An experimental device and system for detecting and testing downhole materials used in oilfield construction sites

Abstract

The present application relates to the technical field of oilfield efficient exploitation, in particular to a kind of well entry material inspection detection experimental device and system for oilfield construction site.The device samples and obtains detection characteristics for liquid and granular well entry materials;based on the detection characteristics, the sampling samples are clustered to form a first sample cluster, and the detection characteristics are weighted according to the number of samples in the cluster, and the standard comparison is carried out after obtaining the comprehensive detection characteristics;based on the number and volume distribution characteristics of particle size, a second sample cluster is formed, impurity interference data is screened out by analyzing the volatility of each particle size distribution characteristic, and the quality proportion in the preset particle size range is weighted and fused by combining the overall detection characteristics similarity of each sample, to obtain the final quality proportion and carry out standard comparison.The present application optimizes the detection data of liquid and granular well entry materials, effectively improves the recognition ability and evaluation accuracy of on-site well entry material detection on transportation and storage interference.

Description

Technical Field

[0001] This invention relates to the field of efficient oilfield exploitation technology, specifically to an experimental device and system for testing and inspecting well-entry materials used in oilfield construction sites. Background Technology

[0002] Drilling, fracturing, and cementing are core components of oilfield construction. Therefore, pre-construction performance testing of materials such as drilling fluids, fracturing fluids, and proppant directly impacts operational safety, oilfield efficiency, and the profitability of oil and gas extraction. In actual oilfield operations, complex and variable geological conditions and harsh working environments, coupled with the susceptibility of materials to environmental changes and impurities during transportation, storage, and use, can lead to significant discrepancies between the performance parameters of these materials and standard parameters, affecting construction safety. Therefore, timely and effective testing of materials before well entry is crucial for quality control. By promptly testing the performance of each material before well entry, adjustments can be made in a timely manner, preventing safety accidents caused by material performance deviations. This is a vital technical guarantee for achieving efficient, safe, and low-cost oilfield extraction.

[0003] However, during on-site construction, traditional testing methods mostly focus on pre-transport testing, neglecting the impact of transportation and storage on the performance of the materials used in the well. This can lead to the materials deviating from standard performance parameters before construction. These materials include both liquid and granular components.

[0004] In actual oilfield operations, both liquid and granular materials used in wellbore fracturing undergo long-distance transportation, multiple loading and unloading processes, and open-air or uncontrolled temperature storage from the factory to the well. During this process, foreign impurities are easily introduced. These impurities often have similar density or appearance to the main components, making them difficult to identify visually or through conventional sieving, but they significantly interfere with the accuracy of sample testing. Existing field testing methods generally assess material quality by directly averaging single or small samples, failing to eliminate interference from transportation or storage. This leads to field test results deviating from the material's true performance, potentially causing misjudgments and severely impacting the safety of fracturing operations and oil and gas recovery efficiency. Summary of the Invention

[0005] To address the technical problem that existing technologies fail to effectively eliminate irrelevant influences during the inspection of materials used in oilfield operations, leading to misjudgments and impacting the safety of fracturing operations and oil and gas recovery efficiency, the present invention aims to provide an experimental device and system for inspecting and testing materials used in oilfield operations. The specific technical solution adopted is as follows: This invention proposes an experimental device for testing and inspecting well-entry materials used in oilfield construction sites. The device includes: The detection data acquisition module is used to sample liquid and granular materials used in wells, obtain samples of each material, and obtain the detection characteristics of each sample. The liquid-type wellbore material detection module is used to cluster liquid-type sampled materials according to detection characteristics for each type of liquid-type wellbore material to obtain a first sample cluster. The detection characteristics are weighted according to the number of samples contained in the first sample cluster to obtain the comprehensive detection characteristics of each type of liquid-type wellbore material and to perform standard comparison. The granular wellbore material detection module is used to cluster granular sampled materials for each type of wellbore material according to detection characteristics to obtain a second sample cluster. The detection characteristics of the granular sampled materials are the distribution characteristics of the quantity and volume of different particle sizes. In each second sample cluster, based on the fluctuation of the distribution characteristics of each particle size, the distribution characteristics of impurity particle sizes are screened out to obtain the optimized distribution characteristics of the final particle size. The mass proportion of each granular sampled material within a preset particle size range is obtained based on the optimized distribution characteristics. The overall detection characteristic similarity of each granular sampled material relative to other granular sampled materials in its second sample cluster is calculated. The overall detection characteristic similarity is used as a weight to perform weighted fusion of the mass proportions to obtain the final mass proportions and perform standard comparison. The test result output module is used to output the results of the standard comparison.

