Oil well equipment state evaluation method based on machine vision

Abstract

The invention discloses an oil well equipment state evaluation method based on machine vision, and relates to the technical field of machine learning, and the method comprises the steps: collecting the image data of an associated assembly of each oil well equipment, and analyzing the real-time state index of the associated assembly of the oil well equipment; acquiring historical operation and maintenance data of the oil well equipment, performing time sequence analysis on the historical operation and maintenance data, and evaluating a historical operation loss coefficient of an associated component of the oil well equipment; correcting the real-time state index of the associated component of the oil well equipment based on the historical operation loss coefficient of the associated component of the oil well equipment to obtain a corrected state index of the associated component of the oil well equipment; obtaining a to-be-mined task of the oil well equipment and the associated component correction state index of the oil well equipment for fitting to evaluate the associated component abnormal risk probability of the oil well equipment; judging whether the abnormal risk probability of the associated component of the oil well equipment is a dangerous critical value; the method has the advantages that the abnormal risk of the associated assembly of the oil field equipment is accurately predicted, and the maintenance cost of the oil field equipment is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of machine learning, and specifically relates to a method for evaluating the state of oil well equipment based on machine vision. Background Art

[0002] Oil well equipment refers to various mechanical equipment required for a series of operations such as oil exploration, extraction, production, and transportation. Together, they form the basis of the oil production system and ensure the extraction of oil from underground.

[0003] At present, the method for monitoring the state of oil well equipment mainly relies on real-time data of various monitoring sensors to judge operation deviations in order to determine whether the equipment needs maintenance. However, the various monitoring sensors and the oil field equipment itself are in the aging process, resulting in inaccurate state monitoring of the equipment after a long time and increased maintenance costs. Summary of the Invention

[0004] To solve the above technical problems, a method for evaluating the state of oil well equipment based on machine vision is provided. This technical solution solves the problem that the current method for monitoring the state of oil well equipment mainly relies on real-time data of various monitoring sensors to judge operation deviations in order to determine whether the equipment needs maintenance. However, the various monitoring sensors and the oil field equipment itself are in the aging process, resulting in inaccurate state monitoring of the equipment after a long time and increased maintenance costs.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for evaluating the state of oil well equipment based on machine vision, comprising: Based on an image sensor, collecting image data of associated components of each oil well equipment and analyzing real-time state indicators of the associated components of the oil well equipment; Obtaining historical operation and maintenance data of the oil well equipment, performing time series analysis on the historical operation and maintenance data, and evaluating the historical operation loss coefficient of the associated components of the oil well equipment; Loss caused by human factors Correcting the real-time state indicators of the associated components of the oil well equipment based on the historical operation loss coefficient of the associated components of the oil well equipment to obtain corrected state indicators of the associated components of the oil well equipment; Obtaining the to-be-exploited tasks of the oil well equipment and fitting and evaluating the abnormal risk probability of the associated components of the oil well equipment with the corrected state indicators of the associated components of the oil well equipment; Judging whether the abnormal risk probability of the associated components of the oil well equipment is at the dangerous critical value. If not, it is determined to execute normally. If so, it is determined that the equipment is abnormal; Wherein, the corrected state indicator of the associated component of the oil well equipment is specifically: G‘ ij =G ij ×H ij ; Wherein, G' ij is the corrected status index of the j-th associated component of the i-th oil well equipment, and G ij is the real-time status index of the j-th associated component of the i-th oil well equipment, and H ij is the historical operation loss coefficient of the j-th associated component of the i-th oil well equipment.

