neural network inference engine MNN is constructed to train a rod-pumped well model, rod-pumped well production abnormal working condition data is collected, and an abnormal working condition indicator diagram data set is obtained after cleaning, noise processing and normalization; then, combining with a rod-pumped well production monitoring video stream feature sequence to form a rod-pumped well abnormal working condition and production monitoring video image combined data set, fusing with an abnormal working condition alarm information sequence obtained through language model preprocessing, and jointly reasoning time sequence data images, video stream data and texts; and an iterative multi-modal modulation neural network model is used as an optimal generalization model, and a prediction result is output. According to the method, the diagnosis real-time performance is improved, the diagnosis accuracy of the abnormal working condition of the rod-pumped well is improved, and meanwhile the production stability of the rod-pumped well is improved.

This invention relates to the field of intelligent fault diagnosis technology, and provides an intelligent fault diagnosis method for reciprocating compressors based on multi-feature fusion, comprising: acquiring dynamometer card data during the operation of the reciprocating compressor and preprocessing it; extracting multi-dimensional features with clear physical meaning from the dynamometer card, including energy, dynamic, geometric, loss, and proportional features; fusing the multi-dimensional features based on an adaptive weight allocation strategy to calculate a comprehensive fault score; combining feature combinations and weight configurations to match a fault mode library and identify the fault type; and determining the fault level based on the comprehensive fault score using fuzzy boundary processing and a dynamic threshold adaptive update mechanism. This method, by constructing a multi-dimensional feature space and integrating physical mechanisms and intelligent algorithms, achieves high-precision, multi-type, and refined fault diagnosis for reciprocating compressors, significantly improving diagnostic accuracy and reducing the risk of false alarms and missed alarms.

The application discloses a fracturing equipment state monitoring and fault diagnosis system and method, and belongs to the intelligent technology field of oil and gas field fracturing equipment. The application aims to solve the problems of single signal monitoring, lagging fault early warning and high misjudgment rate in the prior art. The method comprises the following steps: collecting multi-source operation data of a fracturing pump in real time; establishing a theoretical pressure indicator diagram, and comparing the theoretical pressure indicator diagram with an actual indicator diagram generated by real-time data to realize first-stage fault judgment; processing signals by using an improved wavelet threshold denoising method, and extracting time domain and frequency domain features; and inputting the processed feature data into a joint diagnosis model to realize second-stage fault judgment. The joint diagnosis model combines a principal component analysis (PCA) model for uncalibrated data anomaly detection and a BP neural network model for calibrated data fault classification. Through deep fusion of a mechanism model and a data-driven model, the application effectively resists environmental interference, and significantly improves diagnosis accuracy and operation and maintenance efficiency.

This application relates to an operator performance optimization method, apparatus, computer device, computer-readable storage medium, and computer program product. The method includes: acquiring the original operator code; executing the original operator code and acquiring performance indicator text data output during the execution of the original operator code; generating a graphical performance analysis view corresponding to the original operator code based on the performance indicator text data; acquiring images of the graphical performance analysis view based on performance bottleneck indicators; fusing and analyzing the performance indicator text data and performance indicator image data using a pre-set visual language model to obtain performance bottleneck diagnostic data for the original operator code; analyzing the performance bottleneck diagnostic data using multiple pre-set expert agents; and optimizing the original operator code based on code optimization suggestions. This method can improve the efficiency and effectiveness of code optimization.

The application discloses a gas well plunger gas lift drainage gas indicator diagram drawing method and application thereof, wherein the method comprises the following steps: drawing a first straight line segment according to the relationship between pressure and displacement when the plunger moves downward in the gas column; drawing a second straight line segment according to the relationship between pressure and displacement when the plunger falls in the liquid column; drawing a third straight line segment according to the jump drop value of the pressure on the plunger when the well is opened; drawing a fourth straight line segment according to the relationship between pressure and displacement when the plunger discharges the gas column in the upward process; drawing a fifth straight line segment according to the relationship between pressure and displacement when the plunger discharges the liquid column in the upward process; and drawing a sixth straight line segment according to the pressure drop value of the plunger after reaching the wellhead. The indicator diagram drawn by the application effectively guides real-time analysis of the plunger gas lift process, and plays an important role in plunger operation condition diagnosis, prediction, drainage capacity measurement and system optimization.

