Search

US-12618999-B2 - Photoelectric fusion in-situ detection device and method for thermal volatile gases and thermal evolution stages of underground oil shale

US12618999B2US 12618999 B2US12618999 B2US 12618999B2US-12618999-B2

Abstract

The present disclosure relates to a dust-removed odor sampling device for oil shale and a photoelectric fused method for underground in-situ exploration and thermal evolution stage monitoring of oil shale, belonging to the technical field of oil shale in-situ exploration and exploitation. The present disclosure includes: establishing an dust-removed odor sampling device, which includes an imitation sandfish lizard dustproof chamber, an imitation leaf filter membrane, a single-chip microcomputer controller, a dust collection box, a solenoid valve and an air pump; using a redesigned F β score and a feature selection algorithm to optimize a gas sensor array and reduce the apparatus volume and energy consumption; and combining the optimized gas sensor array with the optical gas detector to realize the underground in-situ exploration and thermal evolution stage monitoring of oil shale by the photoelectric fusion method.

Inventors

  • ZHIYONG CHANG
  • CHENG KONG
  • Chengxin SONG
  • Xiangyu Luan
  • XIAOHUI WENG
  • Zongwei YAO
  • Tao Zhang
  • Ruochen AN
  • SUNHUA DENG

Assignees

  • JILIN UNIVERSITY

Dates

Publication Date
20260505
Application Date
20231127

Claims (4)

