A comprehensive fault diagnosis method and system for intermittent pumping units

Abstract

The present invention discloses a method and system for comprehensive diagnosis of intermittent pumping unit faults, which divides a preset monitoring time period by a predetermined intermittent operation cycle of the pumping unit to determine available electrical parameters and pumping speed parameters, and then converts and calculates the retained electrical parameters and pumping speed parameters to obtain corrected load data and corrected displacement data with higher accuracy, and finally determines the fault type of the pumping unit within the preset monitoring time period by calculation based on a second effective dynamometer diagram, that is, in the entire comprehensive fault diagnosis process, there is no need to set up a special load sensor for dynamometer diagram detection, but a more accurate dynamometer diagram is obtained by parameter conversion, which has better recognition effect and lower cost, and takes into account the actual impact of the intermittent operation cycle of the pumping unit on the preset monitoring time period, ensuring that good fault diagnosis can be performed under any circumstances, which is beneficial to improving the accuracy of pumping unit fault identification.

Description

Technical Field

[0001] The invention relates to the technical field of oilfield production management, in particular to a method and system for comprehensive fault diagnosis of an intermittent pumping unit, and a computer-readable storage medium. Background Art

[0002] In order to improve the oil production efficiency and economic benefits of oil and gas fields and reduce the use burden of pumping units, the pumping unit intermittent control method has been more widely used. In the process of oil well development, the real-time working status of the oil well is usually identified and judged through the indicator diagram to determine the fault condition of the pumping unit. This is a commonly used analysis method in this field. At present, the indicator diagram can be obtained by setting up a special load sensor. However, considering the harsh and changeable environment of oil well development, the data collected by the load sensor is often distorted or affected by environmental interference and the accuracy is reduced, making the obtained indicator diagram more difficult to identify and cannot be applied to a wider range of oil well fault identification scenarios. Summary of the invention

[0003] The present invention aims to solve one of the technical problems in the related art to at least a certain extent. To this end, the present invention proposes a method and system for comprehensive diagnosis of faults of intermittent pumping units, which can effectively improve the recognition accuracy of indicator diagrams and the accuracy of pumping unit fault recognition.

[0004] In a first aspect, an embodiment of the present invention provides a method for comprehensive fault diagnosis of an intermittent pumping unit, comprising the following steps:

[0005] Step S1, continuously acquiring electrical parameters and pumping speed parameters of the oil pumping unit within a preset monitoring time period, dividing the preset monitoring time period according to a predetermined intermittent operation cycle of the oil pumping unit, so as to divide the preset monitoring time period into at least one differentiated monitoring time period;

[0006] Step S2, for each of the distinguished monitoring time periods, judging whether the distinguished monitoring time period is an available monitoring time period according to the electrical parameters and the pumping speed parameters corresponding to the distinguished monitoring time period, if it is not an available monitoring time period, deleting all the electrical parameters and the pumping speed parameters corresponding to the distinguished monitoring time period, otherwise retaining all the electrical parameters and the pumping speed parameters corresponding to the distinguished monitoring time period, and executing step S3;

[0007] Step S3, converting and calculating all the electrical parameters and the pumping speed parameters that have been retained to obtain all corresponding corrected load data and corrected displacement data, and drawing a corresponding first effective dynamometer diagram according to all the corrected load data and the corrected displacement data;

[0008] Step S4, extracting all feature vectors of the first effective indicator diagram, inputting all the feature vectors into a pre-trained feature extraction network, obtaining effective features corresponding to all the feature vectors output by the feature extraction network, and updating the first effective indicator diagram according to the effective features to obtain a second effective indicator diagram;

[0009] Step S5, obtaining several key characteristic values ​​of the second effective indicator diagram, and preprocessing all the key characteristic values ​​to obtain corresponding standard characteristic values;

[0010] Step S6: for each of the standard characteristic values, assign a corresponding fault impact weight to the standard characteristic value according to the difference between the standard characteristic value and the preset reference characteristic value; determine the fault parameter evaluation value corresponding to the second effective indicator diagram according to all the fault impact weights and the standard characteristic values, and determine the fault type of the oil pump within the preset monitoring time period based on the fault parameter evaluation value, the first preset fault parameter reference value and the second preset fault parameter reference value, wherein the second preset fault parameter reference value is greater than the first preset fault parameter reference value.

[0011] Optionally, in an embodiment of the present invention, the step in step S6, assigning a corresponding fault impact weight to the standard feature value according to the difference between the standard feature value and the preset reference feature value, comprises the following steps:

[0012] Step S61, calculating the difference between the standard characteristic value and the preset reference characteristic value;

[0013] Step S62: Based on a preset weight influence calculation formula, in combination with the difference between the standard characteristic value and the preset reference characteristic value and all the standard characteristic values, the fault influence proportion corresponding to the standard characteristic value is calculated, wherein the weight influence calculation formula is as follows:

[0014]

[0015] S i is the ith standard characteristic value, P is the fault impact proportion corresponding to the ith standard characteristic value, S0 is the preset reference characteristic value, and n is the total number of the standard characteristic values.