The present application belongs to the technical field of fault detection of beam pumping unit, and particularly relates to a beam pumping unit crank pin bearing fault diagnosis method. The method comprises: obtaining the horse head displacement curve and the indicator diagram of the beam pumping unit, and determining the crank pin bearing fault of the pumping unit when the horse head displacement curve and the indicator diagram in a certain stroke satisfy any of the following conditions: burrs appear in the corresponding points of the first half of the horse head displacement curve in the horse head upward process, and the loading line in the indicator diagram is twisted; burrs appear in the corresponding points of the first half of the horse head displacement curve in the downward process of the horse head, and the unloading line in the indicator diagram is twisted; burrs appear in the corresponding points of the first half in the upward process of the horse head and in the corresponding points of the first half in the downward process of the horse head, and the loading line and the unloading line in the indicator diagram are twisted. The method does not need to be inspected by a person on the site, can simply, quickly and timely find the crank pin bearing fault of the pumping unit, and ensures the normal operation of the pumping unit.

The present application relates to the technical field of oil well fault monitoring, and discloses an oil well indicator diagram real-time fault prediction method and system. The method comprises: obtaining an oil well sensor data stream, verifying data integrity through a sliding window buffer, applying a dynamic time warping algorithm to align multi-channel sensor data to generate a standardized data stream; extracting time domain features based on the data stream, comparing the features with a historical feature library after dimensionality reduction by principal component analysis to generate a feature difference index; triggering a multi-level threshold strategy according to the difference index, collecting an incremental training data set, fine-tuning the model using an elastic weight maintenance algorithm and generating a hot switching ready model; loading the model and generating a fault probability value through GPU accelerated inference, generating a timestamped early warning event; finally, analyzing the early warning protocol message edge transmission, and dynamically optimizing system resources based on logs. The present application effectively overcomes the delay problem under high-frequency data flow, significantly improving the accuracy of fault prediction and the system response speed.

The application provides a cloud edge joint control system for intelligent variable-speed driving of a pumping unit and a control method thereof, and belongs to the technical field of pumping unit monitoring. According to the indicator diagram obtained through electrical parameter inversion, the application performs working condition diagnosis, automatically judges the continuous pumping mode and the intermittent pumping mode according to the working condition diagnosis result and the environmental temperature, automatically adjusts the pumping frequency in the continuous pumping mode, automatically starts and stops in the intermittent pumping mode, realizes the balance between oil well supply and discharge with low energy consumption, and the whole working mode is completely self-adaptive to the change of the oil well working condition, thereby improving the oil well management efficiency and reducing the production energy consumption.

The sucker-rod pump underground health diagnosis disc comprises a main disc and an auxiliary disc which are coaxially and rotationally connected, the first main disc face of the main disc is provided with an indicator diagram area, an indicator diagram classification area and a first analysis numbering area, and the second main disc face of the main disc is provided with a first indicator diagram analysis area, a second analysis numbering area and a first measure numbering area. The first indicator diagram analysis area, the second analysis numbering area and the first measure numbering area are all annular areas and are sequentially and correspondingly arranged from outside to inside, and a second indicator diagram analysis area, a third analysis numbering area, a second measure numbering area, a measure area and a first transparent area are arranged on the first auxiliary disc face of the auxiliary disc. The sum of the number type of the third analysis number area and the number type of the second analysis number area is equal to the number type of the first analysis number area. According to the method and the system, technicians can quickly diagnose the rod-pumped well, accurately identify potential faults and take correct and effective measures in time, so that continuous development of the faults is avoided, and the possibility of accidents is reduced.

The application discloses a comprehensive working condition diagnosis method based on electric parameter and dynamometer card information fusion, and based on dictionary coding of current state of an oil pumping well motor as "0: stable, 1: rising, 2: falling", the working condition is classified as follows: A class which can use electric parameter data to quickly diagnose specific working condition; B class and C class which can use electric parameter data to quickly diagnose specific working condition category but cannot determine specific working condition, B class such as down pump (B1) and insufficient liquid supply (B2) working condition with current coding all as "0-02"; C class such as fixed valve leakage (C1) and pipe leakage (C2) working condition with current coding all as "2-0"; and D class which cannot use current change to represent working condition. The algorithm fully gives play to the advantages of electric parameter identification and dynamometer card identification method, has higher accuracy than traditional electric parameter identification method, and has better identification accuracy and speed than traditional dynamometer card identification method.