  1. 1 . A dust-removed odor sampling device for oil shale, comprising a dustproof chamber, a filter membrane, a single-chip microcomputer controller, a dust collection box, a solenoid valve I, an air pump, and a solenoid valve II, where the solenoid valve I, the dustproof chamber, the filter membrane, the single-chip microcomputer controller, the air pump, and the solenoid valve II are arranged sequentially from front to rear; the dustproof chamber is a three-way pipe, on which an air inlet channel, a diffusion channel, a dust settlement channel, a sensing channel and a dust collection channel are provided, an upper part of the sensing channel is provided with a jack I, a jack II, a jack III and a jack IV of a gas-sensitive sensor array; the air inlet channel is a circular cross-section straight pipe with an inner diameter d 1 =10-12 mm; the diffusion channel is an inclined pipe formed by lofting a front circular cross-section and a rear circular cross-section, where centers of circles of the front and rear cross-sections are at a same horizontal line and the circles are tangent to each other, an inner diameter of the front cross-section is d 1 =10-12 mm, and an inner diameter of the rear cross-section is d 2 =28-30 mm; the dust settlement channel is formed by lofting a front circular cross-section and a rear semi-circular cross-section, centers of circles of the front and rear cross-sections are concentric, and inner diameters of the front and rear cross-sections are d 2 =28-30 mm; the sensing channel is a circular cross-section bend pipe with an inner diameter d 4 =10-12 mm and an outer diameter d 5 =12-14 mm; distances between centers of circles of the jack I, jack II, jack III and jack IV are L 1 =15-18 mm, inner diameters of the jacks are d 3 =8-9 mm, and heights of all jacks in relative to height of center of cross-section of the air inlet channel are L 2 =60-62 mm; the dust collection channel is formed by lofting an upper elliptical cross-section and a lower circular cross-section, and circles of the upper and lower cross-sections are tangent to each other in their rear directions; the single-chip microcomputer controller comprises a control unit, a signal acquisition unit, a touch display screen, and a power supply, where the control unit comprises the solenoid valve I and the solenoid valve II; the signal acquisition unit comprises the gas-sensitive sensor array and an analog-to-digital acquisition module; the control unit, the signal acquisition unit, and the touch display screen are connected to the power supply through wires; the dust collection box comprises a dust passage pipe, a dust collection chamber, and a filter screen, where the dust passage pipe is fixedly connected to and communicated with a center of an upper end of the dust collection chamber, and the filter screen is fixedly connected to a lower end of the dust collection chamber; an upper end of the dust passage pipe is threaded to a lower end of the dust collection channel, an outer diameter of the dust collection chamber, being a cylinder, is d 6 =40-45 mm, and a diameter of mesh of the filter screen is d 7 =0.3-0.5 mm; the solenoid valve I is communicated with the air inlet channel of the dustproof chamber; the gas-sensitive sensor array is fixedly connected to the jack I, jack II, jack III and jack IV respectively of the sensing channel of the dustproof chamber; the air pump is communicated with a rear part of the sensing channel of the dustproof chamber; the dust collection box is communicated with a lower part of the dust collection channel in the dustproof chamber through the solenoid valve II; the single-chip microcomputer controller is connected with the solenoid valve I and the solenoid valve II through wires, the single-chip microcomputer controller is connected with the gas-sensitive sensor array and powered through the power supply, and a voltage signal transformation generated by the gas-sensitive sensor array is collected, recorded and stored through the analog-to-digital acquisition module; wherein the filter membrane is located at an entrance of the sensing channel and is used to intercept particles in the dust settlement channel from entering the sensing channel, thereby ensuring that the gas-sensitive sensor array is not affected by fine particles and achieving dust-proof sampling.
  2. 2 . The dust-removed odor sampling device for oil shale according to claim 1 , wherein the dustproof chamber is a three-way pipe, on which an air inlet channel, a diffusion channel, a dust settlement channel, a sensing channel and a dust collection channel are provided, an upper part of the sensing channel is provided with a jack I, a jack II, a jack III and a jack IV of a gas-sensitive sensor array; the air inlet channel is a circular cross-section straight pipe with an inner diameter d 1 =10-12 mm; the diffusion channel is an inclined pipe formed by lofting a front circular cross-section and a rear circular cross-section, where centers of circles of the front and rear cross-sections are at a same horizontal line and the circles are tangent to each other, an inner diameter of the front cross-section is d 1 =10-12 mm, and an inner diameter of the rear cross-section is d 2 =28-30 mm; the dust settlement channel is formed by lofting a front circular cross-section and a rear semi-circular cross-section, centers of circles of the front and rear cross-sections are concentric, and inner diameters of the front and rear cross-sections are d 2 =28-30 mm; the sensing channel is a circular cross-section bend pipe with an inner diameter d 4 =10-12 mm and an outer diameter d 5 =12-14 mm; distances between centers of circles of the jack I, jack II, jack III and jack IV are L 1 =15-18 mm, inner diameters of the jacks are d 3 =8-9 mm, and heights of all jacks in relative to height of center of cross-section of the air inlet channel are L 2 =60-62 mm; the dust collection channel is formed by lofting an upper elliptical cross-section and a lower circular cross-section, and circles of the upper and lower cross-sections are tangent to each other in their rear directions; the dust collection box comprises a dust passage pipe, a dust collection chamber, and a filter screen, where the dust passage pipe is fixedly connected to and communicated with a center of an upper end of the dust collection chamber, and the filter screen is fixedly connected to a lower end of the dust collection chamber; an upper end of the dust passage pipe is threaded to a lower end of the dust collection channel, an outer diameter of the dust collection chamber, being a cylinder is d 6 =40-45 mm, and a diameter of mesh of the filter screen is d 7 =0.3-0.5 mm.