[0016] Optionally, in one embodiment of the present invention, the step in step S6, determining the fault parameter evaluation value corresponding to the second effective dynamometer diagram according to all the fault impact proportions and the standard characteristic values, comprises the following steps:

[0017] Step S63, respectively calculating the product between each of the standard characteristic values ​​and the corresponding fault impact proportion to obtain the fault impact amount; screening out the largest standard characteristic value and the smallest standard characteristic value from all the standard characteristic values;

[0018] Step S64: Substitute each of the fault impact quantities, the largest standard characteristic value, and the smallest standard characteristic value into a preset fault parameter evaluation formula to calculate a fault parameter evaluation value corresponding to the second effective dynamometer diagram, wherein the fault parameter evaluation formula is as follows:

[0019]

[0020] Q is the fault parameter evaluation value corresponding to the second effective dynamometer diagram, S max is the largest standard eigenvalue, S min is the smallest one of the standard eigenvalues, O j is the jth fault impact quantity, and m is the total number of fault impact quantities.

[0021] Optionally, in one embodiment of the present invention, the step in step S6, based on the fault parameter evaluation value, the first preset fault parameter reference value and the second preset fault parameter reference value, determines the fault type of the oil pump within the preset monitoring time period, including the following steps:

[0022] Step S65: when the fault parameter evaluation value is less than or equal to the first preset fault parameter reference value, determining that the fault type of the oil pumping unit within the preset monitoring time period is an oil pumping unit communication fault;

[0023] or,

[0024] When the fault parameter evaluation value is greater than the first preset fault parameter reference value and less than the second preset fault parameter reference value, determining that the fault type of the oil pumping unit within the preset monitoring time period is an oil pumping unit instrument fault;

[0025] or,

[0026] When the fault parameter evaluation value is greater than or equal to the second preset fault parameter reference value, it is determined that the fault type of the oil pumping unit within the preset monitoring time period is an electrical fault of the oil pumping unit.

[0027] Optionally, in one embodiment of the present invention, the electrical parameters include a first pumping unit operating voltage parameter, a first pumping unit operating current parameter and a first stroke conversion parameter, and the pumping speed parameter includes a first oil pump efficiency parameter and a first oil pump cleaning parameter. The steps in step S3, converting and calculating all the retained electrical parameters and the pumping speed parameters to obtain all corresponding corrected load data and corrected displacement data, include the following steps:

[0028] Step S31, dividing the electrical parameters and the pumping speed parameters corresponding to any one of the differentiated monitoring time periods into a group of associated data;

[0029] Step S32: for each group of the associated data, normalize the first pumping unit operating voltage parameter, the first pumping unit operating current parameter, the first stroke conversion parameter, the first oil pump efficiency parameter and the first oil pump cleaning parameter to obtain the corresponding second pumping unit operating voltage parameter, the second pumping unit operating current parameter, the second stroke conversion parameter, the second oil pump efficiency parameter and the second oil pump cleaning parameter respectively;

[0030] Step S33, substituting the second pumping unit operating voltage parameter, the second pumping unit operating current parameter, the second stroke conversion parameter, the second oil pump efficiency parameter and the second oil pump cleaning parameter into a preset work indicator data correction formula to calculate corresponding corrected load data and corrected displacement data;

[0031] The correction formula of the indicator data is as follows:

[0032]

[0033] M is the corrected load data, N is the corrected displacement data, V is the operating voltage parameter of the second oil pumping unit, I is the operating current parameter of the second oil pumping unit, H is the second stroke conversion parameter, k is the preset displacement conversion factor, G1 is the second oil pump efficiency parameter, G2 is the second oil pump cleaning parameter, T1 is the corresponding start time of the differentiated monitoring time period, and T2 is the corresponding end time of the differentiated monitoring time period.

[0034] Optionally, in one embodiment of the present invention, the intermittent operation cycle includes a partially open operation cycle and a partially closed operation cycle, wherein in the partially open operation cycle, the oil pump operates normally; in the partially closed operation cycle, the oil pump stops operating;

[0035] The step in step S1, dividing the preset monitoring time period according to the predetermined intermittent operation cycle of the pumping unit to divide the preset monitoring time period into at least one differentiated monitoring time period, includes the following steps:

[0036] Step S11, when the preset monitoring time period completely overlaps with the partially started operation cycle, the preset monitoring time period is used as the first differentiated monitoring time period;

[0037] or,

[0038] When the preset monitoring time period completely overlaps with the partially closed operation cycle, the preset monitoring time period is used as the second differentiated monitoring time period;

[0039] or,

[0040] When a part of the preset monitoring time period coincides with the partially open operating cycle and another part of the preset monitoring time period coincides with the partially closed operating cycle, the preset monitoring time period is divided into a third differentiated monitoring time period located within the partially open operating cycle and a fourth differentiated monitoring time period located within the partially closed operating cycle.

[0041] In a second aspect, an embodiment of the present invention provides a comprehensive fault diagnosis system for an intermittent pumping unit, comprising:

[0042] at least one processor;

[0043] at least one memory for storing at least one program;

[0044] When at least one of the programs is executed by at least one of the processors, the comprehensive fault diagnosis method for the intermittent pumping unit as described in the first aspect is implemented.

[0045] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to implement the comprehensive diagnosis method for faults of intermittent pumping units as described in the first aspect.