The invention discloses a mechanical production well working fluid level prediction method. The method comprises the following steps that oil well static data, working fluid level measurement data and oil well real-time collection data are obtained; the oil well static data, the working fluid level measurement data and the oil well real-time collection data are preprocessed; electric parameter characteristic parameters and indicator diagram characteristic parameters are obtained through calculation based on the preprocessed oil well static data and the preprocessed oil well real-time collected data; performing dimensionless processing on the electrical parameter characteristic parameters and the indicator diagram characteristic parameters to obtain dimensionless characteristic parameters; calculating a Pearson's correlation coefficient and a Spearman correlation coefficient of each dimensionless characteristic parameter, and screening through the Pearson's correlation coefficient and the Spearman correlation coefficient to obtain a working fluid level correlation characteristic parameter; and processing the working fluid level related characteristic parameters through the working fluid level online prediction model to obtain a working fluid level prediction value of the mechanical production well. According to the method for predicting the working fluid level of the mechanical production well, the working fluid level prediction precision is improved, and subtle changes of the production process can be sensed in an all-weather, all-parameter and whole-process mode.

The invention relates to the technical field of intelligent fault diagnosis, and provides a reciprocating compressor indicator diagram intelligent fault diagnosis method based on multi-feature fusion, and the method comprises the steps: obtaining indicator diagram data in the operation process of a reciprocating compressor, and carrying out the preprocessing; extracting multi-dimensional features with clear physical significance from the indicator diagram, wherein the multi-dimensional features comprise energy, dynamic, geometric, loss and proportion features; fusing the multi-dimensional features based on an adaptive weight distribution strategy, and calculating a fault comprehensive score; combining feature combination and weight configuration, matching a fault mode library, and identifying a fault type; and based on the fault comprehensive score, a fuzzy boundary processing and dynamic threshold adaptive updating mechanism is adopted to determine the fault level. According to the method, by constructing the multi-dimensional feature space and fusing a physical mechanism and an intelligent algorithm, high-precision, multi-type and refined diagnosis of faults of the reciprocating compressor is achieved, the diagnosis accuracy is remarkably improved, and the misinformation and missing report risks are reduced.

The invention provides a small sample class increment non-analytic indicator diagram fault diagnosis method based on serialization and comparative learning. The method comprises the following steps of non-analytic indicator diagram serialization; comparative learning pre-training; carrying out incremental learning on small sample classes; performing fault diagnosis; the invention further provides a small sample class increment non-analytic indicator diagram fault diagnosis system based on serialization and comparative learning, wherein the method is applied to the small sample class increment non-analytic indicator diagram fault diagnosis system. According to the small sample class increment non-analytic indicator diagram fault diagnosis system based on serialization and comparative learning and the diagnosis method thereof, the information density of the non-analytic indicator diagram is improved through a serialization method, and efficient and stable dynamic fault diagnosis is achieved in combination with comparative learning pre-training, a lightweight adapter, sample replay and an NME classifier.

The invention provides a small sample class increment non-analytic indicator diagram fault diagnosis method based on serialization and comparative learning. The method comprises the following steps of non-analytic indicator diagram serialization; comparative learning pre-training; carrying out incremental learning on small sample classes; performing fault diagnosis; the invention further provides a small sample class increment non-analytic indicator diagram fault diagnosis system based on serialization and comparative learning, wherein the method is applied to the small sample class increment non-analytic indicator diagram fault diagnosis system. According to the small sample class increment non-analytic indicator diagram fault diagnosis system based on serialization and comparative learning and the diagnosis method thereof, the information density of the non-analytic indicator diagram is improved through a serialization method, and efficient and stable dynamic fault diagnosis is achieved in combination with comparative learning pre-training, a lightweight adapter, sample replay and an NME classifier.

The invention discloses an indicator diagram and friction characteristic measuring method, device and equipment of a low-temperature and high-pressure reciprocating pump, and relates to the technical field of new energy, the method comprises the steps that a convenient rotation angle measuring device is provided, and in-situ online measurement of the rotation angle of a crankshaft of the low-temperature and high-pressure reciprocating pump and the working volume of a hydraulic cylinder is achieved; a stress balance model of piston rod load, piston force, reciprocating inertia force, friction force between a piston ring and the inner surface of a cylinder sleeve and friction force between the piston rod and a piston rod dynamic seal is constructed, and a calculation method of the dynamic friction coefficient between the piston ring and the inner surface of the cylinder sleeve, the friction force between the piston rod and the piston rod dynamic seal and cylinder pressure is provided. Real-time acquisition of cylinder pressure and fluid end friction characteristics of the low-temperature and high-pressure reciprocating pump is realized. On the premise that a pressure guiding device is not needed to be additionally arranged, in-situ online measurement of the indicator diagram and the fluid end friction characteristic of the low-temperature and high-pressure reciprocating pump can be economically and accurately achieved, and a foundation is laid for online monitoring of the safety and service performance of the low-temperature and high-pressure reciprocating pump.