  3. 3 . A production method of filter membrane, comprising the following steps: S1: mixing a PDMS polymer and a curing agent at a ratio of 10:1, and adding them to solvent FC-43 at 20 wt % for mechanical stirring for 10 min, S2: applying solution prepared in step S1 to a pure silicon mold with a pit array, and then placing the mold into a microwave oven for 3 min of radiation, wherein the pure silicon mold has a height of L 5 =25-30 mm, an inner slot thereof has a diameter of d 8 =10-12 mm and a height of L 7 =15-20 mm, and distances between centers of cells of a cylindrical cell array arranged in the inner slot are L 4 =15-20 μm, depths of the cells are L 6 =6-8 μm and diameters of the cells are d 9 =12-16 μm.
  4. 4 . A photoelectric fusion method for underground in-situ exploration and thermal evolution stage monitoring of oil shale, wherein the method is based on the dust-removed odor sampling device for oil shale according to claim 1 , and specifically comprises the following steps: 4.1, according to composition of a volatile gas of oil shale and a change process of a concentration of a gas component during thermal evolution stage, taking a gas whose concentration changes significantly with decomposition of organic matter of oil shale as a target gas of an optical gas detector, and recording the target gas as G; 4.2, constructing the optical gas detector, which comprises: a laser, a gas absorption pool, a detector, a laser driver, a signal generator, a lock-in amplifier, and a data acquisition card; 4.3, constructing an initial gas sensor array: according to type of a gas compound volatilized from oil shale in different states, selecting other gas sensor except the target gas G in step 4.1, to construct an initial gas senor array that is capable of monitoring all heat volatile gases, where the different states of oil shale comprise organic matter abundance, kerogen type and thermal evolution stage; 4.4, obtaining, by using the initial gas sensor array constructed in step 4.3, oil shale odor data in different maturity states, comprising the following steps: 4.4.1, constructing a signal acquisition system by using the initial gas sensor array, together with an air pump, a filter noise reduction circuit and a data acquisition instrument, and heating oil shale sample according to a heat-injection in-situ mining method of oil shale; converting, by the signal acquisition system, an odor into numerical data, where n odor samples are collected and recorded as set S={s 1 , s 2 , s 3 , . . . , s n }; 4.4.2, obtaining the thermal evolution stage of oil shale through a laboratory measurement, recording it as set L={l 1 , l 2 , . . . , l k }, and corresponding it to the odor samples to form a training set T = { s 1 l , s 2 l , s 3 l , … , s n l } , where l∈L; 4.5, extracting and fusing multi-phase features of the training set of oil shale, comprising the following steps: 4.5.1, extracting, by a plurality of steady-state feature extraction methods, a steady-state feature of each sample to form a steady-state feature vector, which is recorded as F s ; 4.5.2, extracting, by a plurality of transient feature extraction methods, a transient feature of each sample to form a transient feature vector, which is recorded as F t ; 4.5.3, fusing the steady-state feature vector and the transient feature vector extracted from each sample to form a final feature vector F=(F s , F t ); 4.6, sorting each feature in the final feature vector based on its importance, comprising the following steps: 4.6.1, calculating the importance of each feature by using a filtering method in feature selection and sorting each feature; 4.6.2, calculating the importance of each feature by using an embedding method in feature selection and sorting each feature; 4.6.3, sorting each feature by obtaining an average value and combining sequence numbers obtained by multiple sorting methods to obtain a final sorting of features; 4.7, utilizing F β ′ score to achieve selection of an optimal feature subset, comprising the following steps: 4.7.1, recognizing, by using numerous classifiers in machine learning, a state of oil shale on the training set, comparing a recognition results of each classifier with a true state of the oil shale, and determining the classifier with a smallest difference between the recognition result and the true state of the oil shale as a classifier with best recognition effect and use it as a feature subset selection classifier, where a 10-fold cross-validation is used for evaluation in an evaluation process of the classifiers; 4.7.2, generating different number of feature subsets based on sorting of the features, evaluating each feature subset by an improved F β score, F β ′ = ( 1 + β 2 ) × d - m d - 1 × accuracy ( β 2 × d - m d - 1 ) + accuracy , and determining a feature subset with a highest F β ′ score as the optimal feature subset, where d is feature number in the final feature vector F, m is feature number in the feature subset, accuracy is a recognition rate of an established model for state prediction of oil shale, and β is integer in [1,10], which is selected through data testing; 4.8, finding gas sensors corresponding to features in the optimal feature subset, to jointly form an optimized sensor array; 4.9, establishing a photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale: the optimized sensor array obtained after optimizing the sensor array, together with the optical gas detector, a cooling device, a temperature sensor, a pressure sensor, an oil shale underground in-situ conversion heating device, and a formation pressure control device constitute the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale; 4.10, using the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale to realize underground in-situ exploration of oil shale, comprising the following steps: 4.10.1, for oil shales with different kerogen types, measuring their kerogen types in laboratory, to constitute a label set, which is recorded as set G 1 = { g 1 1 , g 2 1 , g 3 1 , … , g k 1 } , and meanwhile, collecting, by using the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale, volatile gas information of oil shale, which is recorded as set H 1 = { h 1 1 , h 2 1 , h 3 1 , … , h k 1 } , and combining G 1 and H 1 to constitute a training set J 1 = { j h 1 1 g , j h 2 1 g , j h 3 1 g , … , j h k 1 g } , where g∈G 1 ; 4.10.2, extracting features of the set J 1 using a feature extraction method to constitute a feature set, which is recorded as F j 1 ; 4.10.3, constructing, by using a machine learning algorithm and taking the feature F j 1 as input, a classification model, which is recorded as A 1 , for determining type of organic matter kerogen of oil shale; 4.10.4, for oil shales with different abundances, measuring their abundances in laboratory, to constitute a label set, which is recorded as set G 2 = { g 1 2 , g 2 2 , g 3 2 , … , g k 2 } , and meanwhile, collecting, by using the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale, volatile gas information of oil shale, which is recorded as set H 2 = { h 1 2 , h 2 2 , h 3 2 , … , h k 2 } , and combining G 2 and H 2 to constitute a training set, J 2 = { j h 1 2 g , j h 2 2 g , j h 3 2 g , … , j h k 2 g } , where g∈G 2 ; 4.10.5, extracting features of the set J 2 using a feature