[0006] Preferably, based on an image sensor, the image data of the associated components of each oil well equipment is collected, and the analysis of the real-time status index of the associated components of the oil well equipment specifically includes: Based on the structural parameters of the associated components of the oil well equipment, the initial image data of the associated components of the oil well equipment is determined; Based on the image data of the associated components of each oil well equipment, Gaussian filtering is used to eliminate the noise in the associated component images, and the preprocessed image data of the associated components of the oil well equipment is generated; For the preprocessed image data of the associated components of the oil well equipment, the edge information and shape texture information of the associated component images are extracted to obtain the image feature data of the associated components of the oil well equipment; Based on the image feature data of the associated components of the oil well equipment and the initial image data of the associated components of the oil well equipment, they are packaged into a loss positive and negative sample dataset of the associated components of the oil well equipment; Based on the SVM support vector, a status analysis model of the associated components of the oil well equipment is constructed; The loss positive and negative sample dataset of the associated components of the oil well equipment is substituted into the status analysis model of the associated components of the oil well equipment. The image feature data of the associated components of the oil well equipment is used as the positive sample, and the initial image data of the associated components of the oil well equipment is used as the negative sample (using the image feature data of the associated components as the negative sample) to evaluate the real-time status index of the associated components of the oil well equipment; Among them, the status analysis model of the associated components of the oil well equipment is specifically: ; Wherein, G ij is the real-time status index of the j-th associated component of the i-th oil well equipment, w is the weight vector, and F ij is the feature vector of the j-th associated component of the i-th oil well equipment, b is the bias term, C is the regularization parameter, and ξ ij is the slack variable of the j-th associated component of the i-th oil well equipment, and y ij is the true label of the j-th associated component of the i-th oil well equipment, ξ ij ≥0, which is a non-negative value constraint condition, is the constraint condition, N + is the total number of positive samples of the loss of the associated components of the oil well equipment, and N - is the total number of negative samples of the loss of the associated components of the oil well equipment.

[0007] Preferably, obtaining the historical operation and maintenance data of the oil well equipment, performing time series analysis on the historical operation and maintenance data, and evaluating the historical operation loss coefficient of the associated components of the oil well equipment specifically includes: Obtaining the standardized operation parameters of the associated components of the oil well equipment, and forming an array of the standardized operation parameters of the associated components of the oil well equipment; Based on the historical operation and maintenance data of the oil well equipment, taking the time interval of the historical production tasks as the observation window and the historical operation and maintenance data of the oil well equipment as the observation variables, an observation time series window is established; Based on the observation time series window, obtaining the execution process time series parameters of the associated components of the oil well equipment in the historical production tasks; According to the array of the standardized operation parameters of the associated components of the oil well equipment, screening out the deviation parameters of the operation time series parameters of the associated components of the oil well equipment in the historical production tasks, denoted as the execution process time series parameters of the associated components of the oil well equipment in the deviation tasks; According to the execution process time series parameters of the associated components of the oil well equipment in the historical production tasks and the execution process time series parameters of the associated components of the oil well equipment in the deviation tasks, forming a characteristic data set of the performance loss factors of the associated components of the oil well equipment Based on the characteristic data set of the performance loss factors of the associated components of the oil well equipment, estimating the basic performance loss of the execution process time series parameters of the associated components of the oil well equipment in the historical production tasks as the performance loss constant, and estimating the deviation performance loss of the execution process parameters of the associated components of the oil well equipment in the deviation tasks as the performance loss variable, and evaluating the historical operation loss coefficient of the associated components of the oil well equipment; Among them, the performance decline regression model of the associated components of the oil well equipment is specifically: ; In the formula, H ij is the historical operation loss coefficient of the jth associated component of the ith oil well equipment, α0 is the intercept term, V ij is the basic performance loss constant of the jth associated component of the ith oil well equipment, V' ij is the deviation performance loss variable of the jth associated component of the ith oil well equipment, α1 is the performance loss constant coefficient, α2 is the performance loss variable coefficient, and ∈ is the error term.

[0008] Preferably, obtaining the to-be-exploited tasks of the oil well equipment and the corrected status indicators of the associated components of the oil well equipment for fitting and evaluating the abnormal risk probability of the associated components of the oil well equipment specifically includes: Based on logistic regression, constructing a risk assessment model for the associated components of the oil well equipment; Obtain the to-be-exploited tasks of the oil well equipment, determine the task execution process of the associated components of the oil well equipment for the to-be-exploited tasks, and obtain the execution parameters of the associated components of the oil well equipment for the to-be-exploited tasks; Based on the corrected status indicators of the associated components of the oil well equipment, linearly map each corrected status indicator of the associated components into a corrected status vector indicator of the associated components of the oil well equipment, and substitute it into the risk assessment model of the associated components of the oil well equipment, using the execution parameters of the associated components of the oil well equipment for the to-be-exploited tasks as the input and the abnormal risk probability of the associated components of the oil well equipment as the output; Among them, the specific implementation of the risk assessment model of the associated components of the oil well equipment is as follows: ; In the formula, P ij is the abnormal risk probability of the j-th associated component of the i-th oil well equipment, β0 is the intercept term, S ij is the execution parameter of the j-th associated component of the i-th oil well equipment for the to-be-exploited tasks, β1 is the correlation coefficient vector of the execution parameters, β2 is the correlation coefficient vector of the corrected status vector indicators, and exp is the exponential function.