[0006] Furthermore, the liquid-type wellbore entry materials include drilling fluid and fracturing fluid; the particulate wellbore entry materials include proppant.

[0007] Furthermore, the detection characteristics of the drilling fluid are a feature vector composed of density, viscosity, and filtrate loss; the detection characteristics of the fracturing fluid are a feature vector composed of viscosity and surface tension.

[0008] Furthermore, the method for obtaining the first sample cluster includes: Based on the detected features, the first sample cluster is obtained using an agglomerative hierarchical clustering algorithm.

[0009] Furthermore, the method for obtaining the comprehensive detection features includes: For each first sample cluster, the number of samples is used as the numerator and the total number of liquid samples is used as the denominator to obtain the quantity weight; the average detection feature in each first sample cluster is obtained, and the average detection features in all first sample clusters are weighted and summed using the quantity weight to obtain the comprehensive detection feature.

[0010] Furthermore, the method for obtaining the second sample cluster includes: For each particulate sample, the quantity and volume percentage of each particle size in the particulate sample are obtained, and the particles are arranged according to their size to obtain a quantity percentage sequence and a volume percentage sequence. Obtain the first difference distance of the quantity proportion sequence and the second difference value of the volume proportion sequence between two granular samples; construct a first feature set from all the first difference distances of a granular sample and construct a second feature set from all the second difference distances. For any two granular sampled samples, calculate the Jaccard coefficient between the first feature set and the Jaccard coefficient between the second feature set. Perform a negative correlation mapping on the average of the two Jaccard coefficients to obtain the sample difference. Based on the sample difference, use the agglomerative hierarchical clustering algorithm to obtain the second sample cluster.

[0011] Furthermore, the particle size distribution characteristics of the screened impurities include: In a second sample cluster, for the target particle size, the first variance of the statistical quantity proportion and the second variance of the volume proportion are calculated. If the first variance is greater than a preset first threshold and the second variance is greater than a preset second threshold, then the distribution characteristics of the target particle size in all particulate samples within the second sample cluster are removed.

[0012] Furthermore, the method for obtaining the overall detection feature similarity includes: The sample differences are negatively correlated to obtain sample similarity; for any granular sample, the average sample similarity with other granular samples in the second sample cluster is taken as the overall detection feature similarity.

[0013] Furthermore, the method for obtaining the final mass percentage includes: For any granular sample, the overall detection feature similarity is used as the numerator, and the sum of the overall detection feature similarities of all granular samples is used as the denominator to obtain the similarity weight. In a second sample cluster, the quality proportion is weighted by the similarity weight, and the quality proportions of all granular samples are weighted and summed to obtain the weighted overall quality proportion. The average weighted overall quality proportion in all second sample clusters is used as the final quality proportion.

[0014] This invention also proposes an experimental system for testing and inspecting well-entry materials used in oilfield construction sites, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the following method: Sampling was conducted on liquid and granular materials used in wells to obtain samples of each material, and the detection characteristics of each sample were obtained. For each type of liquid material injected into the well, the sampled liquid materials are clustered according to the detection characteristics to obtain a first sample cluster. The detection characteristics are weighted according to the number of samples contained in the first sample cluster to obtain the comprehensive detection characteristics of each type of liquid material injected into the well and then compared with the standard. For each type of particulate material injected into the well, the particulate samples are clustered according to detection characteristics to obtain a second sample cluster. The detection characteristics of the particulate samples are the distribution characteristics of the number and volume of different particle sizes. In each second sample cluster, based on the fluctuation of the distribution characteristics of each particle size, the distribution characteristics of impurity particle sizes are screened out to obtain the optimized distribution characteristics of the final particle size. The mass proportion of each particulate sample within a preset particle size range is obtained based on the optimized distribution characteristics. The overall detection characteristic similarity of each particulate sample relative to other particulate samples in its second sample cluster is calculated. The mass proportion is weighted and fused using the overall detection characteristic similarity as a weight to obtain the final mass proportion and perform a standard comparison. Output the results of the standard comparison.