[0046] The present invention proposes a method and system for comprehensive diagnosis of faults of intermittent pumping units. The preset monitoring time period is divided by the intermittent operation cycle of the pumping unit, so as to make real-time judgment on whether the corresponding electrical parameters and pumping speed parameters of each distinguished monitoring time period are available, thereby determining the available and effective electrical parameters and pumping speed parameters, and then converting and calculating the retained electrical parameters and pumping speed parameters to obtain corrected load data and corrected displacement data with higher accuracy. On this basis, by extracting the effective features in the preliminary first effective dynamometer diagram, a second effective dynamometer diagram with better recognition accuracy is obtained, and finally the fault type of the pumping unit within the preset monitoring time period can be determined by performing relevant operations based on the second effective dynamometer diagram. That is to say, in the whole comprehensive fault diagnosis process, there is no need to set a special load sensor for dynamometer diagram detection, but a more accurate dynamometer diagram is obtained by parameter conversion, the recognition effect is better and the cost is reduced, and the actual influence of the intermittent operation cycle of the pumping unit on the preset monitoring time period is taken into consideration. By dividing the corresponding distinguished monitoring time periods, it is ensured that good and effective fault diagnosis can be performed under any monitoring conditions, which is conducive to improving the accuracy of pumping unit fault identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of a comprehensive fault diagnosis method for an intermittent pumping unit provided by an embodiment of the present invention;

[0048] Figure 2 yes Figure 1 A partial flow chart of the step S1 of "dividing the preset monitoring time period according to the predetermined intermittent operation cycle of the pumping unit to divide the preset monitoring time period into at least one differentiated monitoring time period";

[0049] Figure 3 yes Figure 1 Partial flow chart of the step S3 of "converting and calculating all the retained electrical parameters and pumping speed parameters to obtain all the corresponding corrected load data and corrected displacement data";

[0050] Figure 4 yes Figure 1 A partial flow chart of the step S6 of “giving a corresponding fault impact weight to the standard characteristic value according to the difference between the standard characteristic value and the preset reference characteristic value”;

[0051] Figure 5 yes Figure 1 A partial flow chart of the step "determining the fault parameter evaluation value corresponding to the second effective dynamometer diagram according to all fault influence proportions and standard characteristic values" in step S6;

[0052] Figure 6 yes Figure 1 A partial flow chart of the step S6 of “determining the fault type of the oil pumping unit within a preset monitoring time period based on the fault parameter evaluation value, the first preset fault parameter reference value and the second preset fault parameter reference value”;

[0053] Figure 7 The present invention is a schematic diagram of the structure of a comprehensive fault diagnosis system for an intermittent pumping unit provided in accordance with an embodiment of the present invention. DETAILED DESCRIPTION

[0054] like Figure 1 As shown, an embodiment of the present invention provides a method for comprehensive fault diagnosis of an intermittent pumping unit, which may include but is not limited to steps S1 to S6.

[0055] Step S1, continuously acquiring electrical parameters and pumping speed parameters of the oil pumping unit within a preset monitoring time period, dividing the preset monitoring time period according to a predetermined intermittent operation cycle of the oil pumping unit, so as to divide the preset monitoring time period into at least one differentiated monitoring time period, wherein the preset monitoring time period can be arbitrarily selected and set, and no matter how the preset monitoring time period is selected and set, the preset monitoring time period can be divided according to the predetermined intermittent operation cycle of the oil pumping unit; the number of differentiated monitoring time periods can be one or more, and when it is one, it is equivalent to taking the preset monitoring time period as a differentiated monitoring time period, and when it is multiple, it is necessary to divide the preset monitoring time period into multiple differentiated monitoring time periods;

[0056] Step S2, for each differentiated monitoring time period, judging whether the differentiated monitoring time period is an available monitoring time period according to the electrical parameters and the pumping speed parameters corresponding to the differentiated monitoring time period, if it is not an available monitoring time period, deleting all the electrical parameters and the pumping speed parameters corresponding to the differentiated monitoring time period, otherwise retaining all the electrical parameters and the pumping speed parameters corresponding to the differentiated monitoring time period, and executing step S3;

[0057] Step S3, convert and calculate all the reserved electrical parameters and pumping speed parameters to obtain all the corresponding corrected load data and corrected displacement data, and draw the corresponding first effective indicator diagram according to all the corrected load data and corrected displacement data, wherein drawing the corresponding indicator diagram according to the load data and the displacement data belongs to the related prior art well known to those skilled in the art, and the indicator diagram can be obtained by selecting the corresponding method in different application scenarios, and will not be described here to avoid redundancy;

[0058] Step S4, extracting all feature vectors of the first effective indicator diagram, inputting all feature vectors into a pre-trained feature extraction network, obtaining effective features corresponding to all feature vectors output by the feature extraction network, and updating the first effective indicator diagram according to the effective features to obtain a second effective indicator diagram;

[0059] Step S5, obtaining several key characteristic values ​​of the second effective indicator diagram, and preprocessing all the key characteristic values ​​to obtain corresponding standard characteristic values;

[0060] Step S6: for each standard characteristic value, assign a corresponding fault impact weight to the standard characteristic value according to the difference between the standard characteristic value and the preset reference characteristic value; determine the fault parameter evaluation value corresponding to the second effective dynamometer diagram according to all fault impact weights and the standard characteristic values, and determine the fault type of the oil pump within the preset monitoring time period based on the fault parameter evaluation value, the first preset fault parameter reference value and the second preset fault parameter reference value, wherein the second preset fault parameter reference value is greater than the first preset fault parameter reference value.