extraction method to constitute a feature set, which is recorded as F j 2 ; 4.10.6, constructing, by using a regression analysis method and taking the feature F j 2 as input, a prediction model, which is recorded as A 2 , for determining abundance of organic matter of oil shale; 4.10.7, for oil shales with different maturities, measuring their maturity stages and maturity values in laboratory, where the maturity stages constitute a label set, which is recorded as set G 3 = { g 1 3 , g 2 3 , g 3 3 , … , g k 3 } , the maturity values constitute a label set, which is recorded as set G 4 = { g 1 4 , g 2 4 , g 3 4 , … , g k 4 } , and meanwhile, collecting, by using the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale, volatile gas information of oil shale, which is recorded as set H 3 = { h 1 3 , h 2 3 , h 3 3 , … , h k 3 } , combining G 3 and H 3 to constitute a training set, J 3 = { j h 1 3 g , j h 2 3 g , j h 3 3 g , … , j h k 3 g } , where g∈G 3 ; and combining G 4 and H 3 to constitute a training set J 4 = { j h 1 3 g , j h 2 3 g , j h 3 3 g , … , j h k 3 g } , where g∈G 4 ; 4.10.8, extracting features of the set J 3 using a feature extraction method to constitute a feature set, which is recorded as F j 3 ; 4.10.9, constructing, by using a machine learning algorithm and taking the feature F j 3 as input, a classification model, which is recorded as A 3 , for determining maturity stage of organic matter of oil shale; 4.10.10, extracting features of the set J 4 using a feature extraction method to constitute a feature set, which is recorded as F j 4 ; 4.10.11, constructing, by using a regression analysis method and taking the feature F j 4 as input, a prediction model, which is recorded as A 4 , for determining maturity of organic matter of oil shale; 4.10.12, heating an oil shale undergoing underground in-situ exploration, and converting, by using the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale, odor into numerical data, which is recorded as sample x; 4.10.13, extracting a feature of a sample x using a feature extraction method and recording it as F x ; 4.10.14, determining, by using the classification model A 1 of the type of organic matter kerogen of oil shale as constructed in step 4.10.3 and taking the feature F x as input, the type of organic matter kerogen of oil shale belongs to which one of kerogen types I, II and III; 4.10.15, determining, by using the prediction model A 2 of the abundance of organic matter of oil shale as constructed in step 4.10.6 and taking the feature F x as input, the abundance of organic matter of oil shale; 4.10.16, determining, by using the classification model A 3 of the maturity stage of organic matter of oil shale as constructed in step 4.10.9 and taking the feature F x as input, the maturity of organic matter of oil shale is in which stage of three stages comprising dehydration, organic matter decomposition and semi-coke mineral decomposition; 4.10.17, determining, by using the prediction model A 4 of the maturity of organic matter of oil shale as constructed in step 4.10.11 and taking the feature F x as input, the maturity of organic matter of oil shale; 4.10.18, according to the kerogen type, abundance, maturity stage and maturity value of organic matter of oil shale obtained in steps 4.10.14 to 4.10.17, evaluating mineral value of oil shale and guiding on-site conversion technology selection and engineering process design to assist efficient in-situ exploitation of underground oil and gas resources; 4.11, using the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale to realize underground in-situ thermal evolution stage monitoring of oil shale, comprising the following steps: 4.11.1, monitoring a pyrolysis process of oil shale undergoing underground in-situ conversion, and converting, by using the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale, odor information into numerical data, which is recorded as sample w; 4.11.2, extracting a feature of a sample w using a feature extraction method and recording it as F w ; 4.11.3, determining, by using the classification model A 3 of maturity stage of organic matter of oil shale as constructed in step 4.10.9 and taking the feature F w as input, maturity of organic matter of oil shale is in which stage of three stages comprising dehydration, organic matter decomposition and semi-coke mineral decomposition; 4.11.4, in the case that the maturity of organic matter of oil shale as determined in step 4.11.3 is in stage of the organic matter decomposition, then separately analyzing a gas concentration signal of the target gas G obtained by a mid-infrared laser gas sensing system in the photoelectric fused system for underground in-situ exploration and thermal evolution stage monitoring of oil shale, where an unit time is recorded as o, a set constituted by p unit times is recorded as O={o 1 , o 2 , o 3 , . . . , o p }, an average concentration of the target gas G in each unit time o is used as a representative gas concentration in this unit time and its set is recorded as C={c 1 , c 2 , c 3 , . . . , c p }, and a change rate of concentration of the target gas G in each unit time o is recorded and its set is recorded as R={r 1 , r 2 , r 3 , . . . , r p }; 4.11.5, in the case of R i >R i−1 , where 1<i<m, then adjusting the oil shale underground in-situ conversion heating device and the formation pressure control device to increase an underground in-situ conversion heating temperature and formation pressure of oil shale, with subsequent steps as follows: 4.11.5.1, in the case of r i+1 >r i , where 1<i<m, and 2<m≤p, where p represents the number of units time, then maintaining the underground in-situ conversion heating temperature and the formation pressure of oil shale; 4.11.5.2, in the case of r i+1 ≤r i , where 1<i<m, and 2<m≤p, where p represents the number of units time, then adjusting the oil shale underground in-situ conversion heating device and the formation pressure control device to increase the underground in-situ conversion heating temperature and the formation pressure of oil shale; 4.11.6, in the case of r i ≤r i−1 , where 1<i<m, and 2<m≤p, where p represents the number of units time, then adjusting the oil shale underground in-situ conversion heating device and the formation pressure control device to increase the underground in-situ conversion heating temperature and the formation pressure of oil shale, with subsequent steps as follows: 4.11.6.1, in the case of r i+1 >r i , where 1<i<m, and 2<m≤p, where p represents the number of units time, then adjusting the oil shale underground in-situ conversion heating device and the formation pressure control device to increase the underground in-situ conversion heating temperature and the formation pressure of oil shale; 4.11.6.2, in the case of r i+1 ≤r i , where 1<i<m, and 2<m≤p, where p represents the number of units time, then maintaining the underground in-situ conversion heating temperature and the formation pressure of oil shale.