[0009] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention proposes a state assessment scheme for oil well equipment based on machine vision. Based on machine vision technology, image data of the associated components of the oil well equipment are collected through an image sensor and their real-time states are analyzed. Combining the time series analysis of historical operation and maintenance data to evaluate component wear, and then correcting the real-time status indicators and fitting the to-be-exploited tasks to evaluate the abnormal risk probability, so as to accurately judge the equipment status. The beneficial effects are as follows: accurately predict the abnormal risks of the associated components of oilfield equipment and reduce the maintenance costs of oilfield equipment. Description of the Drawings

[0010] Figure 1 is a flow chart of a method for assessing the state of oil well equipment based on machine vision; Figure 2 is a flow chart of a method for analyzing the real-time status indicators of the associated components of oil well equipment; Figure 3 is a flow chart of a method for evaluating the historical operation wear coefficient of the associated components of oil well equipment; Figure 4 is a flow chart of a method for fitting and evaluating the abnormal risk probability of the associated components of oil well equipment. Detailed Embodiments

[0011] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations.

[0012] Refer toFigure 1 As shown in the figure, a method for evaluating the status of oil well equipment based on machine vision includes: Based on an image sensor, acquiring the image data of the associated components of each oil well equipment, and analyzing the real-time status indicators of the associated components of the oil well equipment; Obtaining the historical operation and maintenance data of the oil well equipment, performing time series analysis on the historical operation and maintenance data, and evaluating the historical operation loss coefficient of the associated components of the oil well equipment; Based on the historical operation loss coefficient of the associated components of the oil well equipment, correcting the real-time status indicators of the associated components of the oil well equipment to obtain the corrected status indicators of the associated components of the oil well equipment; Obtaining the to-be-exploited tasks of the oil well equipment and fitting them with the corrected status indicators of the associated components of the oil well equipment to evaluate the abnormal risk probability of the associated components of the oil well equipment; Judging whether the abnormal risk probability of the associated components of the oil well equipment is at the dangerous critical value. If not, it is determined to be executed normally. If so, it is determined that the equipment is abnormal; Among them, the corrected status indicator of the associated component of the oil well equipment is specifically: G‘ ij =G ij ×H ij ; In the formula, G‘ ij is the corrected status indicator of the jth associated component of the ith oil well equipment, G ij is the real-time status indicator of the jth associated component of the ith oil well equipment, and H ij is the historical operation loss coefficient of the jth associated component of the ith oil well equipment.

[0013] This solution is based on machine vision technology. By using an image sensor to collect the image data of the associated components of the oil well equipment and analyzing their real-time status, combined with the time series analysis of the historical operation and maintenance data to evaluate the component loss, and then correcting the real-time status indicators and fitting them with the to-be-exploited tasks to evaluate the abnormal risk probability, so as to accurately judge the equipment status. The beneficial effects are as follows: accurately predicting the abnormal risk of the associated components of the oil field equipment and reducing the maintenance cost of the oil field equipment.