[0061] In this step, the preset monitoring time period is divided by the predetermined intermittent operation cycle of the oil pumping unit, so as to make real-time judgment on whether the corresponding electrical parameters and oil pumping speed parameters of each distinguished monitoring time period are available, thereby determining the available and effective electrical parameters and oil pumping speed parameters, and then converting and calculating the retained electrical parameters and oil pumping speed parameters to obtain more accurate corrected load data and corrected displacement data. On this basis, by extracting the effective features in the preliminary first effective indicator diagram, a second effective indicator diagram with better recognition accuracy is updated, and finally the relevant calculation can be performed based on the second effective indicator diagram to determine the fault type of the oil pumping unit within the preset monitoring time period. That is to say, in the whole comprehensive fault diagnosis process, there is no need to set up a special load sensor for indicator diagram detection, but a more accurate indicator diagram is obtained by parameter conversion, the recognition effect is better and the cost is reduced, and the actual influence of the intermittent operation cycle of the oil pumping unit on the preset monitoring time period is taken into account. By dividing the corresponding distinguished monitoring time periods, it is ensured that no matter what the monitoring situation is, good and effective fault diagnosis can be carried out, which is conducive to improving the accuracy of oil pumping unit fault identification.

[0062] It can be understood that, since the merged target management data is a combination of dynamic well opening data and static well shut-in data with similar or close timing characteristics, the merged target management data has a relatively integrated timing characteristic, which indicates that the corresponding dynamic well opening data and static well shut-in data at this time have good timing adaptability. Since the timestamps corresponding to the dynamic well opening data and the static well shut-in data are known, it is equivalent to providing inspiration for the inter-well opening control method of oil and gas fields, that is, it prompts that the dynamic well opening data and static well shut-in data at the timestamp corresponding to such a combination are relatively better, so as to instruct users or background control systems how to better and more effectively perform inter-well opening control of oil and gas fields.

[0063] like Figure 2As shown, in one embodiment of the present invention, when the intermittent operation cycle includes a partially open operation cycle and a partially closed operation cycle, wherein in the partially open operation cycle, the oil pump operates normally; in the partially closed operation cycle, the oil pump stops operating, the steps in step S1, according to the predetermined intermittent operation cycle of the oil pump, divide the preset monitoring time period into at least one differentiated monitoring time period, specifically may include but not limited to the following steps:

[0064] Step S11, when the preset monitoring time period completely overlaps with the partially open operation cycle, the preset monitoring time period is used as the first differentiated monitoring time period; or, when the preset monitoring time period completely overlaps with the partially closed operation cycle, the preset monitoring time period is used as the second differentiated monitoring time period; or, when a part of the preset monitoring time period overlaps with the partially open operation cycle and another part of the preset monitoring time period overlaps with the partially closed operation cycle, the preset monitoring time period is divided into a third differentiated monitoring time period located in the partially open operation cycle and a fourth differentiated monitoring time period located in the partially closed operation cycle.

[0065] In this step, the overlap relationship between the preset monitoring time period and the partially open operation cycle and the partially closed operation cycle is judged to further determine the specific operation status of the oil pump within the preset monitoring time period. That is, if the preset monitoring time period completely overlaps with the partially open operation cycle or the partially closed operation cycle, it means that the oil pump within the preset monitoring time period is fully working or not working at all. In this case, the relevant oil pump data within the preset monitoring time period are integrated and related, and there is not much difference. Therefore, there is no need to divide the preset monitoring time period. Otherwise, it means that the oil pump within the preset monitoring time period has different working states. Therefore, in this case, the oil pump data corresponding to different working states needs to be divided. That is to say, the preset monitoring time period is divided into a third divided monitoring time period located in the partially open operation cycle and a fourth divided monitoring time period located in the partially closed operation cycle, so that the oil pump data are studied and analyzed in the third divided monitoring time period and the fourth divided monitoring time period respectively. This is conducive to more accurate and reliable analysis of the real-time working conditions of the pumping unit. It should be noted that the partial opening operation cycle and the partial closing operation cycle need to be determined accordingly according to the actual scenario. The following is a detailed example. For example, assuming that the intermittent operation cycle is from 09:00 on December 26 to 18:00 on December 27, the partial opening operation cycle is set from 09:00 on December 26 to 10:00 on December 27 according to the production situation, and the partial closing operation cycle is set from 09:00 on December 26 to 10:00 on December 27. From 10:00 on December 27 to 18:00 on December 27, if the preset monitoring time period for sampling is from 20:00 on December 26 to 14:00 on December 27, then the preset monitoring time period will be divided into 20:00 on December 26 to 10:00 on December 27 (i.e. the third zone monitoring time period) and from 10:00 on December 27 to 14:00 on December 27 (i.e. the fourth zone monitoring time period), and then the pumping unit data in these two monitoring time periods will be analyzed respectively.