[0015] The present invention has the following beneficial effects: This invention effectively solves the performance misjudgment problem caused by neglecting the interference of transportation and storage processes in traditional testing methods by constructing a well material inspection and testing system for oilfield construction sites. For liquid well materials, all sampled materials are clustered into a first sample cluster based on detection characteristics. The detection characteristics are then weighted and fused according to the number of samples in each cluster, effectively suppressing the interference of abnormal samples on the overall evaluation and improving the ability of comprehensive detection characteristics to reflect the true performance of the material. For granular well materials, a second sample cluster is obtained by innovatively clustering based on the quantity and volume distribution characteristics of particle size. Then, by analyzing the fluctuation of the distribution characteristics under each particle size, abnormal particle size data affected by impurities are identified and eliminated to obtain an optimized particle size distribution. On this basis, a weight is constructed by combining the overall detection characteristic similarity of each sample with other samples, and the mass proportion within a preset particle size range is weighted and fused, significantly enhancing the robustness to non-ideal factors such as impurity contamination. This invention optimizes the detection data to obtain accurate standard comparison results, thereby improving the accuracy and reliability of on-site well material testing and ensuring the safety of fracturing operations and the efficiency of oil and gas recovery. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a structural block diagram of an experimental device for testing and inspecting well materials used in oilfield construction sites, provided as an embodiment of the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an experimental device and system for testing and inspecting well materials used in oilfield construction sites according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of an experimental device and system for testing and inspecting well materials used in oilfield construction sites, provided by the present invention.

[0021] Please see Figure 1 The diagram shows a structural block diagram of an oilfield construction site well material testing and inspection experimental device according to an embodiment of the present invention. The device includes a data acquisition module 101, a liquid well material testing module 102, a granular well material testing module 103, and a test result output module 104.

[0022] The data acquisition module 101 is the data operation foundation of the device proposed in this embodiment of the invention, and can be implemented through relevant acquisition equipment. The liquid-type wellbore materials in this embodiment include drilling fluid and fracturing fluid; the particulate wellbore materials include proppant. Therefore, it is necessary to pay attention to the density, viscosity, and filtration loss of the drilling fluid; the viscosity and surface tension of the fracturing fluid; and the quantity and volume distribution characteristics of the proppant at various particle sizes. Specifically, a densitometer, rotational viscometer, and fluid loss meter can be used to collect the density, viscosity, and filtration loss data of the drilling fluid; a high-temperature, high-pressure viscometer and a surface tension meter can be used to collect the viscosity and surface tension data of the fracturing fluid; and a laser particle size analyzer can be used to obtain the particle size distribution data of the proppant, wherein the particle size monitoring range of the laser particle size analyzer is 10μm-2000μm.

[0023] In this embodiment of the invention, the material inspection adopts smoke sampling inspection. For each material, this embodiment of the invention uses a fixed sampling number, which is set to 500 samples as a group of sampling samples. In other implementations of this embodiment of the invention, the implementer can increase or decrease the number of sampling samples according to the actual amount of material used in the well.

[0024] It should be noted that due to the complex environment of the oilfield construction site, the data quality is poor due to equipment errors and environmental interference during the relevant testing data collection process. Therefore, for each type of testing data of each type of material injected into the well, a Wiener filter is used to filter and reduce noise, thereby reducing the impact of environmental interference on the collected testing data. The detailed process of Wiener filtering for noise reduction is well known to those skilled in the art and will not be elaborated here. The Z-score standardization algorithm is used to standardize the collected drilling fluid and fracturing fluid testing data to avoid the influence of dimensional differences on the testing and analysis results. The preprocessed data is transmitted to the subsequent testing module for testing and analysis through the testing modules corresponding to different materials injected into the well. The data preprocessing methods are specific to those skilled in the art. In other implementations of this invention, data standardization and normalization can also be achieved through methods such as linear normalization, which will not be elaborated here.