[0014] Referring to Figure 2 shown in the figure, based on an image sensor, acquiring the image data of the associated components of each oil well equipment and analyzing the real-time status indicators of the associated components of the oil well equipment specifically include: Based on the structural parameters of the associated components of the oil well equipment, determining the initial image data of the associated components of the oil well equipment; Based on the image data of the associated components of each oil well equipment, using Gaussian filtering to eliminate the noise in the associated component images and generating the preprocessed image data of the associated components of the oil well equipment; For the image data of the associated components of the preprocessed oil well equipment, extract the edge information and shape texture information of the associated component images to obtain the image feature data of the associated components of the oil well equipment; Based on the image feature data of the associated components of the oil well equipment and the initialization image data of the associated components of the oil well equipment, package them into a positive and negative sample dataset for the loss of the associated components of the oil well equipment; Based on the SVM support vector, construct a state analysis model for the associated components of the oil well equipment; Substitute the positive and negative sample dataset for the loss of the associated components of the oil well equipment into the state analysis model for the associated components of the oil well equipment. Use the image feature data of the associated components of the oil well equipment as positive samples, and use the initialization image data of the associated components of the oil well equipment as negative samples (using the image feature data of the associated components as negative samples) to evaluate the real-time state index of the associated components of the oil well equipment; Among them, the state analysis model for the associated components of the oil well equipment is specifically: ; In the formula, G ij is the real-time state index of the jth associated component of the ith oil well equipment, w is the weight vector, F ij is the feature vector of the jth associated component of the ith oil well equipment, b is the bias term, C is the regularization parameter, ξ ij is the slack variable of the jth associated component of the ith oil well equipment, y ij is the true label of the jth associated component of the ith oil well equipment, ξ ij ≥0, which is a non-negative value constraint condition, is the constraint condition, N + is the total number of positive samples for the loss of the associated components of the oil well equipment, N - is the total number of negative samples for the loss of the associated components of the oil well equipment.

[0015] This solution collects the images of the associated components of the oil well equipment through an image sensor, extracts features after denoising with Gaussian filtering, constructs a positive and negative sample dataset for loss, and trains and analyzes based on the SVM model to achieve real-time state evaluation. The beneficial effects are as follows: It can accurately capture the state of the oil well equipment and adapt to the monitoring needs of different oil well equipment and components.

[0016] Refer to Figure 3 As shown, obtain the historical operation and maintenance data of the oil well equipment, conduct time series analysis on the historical operation and maintenance data, and evaluate the historical operation loss coefficient of the associated components of the oil well equipment, specifically including: Obtain the standardized operation parameters of the associated components of the oil well equipment and form an array of standardized operation parameters of the associated components of the oil well equipment; Based on the historical operation and maintenance data of oil well equipment, taking the time interval of historical mining tasks as the observation window and the historical operation and maintenance data of oil well equipment as the observation variables, an observation time series window is established; Based on the observation time series window, obtain the execution process time series parameters of the associated components of the oil well equipment in historical mining tasks; According to the standardized operation parameter array of the associated components of the oil well equipment, screen out the deviation parameters of the operation time series parameters of the associated components of the oil well equipment in historical mining tasks, denoted as the execution process time series parameters of the associated components of the oil well equipment in the deviation tasks; According to the execution process time series parameters of the associated components of the oil well equipment in historical mining tasks and the execution process time series parameters of the associated components of the oil well equipment in deviation tasks, form a characteristic data set of performance loss factors of the associated components of the oil well equipment Based on the characteristic data set of performance loss factors of the associated components of the oil well equipment, estimate the basic performance loss of the execution process time series parameters of the associated components of the oil well equipment in historical mining tasks as the performance loss constant, estimate the deviation performance loss of the execution process parameters of the associated components of the oil well equipment in deviation tasks as the performance loss variable, and evaluate the historical operation loss coefficient of the associated components of the oil well equipment; Among them, the performance degradation regression model of the associated components of the oil well equipment is specifically: ; In the formula, H ij is the historical operation loss coefficient of the jth associated component of the ith oil well equipment, α0 is the intercept term, V ij is the basic performance loss constant of the jth associated component of the ith oil well equipment, V' ij is the deviation performance loss variable of the jth associated component of the ith oil well equipment, α1 is the performance loss constant coefficient, α2 is the performance loss variable coefficient, and ∈ is the error term.

[0017] This solution collects the standardized operation parameters and historical operation and maintenance data of the associated components of the oil well equipment, constructs an observation time series window and analyzes the execution process time series parameters, screens out the deviation parameters, and then forms a characteristic data set of performance loss factors, thereby estimating the basic performance loss and deviation performance loss, and evaluating the historical operation loss coefficient of the associated components of the oil well equipment. The beneficial effect is that it can accurately quantify the loss situation of the associated components of the oil well equipment and provide a basis for subsequent equipment failure warning.