[0066] In one embodiment, in the step S2, for each differentiated monitoring time period, according to each electrical parameter and oil pumping speed parameter corresponding to the differentiated monitoring time period, there may be multiple specific ways to determine whether the differentiated monitoring time period is an available monitoring time period, which are illustrated by the following examples. For example, when the preset monitoring time period is used as the first differentiated monitoring time period, it can be but not limited to simultaneously determining whether the electrical parameter falls within the electrical preset threshold range and whether the oil pumping speed parameter falls within the oil pumping speed preset threshold range. If both fall within, the first differentiated monitoring time period is determined to be an available monitoring time period, otherwise all electrical parameters and oil pumping speed parameters corresponding to the first differentiated monitoring time period are deleted; when the preset monitoring time period is used as the second differentiated monitoring time period, it can be but not limited to only determining whether the electrical parameter falls within the electrical preset threshold range. If so, the second differentiated monitoring time period can be determined to be an available monitoring time period, otherwise all electrical parameters and oil pumping speed parameters corresponding to the second differentiated monitoring time period are deleted; when the preset monitoring time period is divided into a third differentiated monitoring time period and a fourth differentiated monitoring time period, the third differentiated monitoring time period can refer to the determination method of the first differentiated monitoring time period, and the fourth differentiated monitoring time period can refer to the determination method of the second differentiated monitoring time period, which will not be elaborated here.

[0067] It should be noted that the specific types and values ​​of electrical parameters and oil pumping speed parameters in different situations can be set accordingly according to the actual scenario. For example, the electrical parameters may include but are not limited to the first oil pumping unit operating voltage parameter, the first oil pumping unit operating current parameter and the first stroke conversion parameter, and the oil pumping speed parameters may include but are not limited to the first oil pump efficiency parameter and the first oil pump cleaning parameter. Accordingly, the electrical preset threshold range and the oil pumping speed preset threshold range are also set accordingly. For example, when the electrical parameter adopts the first oil pumping unit operating voltage parameter, the electrical preset threshold range is set to the voltage preset threshold range, and so on. The oil pumping speed preset threshold range can be set accordingly, which will not be elaborated here.

[0068] In one embodiment, in step S4, there may be multiple ways to extract all feature vectors of the first effective dynamometer diagram, such as obtaining the histogram features in the first effective dynamometer diagram, and then reading the corresponding feature vector through the histogram features, or directly collecting image features of the first effective dynamometer diagram, and then encoding the collected image features and converting the encoded features into corresponding numerical vectors, so as to obtain the feature vectors of the first effective dynamometer diagram, which is not limited here; the feature extraction network in step S4 may be, but is not limited to, a convolutional neural network, a recurrent neural network, and a Transformer, etc. Taking the convolutional neural network as an example, firstly, the feature vectors of the first effective dynamometer diagram corresponding to the historical monitoring time period are normalized, and then all the normalized feature vectors are input into the input layer of the convolutional neural network, and finally, the trained feature values ​​are output through the output layer of the convolutional neural network, and the feature values ​​are the effective features of the corresponding feature vectors. It should be noted that the effective features finally outputted do not necessarily correspond to each feature vector one-to-one, because some feature vectors may not have corresponding effective features, and then for these feature vectors, the convolutional neural network does not output the corresponding effective features.

[0069] In one embodiment, the step in step S4, updating the first effective dynamometer diagram according to the effective features, and obtaining the second effective dynamometer diagram in a specific manner can be multiple, for example, adding the acquired effective features to the remaining portion of the first effective dynamometer diagram, thereby fusing the effective features with the first effective dynamometer diagram, which is equivalent to updating the first effective dynamometer diagram to obtain the second effective dynamometer diagram, or decomposing the effective features to obtain multiple separate more refined features, and then adding these features to the first effective dynamometer diagram for fusion, thereby obtaining the second effective dynamometer diagram, etc., which is not limited here.

[0070] In one embodiment, the step in step S5 obtains several key characteristic values ​​of the second effective dynamometer diagram, and preprocesses all the key characteristic values. There may be multiple ways to obtain the corresponding standard characteristic values, which are not limited here. For example, the coordinate value in at least one cycle corresponding to the beginning of the downstroke in the second effective dynamometer diagram may be extracted as a characteristic value, or the second effective dynamometer diagram may be represented by a multi-value image, and the background and shape therein may be distinguished and shown, so as to extract the pixel values ​​therein as key characteristic values, etc., which are not limited here; the way of preprocessing the key characteristic values ​​may be, but not limited to, data cleaning, normalization, and data filtering, or may be selected by technicians in this field according to different scenarios, which are not limited here.

[0071] like Figure 3As shown, in one embodiment of the present invention, when the electrical parameters include a first pumping unit operating voltage parameter, a first pumping unit operating current parameter and a first stroke conversion parameter, and the pumping speed parameter includes a first oil pump efficiency parameter and a first oil pump cleaning parameter, the steps in step S3, converting and calculating all the retained electrical parameters and pumping speed parameters to obtain all corresponding corrected load data and corrected displacement data, may include but are not limited to the following steps:

[0072] Step S31, dividing the electrical parameters and pumping speed parameters corresponding to any one of the monitoring time periods into a group of related data;

[0073] Step S32: for each group of associated data, normalize the first pumping unit operating voltage parameter, the first pumping unit operating current parameter, the first stroke conversion parameter, the first oil pump efficiency parameter, and the first oil pump cleaning parameter to obtain the corresponding second pumping unit operating voltage parameter, the second pumping unit operating current parameter, the second stroke conversion parameter, the second oil pump efficiency parameter, and the second oil pump cleaning parameter, respectively;

[0074] Step S33, substitute the second pumping unit operating voltage parameter, the second pumping unit operating current parameter, the second stroke conversion parameter, the second oil pump efficiency parameter and the second oil pump cleaning parameter into the preset power data correction formula, and calculate the corresponding corrected load data and corrected displacement data.