[0025] The liquid-type wellbore material detection module 102 is used to detect liquid-type wellbore materials. Because liquid-type wellbore materials have significant characteristics, their properties can be characterized mainly through physical properties such as viscosity and surface tension. Therefore, liquid-type sampled materials can be directly clustered according to the detection characteristics to obtain the first sample cluster. In the division results of the first sample cluster, if the number of first sample clusters is small, and the number of samples in one or a few first sample clusters is larger, it indicates that the consistency of the drilling fluid performance parameters is higher, and the influence of transportation and storage interference may be less, and the accuracy of its data reflecting the performance of the drilling fluid may be higher. Therefore, in this embodiment of the invention, the number of samples contained in the first sample cluster is used as the confidence weight. The detection features in the first cluster are weighted using this weight, so that the detection features in the first cluster with high consistency retain more information, while the detection features in the first cluster with low consistency (low consistency means that the number of samples is small, which may be due to impurities added during transportation and storage) retain less information. Then, the information of all samples is weighted and integrated to obtain a comprehensive detection feature that can effectively and accurately characterize the detection characteristics of the actual sample. The quality of the batch of materials can be determined by comparing it with the standard feature.

[0026] Preferably, in one embodiment of the present invention, the liquid-type wellbore materials targeted include drilling fluid and fracturing fluid. Drilling fluid is more concerned with density, viscosity, and fluid loss; however, because fracturing fluid has wetting properties and oil displacement effects on the formation, changes in surface tension and viscosity have a significant impact on the compatibility between the fracturing fluid and the formation, leading to poor oil and gas flow. Therefore, for fracturing fluid, viscosity and surface tension are more important. Thus, the detection characteristics of drilling fluid are a feature vector composed of density, viscosity, and fluid loss; the detection characteristics of fracturing fluid are a feature vector composed of viscosity and surface tension.

[0027] Preferably, in this embodiment of the invention, the first sample cluster is obtained based on the detected features using a hierarchical clustering algorithm. It should be noted that, taking drilling fluid samples as an example, their detected features are three-dimensional feature vectors, and the Euclidean distance between feature vectors can be used as the clustering distance for clustering during the clustering process; similarly, fracturing fluid can also be clustered using the Euclidean distance between feature vectors. Specific clustering methods are well-known to those skilled in the art and will not be elaborated upon here.

[0028] Preferably, in this embodiment of the invention, the method for obtaining comprehensive detection features includes: For each first sample cluster, the sample count is used as the numerator, and the total number of liquid samples is used as the denominator to obtain the quantity weights. The average detection features in each first sample cluster are then obtained. The average detection features from all first sample clusters are weighted and summed using the aforementioned quantity weights to obtain the comprehensive detection features. It should be noted that the detection features may have multiple dimensions. For example, drilling fluid contains information on three features. In the weighting process, a normal sampling vector weighted fusion method can be used, specifically a basic linear algebra processing technique, which will not be elaborated upon here.

[0029] It should be noted that after obtaining the comprehensive detection characteristics, conventional comparison methods can be used to compare the detection characteristics of each dimension of the comprehensive detection characteristics with the standard parameters. If they are within the range of the standard parameters, it indicates that the batch of liquid wellbore material is qualified; otherwise, it is judged as unqualified. The range of the standard parameters is prior knowledge in this field and will not be elaborated or limited in detail.