The invention belongs to the technical field of beam-pumping unit fault diagnosis, and particularly relates to a beam-pumping unit tail bearing fault diagnosis method. Comprising the following steps that a horse head displacement curve and an indicator diagram of the beam-pumping unit are obtained, and when the horse head displacement curve and the indicator diagram in a certain stroke meet the following conditions at the same time, it is judged that a tail bearing of the beam-pumping unit breaks down: burrs appear at corresponding points of the second half stroke in the horse head ascending process in the horse head displacement curve; the rear half section of the load stability line in the ascending process of the oil pumping unit indicator diagram has typical characteristics; wherein when one displacement value corresponding to at least two load values appears in the rear half section of the load stability line in the uplink process of the indicator diagram, it is judged that typical features appear in the rear half section of the load stability line in the uplink process of the indicator diagram. The fault of the tail bearing of the oil pumping unit can be found simply, quickly and timely without the need for inspection personnel to inspect on site, so that the normal operation of the oil pumping unit is ensured.

The application discloses a metal quantitative detection method based on pressure feedback, comprising the following steps: obtaining the pressure curve of the metal to be detected in unit time, extracting the upper envelope and the lower envelope of the pressure curve through Hilbert-Huang transform, determining the indicator diagram of the pressure curve at the current moment, and performing normalization processing on the indicator diagram; extracting the geometric features of the indicator diagram, extracting the moment features of the indicator diagram, extracting the statistical features of the indicator diagram, performing set processing on the statistical features, the geometric features and the moment features as a feature set, performing polynomial feature analysis on the feature set, inputting the result after the polynomial feature analysis into a regression prediction model, determining the soup output of the metal to be detected at the current moment in unit time, and comparing the accumulated soup output with a set value to quantitatively detect the metal. The application realizes accurate and rapid prediction of the soup output of the metal, simplifies the control process, and reduces the control cost.

The present application belongs to the technical field of oil extraction equipment information processing method, and discloses an oil pumping unit operation data processing system based on edge cloud cooperation, comprising a plurality of edge end nodes deployed at different oil pumping well sites, each edge end node being in communication connection with a cloud server, the edge end node and the cloud server being in communication connection with a visualization platform, the visualization platform receiving and displaying working condition diagnosis results; each edge end node is used for collecting operation data of a local oil pumping unit, constructing a local training data set, training a local working condition diagnosis model, obtaining a local model weight and uploading it to the cloud server; the cloud server aggregates each local model weight received, generates a global model weight, and issues it to each edge end node to update the local working condition diagnosis model of each edge end node, and the diagnosis model output obtains a visual working condition diagnosis result. The system solves the problem of scattered pumping unit indicator diagram data, which is difficult to model and diagnose.

The embodiment of the invention discloses an indicator diagram determination method and device based on a load sensor, the load sensor, a storage medium and a program product. The method is used in cooperation with a target oil pumping unit, the target oil pumping unit comprises a roller, a head sheave and a steel wire rope connecting the roller and the head sheave, a load sensor is installed on a base of the head sheave, and the method comprises the steps that acting force collected by the load sensor and the gravity of the head sheave are obtained; according to the included angle between the steel wire rope and the vertical direction, the trigonometric function value of the included angle is determined, and the vertical direction is the vertical direction of the suspension point of the target oil pumping unit; according to the acting force, the gravity and the trigonometric function value, the suspension center load of the suspension center is determined, and an indicator diagram of the target pumping unit is determined according to the suspension center load. According to the technical scheme of the embodiment of the invention, high-precision determination of the indicator diagram can be realized.

1. Name of this design product: Graphical user interface for anomaly analysis and monitoring of operation and maintenance log fields of electronic equipment. 2. Purpose of this design: An electronic device. 3. The key design feature of this product is its graphical user interface. 4. The image or photograph that best illustrates the design's key features: the front view. 5. Purpose of the graphical user interface: Used in operation and maintenance monitoring products, it can monitor for anomalies in link log fields. 6. Human-computer interaction method of graphical user interface: The main view is the initial interface; in the main view, clicking the "11.2.5.6" icon in the lower left corner list will enter interface state diagram 1; in interface state diagram 1, clicking "View Entity Details" in the floating window icon will enter interface state diagram 2; in interface state diagram 2, clicking the "X" in the upper right corner will enter interface state diagram 3; in interface state diagram 3, clicking the "Configuration" icon on the right side of the middle of the page will enter interface state diagram 4; in interface state diagram 4, clicking "Show" after "Total", "Error Count", and "Root Error Count" in the display indicator list will enter interface state diagram 5; in interface state diagram 5, clicking... After clicking the "Configure Metrics" icon, you will enter interface state diagram 6; in interface state diagram 6, enter the metric name, field variable, and SPL, then click "OK" to enter interface state diagram 7; in interface state diagram 7, click "Hide" next to "test1" to enter interface state diagram 8; in interface state diagram 8, click "Abnormal Score Weight" to enter interface state diagram 9; in interface state diagram 9, change the difference value to 0.1 to enter interface state diagram 10; in interface state diagram 10, click the "Confirm" icon to enter interface state diagram 11; in interface state diagram 11, click the last item in the list, "Abnormal Score," to automatically sort the data, then enter interface state diagram 12. 7. Other situations requiring explanation: This display panel is used in electronic devices.