[0075] In this step, the electrical parameters and pumping speed parameters corresponding to any differentiated monitoring time period are divided into a group of related data to distinguish and identify different electrical parameters and pumping speed parameters in different differentiated monitoring time periods, and then each group of related data is normalized to improve the processing accuracy of each group of related data, and finally the obtained second pumping unit operating voltage parameter, second pumping unit operating current parameter, second stroke conversion parameter, second oil pump efficiency parameter and second oil pump cleaning parameter are substituted into the preset power indicator data correction formula to calculate the required corrected load data and corrected displacement data.

[0076] In one embodiment, the dynamometer data correction formula may be, but is not limited to, as follows:

[0077]

[0078] M is the corrected load data, N is the corrected displacement data, V is the operating voltage parameter of the second oil pumping unit, I is the operating current parameter of the second oil pumping unit, H is the second stroke conversion parameter, k is the preset displacement conversion factor, G1 is the second oil pump efficiency parameter, G2 is the second oil pump cleaning parameter, T1 is the start time of the corresponding differentiated monitoring time period, and T2 is the end time of the corresponding differentiated monitoring time period.

[0079] like Figure 4 As shown, in one embodiment of the present invention, the steps in step S6, according to the difference between the standard characteristic value and the preset reference characteristic value, assigning the corresponding fault impact proportion to the standard characteristic value, may include but is not limited to the following steps:

[0080] Step S61, calculating the difference between the standard characteristic value and the preset reference characteristic value;

[0081] Step S62: Based on a preset weighted influence calculation formula, the difference between the standard characteristic value and the preset reference characteristic value and all standard characteristic values ​​are combined to calculate the fault influence proportion corresponding to the standard characteristic value, wherein the weighted influence calculation formula is as follows:

[0082]

[0083] I i is the i-th standard eigenvalue, P is the fault impact proportion corresponding to the i-th standard eigenvalue, S0 is the preset reference eigenvalue, and n is the total number of standard eigenvalues.

[0084] In this step, the difference between the standard eigenvalue and the preset reference eigenvalue, i.e., S i -S0 represents the degree of fluctuation of the standard characteristic value, while It represents the degree of fluctuation of all standard eigenvalues ​​relative to the whole. By calculating the quotient of the two, the fault impact proportion corresponding to the standard eigenvalue can be obtained. The purpose of taking the absolute value is to adapt to the two different situations that the standard eigenvalue is greater than the preset reference eigenvalue or the standard eigenvalue is less than the preset reference eigenvalue, so as to obtain an accurate and reliable fault impact proportion.

[0085] like Figure 5 As shown, in one embodiment of the present invention, the step in step S6, determining the fault parameter evaluation value corresponding to the second effective dynamometer diagram according to all fault influence proportions and standard characteristic values, includes the following steps:

[0086] Step S63, respectively calculating the product of each standard characteristic value and the corresponding fault impact ratio to obtain the fault impact amount; screening out the largest standard characteristic value and the smallest standard characteristic value from all standard characteristic values;

[0087] Step S64: Substitute each fault influence amount, the largest standard characteristic value and the smallest standard characteristic value into a preset fault parameter evaluation formula to calculate a fault parameter evaluation value corresponding to the second effective dynamometer diagram, wherein the fault parameter evaluation formula is as follows:

[0088]

[0089] Q is the fault parameter evaluation value corresponding to the second effective dynamometer diagram, S max is the largest standard eigenvalue, S min is the smallest standard eigenvalue, P j is the jth fault impact, and m is the total number of fault impacts.

[0090] In this step, the fault impact amount is obtained by calculating the product of each standard eigenvalue and the corresponding fault impact ratio, and the fault impact amount characterizes the actual influence of the standard eigenvalue on the fault assessment. The largest standard eigenvalue and the smallest standard eigenvalue are further screened from all the standard eigenvalues. The largest standard eigenvalue and the smallest standard eigenvalue are used to cooperate with the differentiated influence degree of the standard eigenvalue itself. Finally, each fault impact amount, the largest standard eigenvalue and the smallest standard eigenvalue are substituted into the preset fault parameter evaluation formula to calculate the fault parameter evaluation value corresponding to the second effective dynamometer diagram with high accuracy.

[0091] like Figure 6 As shown, in one embodiment of the present invention, the step in step S6, based on the fault parameter evaluation value, the first preset fault parameter reference value and the second preset fault parameter reference value, determines the fault type of the pumping unit within the preset monitoring time period, including the following steps:

[0092] Step S65: when the fault parameter evaluation value is less than or equal to the first preset fault parameter reference value, determine that the fault type of the oil pump within the preset monitoring time period is an oil pump communication fault; or, when the fault parameter evaluation value is greater than the first preset fault parameter reference value and less than the second preset fault parameter reference value, determine that the fault type of the oil pump within the preset monitoring time period is an oil pump instrument fault; or, when the fault parameter evaluation value is greater than or equal to the second preset fault parameter reference value, determine that the fault type of the oil pump within the preset monitoring time period is an oil pump electrical fault.