[0030] The granular wellbore material detection module 103 further detects the granular wellbore material. The risks associated with granular wellbore material during construction include: excessively large particles may block fracture channels, hindering the extension of fracturing fluid and reducing fracture conductivity; secondly, excessively small particles may be affected by formation fluids, leading to particle loss and reducing fracture support. Therefore, more attention should be paid to the distribution characteristics of different particle sizes in granular wellbore material. Therefore, in this embodiment of the invention, the detection of granular sampled material specifically focuses on the distribution characteristics of the quantity and volume of different particle sizes. It should be noted that in one specific implementation of this embodiment, the granular wellbore material is selected using a proppant; therefore, the following description of this embodiment can use a proppant as an example.

[0031] In the particulate wellbore material detection module 103, similar to the liquid wellbore material detection module 102, the particulate samples are first clustered according to detection characteristics to obtain second sample clusters. Since the composition of particulate samples is complex, different second sample clusters may be distinguished by different impurity distributions. Therefore, targeted impurity analysis is required for each second sample cluster. Thus, unlike the liquid wellbore material analysis process, which uses a large cluster-dominated approach to optimize detection characteristics, localized and detailed feature analysis is needed for each second sample cluster.

[0032] To achieve accurate impurity impact analysis, this embodiment of the invention analyzes and filters out impurities based on their corresponding particle sizes within each second sample cluster. Therefore, in each second sample cluster, for each particle size, this embodiment analyzes the distribution characteristic fluctuations of the samples contained in the second sample cluster. If the fluctuations are significantly large, indicating a lack of consistency within the cluster, it suggests that the statistical deviation of the particle size is large due to the influence of impurity distribution and volume differences. Therefore, it is unsuitable for subsequent quality testing and can be directly filtered out. In other words, based on the distribution characteristic fluctuations of each particle size, the impurity particle size distribution characteristics are filtered out to obtain the optimized distribution characteristics of the final particle size.

[0033] Based on the optimized distribution characteristics obtained from the final retention, the mass proportion within the preset particle size range can be statistically analyzed. This mass proportion is a direct representation of the quality of each particulate sample. Similarly, as with the detection process of liquid-type wellbore materials, it is necessary to further analyze the confidence level of each sample. Therefore, the particulate wellbore material detection module 103 further calculates the overall detection feature similarity of each particulate sample relative to other particulate samples in its second sample cluster. The greater the overall detection feature similarity, the stronger the consistency of the sample, and the more likely it is to be a normal material sample; thus, the higher the confidence level should be. Therefore, the overall detection feature similarity can be used as a weight to perform weighted fusion of the mass proportions to obtain the final mass proportion and perform standard comparison.

[0034] It should be noted that, taking proppant as an example, the preset particle size range is 40 to 140 mesh. For other granular materials used in wells, specific settings can be made based on the specific type, and no limitations are imposed here. Standard comparisons also need to be made with the standard. It should be noted that, since the detection characteristics collected during the testing of particulate samples are the number of particle sizes and the distribution characteristics of volume, after obtaining the final optimized distribution characteristics, the mass distribution of each particulate sample can be directly obtained through the volume distribution information and the prior density value of the proppant. This is a calculation method well known to those skilled in the art, and will not be elaborated or limited here.

[0035] Preferably, in this embodiment of the invention, the method for obtaining the second sample cluster includes: For each particulate sample, since each particle size yields two data characteristics: quantity distribution and volume distribution, this embodiment uses proportion as the quantification method. The quantity and volume proportions of each particle size in the particulate sample are obtained, and then arranged according to particle size to obtain a quantity proportion sequence and a volume proportion sequence. That is, each particulate sample yields two sequences: a quantity proportion sequence and a volume proportion sequence. In this embodiment, the sequences are arranged in ascending order of particle size.

[0036] The method obtains a first difference distance for the quantity percentage sequence and a second difference value for the volume percentage sequence between two granular sampled samples. All first difference distances of a granular sample are used to form a first feature set, and all second difference distances are used to form a second feature set. In this embodiment, the first and second difference distances are represented using Manhattan distance. This set-based approach allows for the statistical display of all difference results obtained for each granular sample, characterizing the detailed difference distribution between a sample and the overall sample.