[0018] Refer to Figure 4 As shown, obtaining the to-be-mined tasks of the oil well equipment and the corrected status indicators of the associated components of the oil well equipment for fitting and evaluating the abnormal risk probability of the associated components of the oil well equipment specifically includes: Based on logistic regression, construct a risk assessment model for the associated components of the oil well equipment; Obtain the to-be-exploited tasks of the oil well equipment, determine the task execution process of the associated components of the oil well equipment for the to-be-exploited tasks, and obtain the execution parameters of the associated components of the oil well equipment for the to-be-exploited tasks; Based on the corrected status indicators of the associated components of the oil well equipment, linearly map each corrected status indicator of the associated components into the corrected status vector indicators of the associated components of the oil well equipment, and substitute them into the risk assessment model of the associated components of the oil well equipment, using the execution parameters of the associated components of the oil well equipment for the to-be-exploited tasks as the input and the abnormal risk probability of the associated components of the oil well equipment as the output; Among them, the risk assessment model of the associated components of the oil well equipment is specifically: ; In the formula, P ij is the abnormal risk probability of the j-th associated component of the i-th oil well equipment, β0 is the intercept term, S ij is the execution parameter of the j-th associated component of the i-th oil well equipment for the to-be-exploited tasks, β1 is the correlation coefficient vector of the execution parameters, β2 is the correlation coefficient vector of the corrected status vector indicators, and exp is the exponential function.

[0019] This solution determines the execution parameters of the associated components during the task execution process, linearly maps the corrected status indicators of each associated component into corrected status vector indicators, and uses the to-be-executed parameters and corrected values to train the Logistic regression risk assessment model. The model calculates the abnormal risk probability of the associated components of the oil well equipment according to the input data. It can accurately quantify the abnormal risks of the associated components and provide rationalized data for equipment maintenance.

[0020] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the state of oil well equipment based on machine vision, characterized in that: include: Based on image sensors, image data of associated components of each oil well equipment is collected, and real-time status indicators of associated components of the oil well equipment are analyzed; Obtain historical operation and maintenance data of oil well equipment, conduct time series analysis on the historical operation and maintenance data, and evaluate the historical operation loss coefficients of related components of oil well equipment; Correcting the real-time status index of the associated components of the oil well equipment based on the historical operation loss coefficient of the associated components of the oil well equipment to obtain a corrected status index of the associated components of the oil well equipment; Obtaining the pending mining tasks of the oil well equipment and the correction status indicators of the associated components of the oil well equipment to perform fitting and evaluate the abnormal risk probability of the associated components of the oil well equipment; Determine whether the abnormal risk probability of the associated components of the oil well equipment is at a critical value. If not, determine that the operation is normal. If so, determine that the equipment is abnormal. Among them, the correction status index of the associated components of the oil well equipment is specifically: G' ij =G ij ×H ij ; In the formula, G' ij Correct the status index of the jth associated component of the i-th oil well equipment, G ij is the real-time status indicator of the jth associated component of the i-th oil well equipment, H ij is the historical operating loss coefficient of the jth associated component of the i-th oil well equipment.

2. The method for evaluating the state of oil well equipment based on machine vision according to claim 1, characterized in that: Based on the image sensor, the image data of the associated components of each oil well equipment is collected, and the real-time status indicators of the associated components of the oil well equipment are analyzed, including: Determining initialization image data of the associated components of the oil well equipment based on the construction parameters of the associated components of the oil well equipment; Based on the associated component image data of each oil well equipment, the noise in the associated component image is eliminated by using Gaussian filtering to generate the associated component image data of the pre-processed oil well equipment; Preprocessing the image data of the associated components of the oil well equipment, extracting the edge information and shape texture information of the associated component images, and obtaining the image feature data of the associated components of the oil well equipment; Based on the image feature data of the associated components of the oil well equipment and the initialization image data of the associated components of the oil well equipment, the data are packaged into a loss positive and negative sample data set of the associated components of the oil well equipment; Based on SVM support vector, a state analysis model of associated components of oil well equipment is constructed; The positive and negative sample data sets of associated component loss of oil well equipment are substituted into the associated component status analysis model of oil well equipment. The image feature data of the associated components of the oil well equipment are used as positive samples, and the initialized image data of the associated components of the oil well equipment are used as negative samples to evaluate the real-time status indicators of the associated components of the oil well equipment.