[0093] In this step, according to the magnitude relationship between the fault parameter evaluation value and the first preset fault parameter reference value and the second preset fault parameter reference value, the specific fault type of the oil pump within the preset monitoring time period can be determined. This makes the judgment very convenient, intuitive and effective, and reduces the difficulty of determining the fault type of the oil pump in the related prior art. Specifically, the communication failure of the oil pump is reflected in the failure of the communication system itself, such as the failure of the RTU itself, data collection failure, channel failure and data transmission failure, etc., and a fault prompt alarm is issued accordingly and stored locally, and uploaded to the upper computer of the oil pump, wherein For the fault information of communication interruption, the fault information will be uploaded after the communication is normal; the failure of the pumping unit instrument can be but not limited to the failure of the instrument connected to the RTU and diagnosis, such as the instrument data exceeding the limit, data mutation, etc., which will be prompted and alarmed, and uploaded to the upper computer of the pumping unit; the electrical failure of the pumping unit can be but not limited to the electrical failure of the RTU, such as the failure information of overcurrent, overvoltage or phase loss, and the failure information of the main components such as contactors, inverters, switch protection devices, etc., which will be prompted and alarmed, and the acquired fault information will be uploaded to the upper computer of the pumping unit.

[0094] It should be noted that the specific values ​​of the first preset fault parameter reference value and the second preset fault parameter reference value can be set accordingly according to different application scenarios, and are not limited here.

[0095] Figure 7 FIG. 1 is a schematic diagram of a comprehensive fault diagnosis system 1000 for an intermittent pumping unit according to an embodiment of the present invention. Figure 7 As shown, the integrated fault diagnosis system 1000 for the pumping unit includes a memory 1100 and a processor 1200. The number of the memory 1100 and the processor 1200 can be one or more. Figure 7 In the embodiment, a memory 1100 and a processor 1200 are taken as an example; the memory 1100 and the processor 1200 in the device may be connected via a bus or other means. Figure 7 The example of connecting through bus is taken in the following.

[0096] The memory 1100 is a computer-readable storage medium that can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the method for comprehensive fault diagnosis of an intermittent pumping unit provided in any embodiment of the present invention. The processor 1200 implements the above-mentioned method for comprehensive fault diagnosis of an intermittent pumping unit by running the software programs, instructions and modules stored in the memory 1100.

[0097] The memory 1100 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function. In addition, the memory 1100 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 1100 may further include a memory remotely arranged relative to the processor 1200, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0098] An embodiment of the present invention further provides a computer-readable storage medium storing computer-executable instructions for executing the comprehensive fault diagnosis method for an intermittent pumping unit provided in any embodiment of the present invention.

[0099] An embodiment of the present invention also provides a computer program product, including a computer program or computer instructions, wherein the computer program or computer instructions are stored in a computer-readable storage medium, and a processor of a computer device reads the computer program or computer instructions from the computer-readable storage medium, and the processor executes the computer program or computer instructions, so that the computer device executes the comprehensive diagnosis method for intermittent pumping unit faults provided in any embodiment of the present invention.

[0100] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0101] In hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0102] The terms "component", "module", "system", etc. used in this specification are used to represent computer-related entities, hardware, firmware, a combination of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, or a computer. By way of illustration, both applications running on a computing device and a computing device can be components. One or more components may reside in a process or an execution thread, and a component may be located on a computer or distributed between two or more computers. In addition, these components may be executed from various computer-readable media having various data structures stored thereon. Components may communicate, for example, through local or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, or a network, such as the Internet interacting with other systems through signals).