The application discloses a kind of based on unsupervised learning's indicator diagram sample set preparation method and device, comprising: obtaining oilfield data, and constructing fault diagnosis big data system according to oilfield data;According to the fault diagnosis big data system, obtain detection sample vector and carry out pre-processing operation;The sample vector after pre-processing is clustered using clustering algorithm to find optimal classification number, and the marking of fault diagnosis category is realized according to optimal classification result.The indicator diagram sample set preparation method provided by the application, when there are fewer marked examples, the learning performance is improved by a large number of unmarked examples, the optimal classification number is determined using the profile coefficient in clustering algorithm, the same vector is plotted into indicator diagram according to classification result, and judgment is carried out in combination with expert experience, the problem that learning performance is poor in sample marking and artificial marking is prone to error is avoided, the human and time cost is reduced, and the accuracy of fault diagnosis classification is improved.

The invention relates to an oil field indicator diagram working condition identification method, system and device based on lightweight double-path attention and a medium, and the method comprises the steps: collecting indicator diagram images generated in the operation process of a pumping unit from a plurality of main oil field sites, carrying out the manual marking of the working condition type of each indicator diagram image, constructing an indicator diagram data set, and carrying out the manual marking of the working condition type of each indicator diagram image; dividing into an indicator diagram training set, a verification set and a test set; a MobileNetDual oil field indicator diagram recognition model is constructed; inputting the indicator diagram training set into a MobileNetDual oilfield indicator diagram recognition model, performing training by adopting a gradient descent method, and monitoring loss and accuracy on the indicator diagram verification set to obtain a trained MobileNetDual oilfield indicator diagram recognition model; converting the data into a format adaptive to the edge computing equipment, and carrying out quantitative compression; inputting the indicator diagram test set into the quantized MobileNetDual oilfield indicator diagram recognition model for testing to obtain a recognition result; according to the method, efficient and stable operation of the model on the resource-limited edge equipment is ensured, and a closed loop from theory to industry is really realized.

The invention discloses a pumping unit edge fault diagnosis method, and relates to the field of crude oil exploitation, and the method comprises the steps: respectively building a pumping unit indicator diagram fault diagnosis model and a pumping unit electric indicator diagram fault diagnosis model; collecting suspension center displacement data, suspension center load data and electric power data of the pumping unit to be diagnosed; and judging whether an indicator diagram and an electric indicator diagram of the to-be-diagnosed oil pumping unit under the current working condition can be obtained, and outputting whether a fault occurs and a fault type. The invention provides an edge fault diagnosis method for an oil pumping unit, and aims to solve the problems that an oil pumping unit fault diagnosis method in the prior art is insufficient in accuracy, real-time performance, complex fault processing capability, field adaptability and the like. The purposes of improving the effectiveness and accuracy of complex fault diagnosis of the oil pumping unit and promoting the intelligent development of an oil field are achieved.

The invention provides an intelligent fault diagnosis method for an air valve of a reciprocating machine, which comprises the following steps of: on the basis of a logarithm p-V diagram, performing equal parameter normalization processing on an indicator diagram of a reciprocating compressor; constructing a convolutional neural network of four layers of convolution to identify and classify p-V maps; a self-learning sample database is constructed, original p-V graph data in the training sample database is subjected to three kinds of common normalization and equal parameter normalization processing, a convolutional neural network structure is used for training, and test set accuracy results are compared; based on a fault experiment, dynamic pressure data and valve cover vibration frequency domain data which are subjected to equal parameter element normalization processing in the same work period are simultaneously input into the constructed convolutional neural network for learning and recognition, and the specific fault state of the air valve is further classified. According to the method, intelligent diagnosis is carried out based on the convolutional neural network, automatic induction and classification of fault features are realized through a self-learning function, and a dynamic updating model is supported to adapt to a new fault type.