[0037] For any two granular sampled samples, the Jaccard coefficients between the first feature set and the second feature set are calculated. The average of the two Jaccard coefficients is then negatively correlated to obtain the sample difference. Based on this sample difference, agglomerative hierarchical clustering algorithm is used to obtain the second sample cluster. A larger Jaccard coefficient indicates a higher degree of overlap between the two sets. Therefore, in this embodiment of the invention, the Jaccard coefficients of the two dimensions are averaged and fused, and then the sample difference is obtained through negative correlation mapping, which is then used as the clustering distance for clustering.

[0038] It should be noted that the Jaccard coefficient ranges from 0 to 1, so the average value of the Jaccard coefficient can be directly subtracted from the positive integer 1 as the result of the negative correlation mapping.

[0039] Preferably, in this embodiment of the invention, the particle size distribution characteristics of the impurities screened out include: In a second sample cluster, for the target particle size, the distribution characteristics of the target particle size in all particulate samples are discarded if the first variance of the statistical quantity proportion is greater than a preset first threshold and the second variance of the volume proportion is greater than a preset second threshold. Specifically, the first threshold is set as the average of the first variances of the target particle size in all particulate samples; the second threshold is set as the average of the second variances of the target particle size in all particulate samples. In other words, the average value is used as a threshold to characterize an overall level, thereby identifying impurity particle sizes that exceed the fluctuations of the overall level.

[0040] Preferably, in this embodiment of the invention, the method for obtaining the overall detection feature similarity includes: The sample differences are negatively correlated to obtain sample similarity. For any granular sample, the average sample similarity with other granular samples in the same second sample cluster is taken as the overall detection feature similarity. It should be noted that, since the sample difference in this embodiment is the result of a negative correlation mapping of the average values ​​of two Jaccard coefficients, this average value can be directly used as the sample similarity, and its value range is between 0 and 1.

[0041] Preferably, in this embodiment of the invention, the method for obtaining the final mass percentage includes: For any granular sample, the overall detection feature similarity is used as the numerator, and the sum of the overall detection feature similarities of all granular samples is used as the denominator to obtain the similarity weight. Within a second sample cluster, the quality proportion is weighted using the similarity weight, and the weighted sum of the quality proportions of all granular samples is obtained to obtain the weighted overall quality proportion. That is, the weighted overall quality proportion represents the sample statistical result of the second sample cluster after eliminating the influence of impurities. Therefore, the average weighted overall quality proportion of all second sample clusters is used as the final quality proportion.

[0042] It should be noted that the Euclidean distance, Manhattan distance, Jaccard coefficient, hierarchical clustering, and other algorithms designed in the embodiments of the present invention are all technical means well known to those skilled in the art, and will not be described in detail.

[0043] The test result output module 104 is used to output the final standard comparison result. In this embodiment of the invention, based on the test results of each material received, the test indicators, measured data, standard thresholds, and qualification judgment results of each material can be marked, and the data comparison can be presented in a visual form such as tables, line graphs, or bar charts. Specifically, in this application, bar charts can be used to statistically analyze the proportion of proppant in different particle size ranges and mark the comparison results with the standard requirements, so that construction personnel can quickly and intuitively grasp the material performance status.

[0044] In summary, the data acquisition module of this invention is used to sample liquid and granular materials injected into wells and obtain detection features. The liquid detection module clusters the sampled samples based on the detection features to form a first sample cluster, and weights the detection features according to the number of samples in each cluster to obtain comprehensive detection features for standard comparison. The granular detection module clusters the samples based on the quantity and volume distribution features of particle size to form a second sample cluster. By analyzing the fluctuations in the distribution features of each particle size, it filters out impurities and obtains an optimized distribution. It then combines the overall similarity of the detection features of each sample with a weighted fusion of the mass proportion within a preset particle size range to obtain the final mass proportion for standard comparison. The detection result output module outputs the comparison results. This invention, through targeted optimization of the detection data for liquid and granular materials injected into wells, effectively improves the ability to identify and assess transportation and storage interference in on-site well material detection.