Claims

1. A comprehensive fault diagnosis method for an intermittent pumping unit, characterized in that: The steps include: Step S1, continuously acquiring electrical parameters and pumping speed parameters of the oil pumping unit within a preset monitoring time period, dividing the preset monitoring time period according to a predetermined intermittent operation cycle of the oil pumping unit, so as to divide the preset monitoring time period into at least one differentiated monitoring time period; Step S2, for each of the distinguished monitoring time periods, judging whether the distinguished monitoring time period is an available monitoring time period according to the electrical parameters and the pumping speed parameters corresponding to the distinguished monitoring time period, if it is not an available monitoring time period, deleting all the electrical parameters and the pumping speed parameters corresponding to the distinguished monitoring time period, otherwise retaining all the electrical parameters and the pumping speed parameters corresponding to the distinguished monitoring time period, and executing step S3; Step S3, converting and calculating all the electrical parameters and the pumping speed parameters that have been retained to obtain all corresponding corrected load data and corrected displacement data, and drawing a corresponding first effective dynamometer diagram according to all the corrected load data and the corrected displacement data; Step S4, extracting all feature vectors of the first effective indicator diagram, inputting all the feature vectors into a pre-trained feature extraction network, obtaining effective features corresponding to all the feature vectors output by the feature extraction network, and updating the first effective indicator diagram according to the effective features to obtain a second effective indicator diagram; Step S5, obtaining several key characteristic values ​​of the second effective indicator diagram, and preprocessing all the key characteristic values ​​to obtain corresponding standard characteristic values; Step S6: for each of the standard characteristic values, assigning a corresponding fault impact weight to the standard characteristic value according to the difference between the standard characteristic value and a preset reference characteristic value; Determine a fault parameter evaluation value corresponding to the second effective dynamometer diagram according to all the fault impact proportions and the standard characteristic value, and determine the fault type of the oil pumping unit within the preset monitoring time period based on the fault parameter evaluation value, a first preset fault parameter reference value, and a second preset fault parameter reference value, wherein the second preset fault parameter reference value is greater than the first preset fault parameter reference value; The electrical parameters include a first pumping unit operating voltage parameter, a first pumping unit operating current parameter and a first stroke conversion parameter, and the pumping speed parameters include a first oil pump efficiency parameter and a first oil pump cleaning parameter. The steps in step S3, converting and calculating all the reserved electrical parameters and the pumping speed parameters to obtain all corresponding corrected load data and corrected displacement data, include the following steps: Step S31, dividing the electrical parameters and the pumping speed parameters corresponding to any one of the differentiated monitoring time periods into a group of associated data; Step S32: for each group of the associated data, normalize the first pumping unit operating voltage parameter, the first pumping unit operating current parameter, the first stroke conversion parameter, the first oil pump efficiency parameter and the first oil pump cleaning parameter to obtain the corresponding second pumping unit operating voltage parameter, the second pumping unit operating current parameter, the second stroke conversion parameter, the second oil pump efficiency parameter and the second oil pump cleaning parameter respectively; Step S33, substituting the second pumping unit operating voltage parameter, the second pumping unit operating current parameter, the second stroke conversion parameter, the second oil pump efficiency parameter and the second oil pump cleaning parameter into a preset work indicator data correction formula to calculate corresponding corrected load data and corrected displacement data; The correction formula of the indicator data is as follows: M is the corrected load data, N is the corrected displacement data, V is the operating voltage parameter of the second pumping unit, I is the operating current parameter of the second pumping unit, H is the second stroke conversion parameter, k is a preset displacement conversion factor, G1 is the second oil pump efficiency parameter, G2 is the second oil pump cleaning parameter, T1 is the corresponding start time of the differentiated monitoring time period, and T2 is the corresponding end time of the differentiated monitoring time period; The step S6 comprises the following steps: Step S61, calculating the difference between the standard characteristic value and the preset reference characteristic value; Step S62: Based on a preset weight influence calculation formula, in combination with the difference between the standard characteristic value and the preset reference characteristic value and all the standard characteristic values, the fault influence proportion corresponding to the standard characteristic value is calculated, wherein the weight influence calculation formula is as follows: S i is the i-th standard characteristic value, P is the fault impact proportion corresponding to the i-th standard characteristic value, S0 is the preset reference characteristic value, and n is the total number of the standard characteristic values; Step S63, respectively calculating the product between each of the standard characteristic values ​​and the corresponding fault impact proportion to obtain the fault impact amount; screening out the largest standard characteristic value and the smallest standard characteristic value from all the standard characteristic values; Step S64: Substitute each of the fault impact quantities, the largest standard characteristic value, and the smallest standard characteristic value into a preset fault parameter evaluation formula to calculate a fault parameter evaluation value corresponding to the second effective dynamometer diagram, wherein the fault parameter evaluation formula is as follows: Q is the fault parameter evaluation value corresponding to the second effective dynamometer diagram, S max is the largest standard eigenvalue, S min is the smallest one of the standard eigenvalues, P j is the jth fault impact quantity, and m is the total number of fault impact quantities.

2. The method for comprehensive fault diagnosis of an intermittent pumping unit according to claim 1, characterized in that: The step in step S6, based on the fault parameter evaluation value, the first preset fault parameter reference value and the second preset fault parameter reference value, determines the fault type of the oil pump within the preset monitoring time period, including the following steps: Step S65: when the fault parameter evaluation value is less than or equal to the first preset fault parameter reference value, determining that the fault type of the oil pumping unit within the preset monitoring time period is an oil pumping unit communication fault; or, When the fault parameter evaluation value is greater than the first preset fault parameter reference value and less than the second preset fault parameter reference value, determining that the fault type of the oil pumping unit within the preset monitoring time period is an oil pumping unit instrument fault; or, When the fault parameter evaluation value is greater than or equal to the second preset fault parameter reference value, it is determined that the fault type of the oil pumping unit within the preset monitoring time period is an electrical fault of the oil pumping unit.

3. The method for comprehensive fault diagnosis of an intermittent pumping unit according to claim 1, characterized in that: The intermittent operation cycle includes a partially open operation cycle and a partially closed operation cycle, wherein in the partially open operation cycle, the oil pump operates normally; in the partially closed operation cycle, the oil pump stops operating; The step in step S1, dividing the preset monitoring time period according to the predetermined intermittent operation cycle of the pumping unit to divide the preset monitoring time period into at least one differentiated monitoring time period, includes the following steps: Step S11, when the preset monitoring time period completely overlaps with the partially started operation cycle, the preset monitoring time period is used as the first differentiated monitoring time period; or, When the preset monitoring time period completely overlaps with the partially closed operation cycle, the preset monitoring time period is used as the second differentiated monitoring time period; or, When a part of the preset monitoring time period coincides with the partially open operating cycle and another part of the preset monitoring time period coincides with the partially closed operating cycle, the preset monitoring time period is divided into a third differentiated monitoring time period located within the partially open operating cycle and a fourth differentiated monitoring time period located within the partially closed operating cycle.

4. A comprehensive fault diagnosis system for intermittent pumping units, characterized in that: include: at least one processor; at least one memory for storing at least one program; When at least one of the programs is executed by at least one of the processors, the comprehensive fault diagnosis method for the intermittent pumping unit according to any one of claims 1 to 3 is implemented.

5. A computer-readable storage medium, characterized in that: A program executable by a processor is stored therein, and when the program executable by the processor is executed by the processor, it is used to implement the comprehensive fault diagnosis method of the intermittent pumping unit as described in any one of claims 1 to 3.

Images