3. The method for evaluating the state of oil well equipment based on machine vision according to claim 2, characterized in that: The state analysis model of the associated components of oil well equipment is as follows: ； In the formula, G ij is the real-time status index of the jth associated component of the i-th oil well equipment, w is the weight vector, F ij is the eigenvector of the jth associated component of the i-th oil well equipment, b is the bias term, C is the regularization parameter, ξ ij is the slack variable of the jth associated component of the i-th oil well equipment, y ij is the true label of the jth associated component of the i-th oil well equipment, ξ ij ≥0, is the value non-negative constraint, is the constraint condition, N + is the total number of positive samples of associated component loss of oil well equipment, N - is the total number of negative samples lost for associated components of oil well equipment.

4. The method for evaluating the state of oil well equipment based on machine vision according to claim 3, characterized in that: Obtain the historical operation and maintenance data of oil well equipment, conduct time series analysis on the historical operation and maintenance data, and evaluate the historical operation loss coefficients of the associated components of the oil well equipment. Specifically include: Obtaining standardized operating parameters of associated components of oil well equipment, and forming an array of standardized operating parameters of associated components of oil well equipment; Based on the historical operation and maintenance data of oil well equipment, the time interval of historical mining tasks is used as the observation window, and the historical operation and maintenance data of oil well equipment is used as the observation variable to establish the observation time series window; Based on the observation time series window, the execution process time series parameters of the associated components of the oil well equipment in the historical mining tasks are obtained; According to the standardized operation parameter array of the associated components of the oil well equipment, the deviation parameters of the operation timing parameters of the associated components of the oil well equipment in the historical mining tasks are screened out, and recorded as the execution process timing parameters of the associated components of the oil well equipment in the deviation tasks; According to the execution process timing parameters of the associated components of oil well equipment in historical mining tasks and the execution process timing parameters of the associated components of oil well equipment in deviation tasks, a characteristic dataset of performance loss factors of associated components of oil well equipment is established Based on the characteristic data set of performance loss factors of associated components of oil well equipment, the basic performance loss of execution process timing parameters of associated components of oil well equipment in historical mining tasks is estimated as the performance loss constant, the deviation performance loss of execution process parameters of associated components of oil well equipment in deviation tasks is estimated as the performance loss variable, and the historical operation loss coefficient of associated components of oil well equipment is evaluated.

5. The method for evaluating the state of oil well equipment based on machine vision according to claim 4, characterized in that: The performance degradation regression model of associated components of oil well equipment is as follows: ； In the formula, H ij is the historical operating loss coefficient of the jth associated component of the i-th oil well equipment, α0 is the intercept term, V ij is the basic performance loss constant of the jth associated component of the i-th oil well equipment, V' ij is the deviation performance loss variable of the jth associated component of the i-th oil well equipment, α1 is the performance loss constant coefficient, α2 is the performance loss variable coefficient, and ∈ is the error term.

6. The method for evaluating the state of oil well equipment based on machine vision according to claim 5, characterized in that: Obtaining the pending mining tasks of the oil well equipment and the correction status indicators of the associated components of the oil well equipment to perform fitting evaluation on the abnormal risk probability of the associated components of the oil well equipment specifically includes: Based on logistic regression, a risk assessment model for associated components of oil well equipment is constructed; Acquire the to-be-mined tasks of the oil well equipment, determine the execution process of the associated components of the oil well equipment for the to-be-mined tasks, and obtain the execution parameters of the associated components of the oil well equipment for the to-be-mined tasks; Based on the corrected state indicators of the associated components of the oil well equipment, each corrected state indicator of the associated components is linearly mapped into the corrected state vector indicator of the associated components of the oil well equipment, and substituted into the associated component risk assessment model of the oil well equipment. The associated component execution parameters of the oil well equipment to be mined are taken as input, and the abnormal risk probability of the associated components of the oil well equipment is taken as output.

7. The method for evaluating the state of oil well equipment based on machine vision according to claim 6, characterized in that: The risk assessment model for the associated components of the oil well equipment is specifically as follows: ； Where P ij is the abnormal risk probability of the jth associated component of the i-th oil well equipment, β0 is the intercept term, S ij is the execution parameter of the jth associated component of the i-th oil well equipment to be mined, β1 is the execution parameter correlation coefficient vector, β2 is the correction state vector indicator correlation coefficient vector, and exp is the exponential function.

Images