[0045] Based on the same inventive concept, this invention also proposes an experimental system for testing and inspecting well-entry materials used in oilfield construction sites, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the following method: Sampling was conducted on liquid and granular materials used in wells to obtain samples of each material, and the detection characteristics of each sample were obtained. For each type of liquid material injected into the well, the sampled liquid materials are clustered according to the detection characteristics to obtain a first sample cluster. The detection characteristics are weighted according to the number of samples contained in the first sample cluster to obtain the comprehensive detection characteristics of each type of liquid material injected into the well and then compared with the standard. For each type of particulate material injected into the well, the particulate samples are clustered according to detection characteristics to obtain a second sample cluster. The detection characteristics of the particulate samples are the distribution characteristics of the number and volume of different particle sizes. In each second sample cluster, based on the fluctuation of the distribution characteristics of each particle size, the distribution characteristics of impurity particle sizes are screened out to obtain the optimized distribution characteristics of the final particle size. The mass proportion of each particulate sample within a preset particle size range is obtained based on the optimized distribution characteristics. The overall detection characteristic similarity of each particulate sample relative to other particulate samples in its second sample cluster is calculated. The mass proportion is weighted and fused using the overall detection characteristic similarity as a weight to obtain the final mass proportion and perform a standard comparison. Output the results of the standard comparison.

[0046] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0047] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A testing and inspection apparatus for well-entry materials used in oilfield construction sites, characterized in that, The device includes: The detection data acquisition module is used to sample liquid and granular materials injected into the well, obtain samples of each material, and obtain the detection characteristics of each sample. The liquid-type wellbore material detection module is used to cluster liquid-type sampled materials according to detection characteristics for each type of liquid-type wellbore material to obtain a first sample cluster. The detection characteristics are weighted according to the number of samples contained in the first sample cluster to obtain the comprehensive detection characteristics of each type of liquid-type wellbore material and to perform standard comparison. The granular wellbore material detection module is used to cluster granular sampled materials for each type of wellbore material according to detection characteristics to obtain a second sample cluster. The detection characteristics of the granular sampled materials are the distribution characteristics of the quantity and volume of different particle sizes. In each second sample cluster, based on the fluctuation of the distribution characteristics of each particle size, the distribution characteristics of impurity particle sizes are screened out to obtain the optimized distribution characteristics of the final particle size. The mass proportion of each granular sampled material within a preset particle size range is obtained based on the optimized distribution characteristics. The overall detection characteristic similarity of each granular sampled material relative to other granular sampled materials in its second sample cluster is calculated. The overall detection characteristic similarity is used as a weight to perform weighted fusion of the mass proportions to obtain the final mass proportions and perform standard comparison. The test result output module is used to output the results of the standard comparison; The method for obtaining the second sample cluster includes: For each particulate sample, the quantity and volume percentage of each particle size in the particulate sample are obtained, and the particles are arranged according to their size to obtain a quantity percentage sequence and a volume percentage sequence. Obtain the first difference distance of the quantity proportion sequence and the second difference value of the volume proportion sequence between two granular samples; construct a first feature set from all the first difference distances of a granular sample and construct a second feature set from all the second difference distances. For any two granular sampled samples, calculate the Jaccard coefficient between the first feature set and the Jaccard coefficient between the second feature set, perform a negative correlation mapping on the average of the two Jaccard coefficients to obtain the sample difference, and use the agglomerative hierarchical clustering algorithm to obtain the second sample cluster based on the sample difference. The particle size distribution characteristics of the screened impurities include: In a second sample cluster, for the target particle size, the first variance of the statistical quantity proportion and the second variance of the volume proportion are calculated. If the first variance is greater than a preset first threshold and the second variance is greater than a preset second threshold, then the distribution characteristics of the target particle size in all particulate samples within the second sample cluster are removed.

2. The testing and inspection apparatus for well-entry materials used in oilfield construction sites according to claim 1, characterized in that, The liquid-type wellbore entry materials include drilling fluid and fracturing fluid; the particulate wellbore entry materials include proppant.

3. The testing and inspection apparatus for well-entry materials used in oilfield construction sites according to claim 2, characterized in that, The detection characteristics of drilling fluid are a feature vector composed of density, viscosity, and filtrate loss; the detection characteristics of fracturing fluid are a feature vector composed of viscosity and surface tension.

4. The testing and inspection apparatus for well-entry materials used in oilfield construction sites according to claim 1, characterized in that, The method for obtaining the first sample cluster includes: Based on the detected features, the first sample cluster is obtained using an agglomerative hierarchical clustering algorithm.

5. The testing and inspection apparatus for well-entry materials used in oilfield construction sites according to claim 1, characterized in that, The method for obtaining the comprehensive detection features includes: For each first sample cluster, the number of samples is used as the numerator and the total number of liquid samples is used as the denominator to obtain the quantity weight; the average detection feature in each first sample cluster is obtained, and the average detection features in all first sample clusters are weighted and summed using the quantity weight to obtain the comprehensive detection feature.

6. The testing and inspection apparatus for well-entry materials used in oilfield construction sites according to claim 1, characterized in that, The method for obtaining the overall detection feature similarity includes: The sample differences are negatively correlated to obtain sample similarity; for any granular sample, the average sample similarity with other granular samples in the second sample cluster is taken as the overall detection feature similarity.

7. The testing and inspection apparatus for well-entry materials used in oilfield construction sites according to claim 1, characterized in that, The method for obtaining the final mass percentage includes: For any granular sample, the overall detection feature similarity is used as the numerator, and the sum of the overall detection feature similarities of all granular samples is used as the denominator to obtain the similarity weight. In a second sample cluster, the quality proportion is weighted by the similarity weight, and the quality proportions of all granular samples are weighted and summed to obtain the weighted overall quality proportion. The average weighted overall quality proportion in all second sample clusters is used as the final quality proportion.

8. A testing and inspection system for well-entry materials used in oilfield construction, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the following methods: Sampling was conducted on liquid and granular materials used in wells to obtain samples of each material, and the detection characteristics of each sample were obtained. For each type of liquid material injected into the well, the sampled liquid materials are clustered according to the detection characteristics to obtain a first sample cluster. The detection characteristics are weighted according to the number of samples contained in the first sample cluster to obtain the comprehensive detection characteristics of each type of liquid material injected into the well and then compared with the standard. For each type of particulate material injected into the well, the particulate samples are clustered according to the detection characteristics to obtain a second sample cluster; the detection characteristics of the particulate samples are the distribution characteristics of the number and volume of different particle sizes. In each second sample cluster, based on the fluctuation of the distribution characteristics of each particle size, the distribution characteristics of impurity particle sizes are screened out to obtain the optimized distribution characteristics of the final particle size; the mass proportion of each particulate sample within a preset particle size range is obtained based on the optimized distribution characteristics; the overall detection feature similarity of each particulate sample relative to other particulate samples in its second sample cluster is calculated, and the mass proportion is weighted and fused using the overall detection feature similarity as a weight to obtain the final mass proportion and perform a standard comparison; Output the results of the standard comparison; The method for obtaining the second sample cluster includes: For each particulate sample, the quantity and volume percentage of each particle size in the particulate sample are obtained, and the particles are arranged according to their size to obtain a quantity percentage sequence and a volume percentage sequence. Obtain the first difference distance of the quantity proportion sequence and the second difference value of the volume proportion sequence between two granular samples; construct a first feature set from all the first difference distances of a granular sample and construct a second feature set from all the second difference distances. For any two granular sampled samples, calculate the Jaccard coefficient between the first feature set and the Jaccard coefficient between the second feature set, perform a negative correlation mapping on the average of the two Jaccard coefficients to obtain the sample difference, and use the agglomerative hierarchical clustering algorithm to obtain the second sample cluster based on the sample difference. The particle size distribution characteristics of the screened impurities include: In a second sample cluster, for the target particle size, the first variance of the statistical quantity proportion and the second variance of the volume proportion are calculated. If the first variance is greater than a preset first threshold and the second variance is greater than a preset second threshold, then the distribution characteristics of the target particle size in all particulate samples within the second sample cluster are removed.

Images