Description

BACKGROUND Technical Field The present disclosure belongs to the technical field of oil shale in-situ exploration and exploitation, and in particular, relates to a photoelectric fusion in-situ detection device and method for thermal volatile gases and thermal evolution stage (i.e., cracking state) of underground oil shale. Description of Related Art Oil shale is a sedimentary rock rich in kerogen organic matter, which is expected to be commercially developed through in-situ conversion technology and can be used as a supplementary energy source for traditional oil and gas. The oil shale has huge reserves on earth, and the exploitation and utilization of the oil shale are mainly affected by the effect of exploration and the thermal evolution stage monitoring of oil shale during exploitation. On the one hand, exploration of oil shale can evaluate the abundance of organic matter inside the oil shale, and type and evolution of the kerogen, so as to determine the exploitation value of the mine and guide subsequent exploitation. On the other hand, real-time thermal evolution stage monitoring of oil shale is conducive to quickly correcting the heating and pressure control process, reducing costs and improving energy return rate. Therefore, underground in-situ exploration and thermal evolution stage monitoring of oil shale are conducive to achieving accurate evaluation of oil shale deposit value and real-time monitoring of the mining process, guiding the selection of on-site conversion technology and regulation of the engineering process, to assist the efficient exploitation of underground oil and gas resources. Traditional exploration and evaluation of oil shale generally rely on on-site geological survey, drilling coring and further laboratory testing and analysis. The methods widely used for qualitative or quantitative monitoring of cracking process of oil shale include directly measuring vitrinite reflectance of rock samples, using spectral techniques to analyze structure and composition of rock, and establishing a numerical simulation method based on dynamic data obtained from pyrolysis experiments or data from actual mining. At present, these methods for the underground in-situ exploration and the thermal evolution stage monitoring of oil shale usually require complex sample preparation and analysis in the laboratory, which is time-consuming, expensive and cannot meet the requirements of in-situ real-time monitoring. The organic matter type, abundance, maturity and thermal evolution stage of oil shale can be accurately judged and monitored in real time through the qualitative and quantitative accurate detection of oil shale with different qualities and a multi-component gas generated in the process of cracking. As a rapidly developing gas detection technology, a gas sensor array has the advantages of high efficiency, fast speed, low cost, simple operation, and on-site real-time detection, and so on, and is receiving increasing attention in energy field. Optical gas detection technology converts a signal of gas molecules interacting with photons into an electrical signal, which has the advantages of good selectivity, high sensitivity, strong stability, and short response time and has been widely used in real-time monitoring of various gases in various environments. By using the gas sensor array and the optical gas detection technology, it is expected to realize rapid underground in-situ exploration and real-time thermal evolution stage monitoring of oil shale. However, the application of the two gas detection technologies in the field of in-situ exploration and in-situ real-time thermal evolution stage monitoring of oil shale still faces three key technical problems that need to be solved urgently. Firstly, in the process of the underground in-situ exploration and thermal evolution stage monitoring of oil shale, particles such as rock debris will adhere to surfaces of gas-sensitive detectors and an optical apparatus, which will affect the accuracy of the sensors and greatly damage the effect of exploration and monitoring. Secondly, too many sensors are easy to cause problems such as large apparatus volume and high energy consumption, which is not conducive to assembling of apparatus during oil shale exploration and exploitation. More importantly, currently the gas sensor arrays alone or the optical gas detection apparatus alone cannot achieve reliable underground in-situ exploration and thermal evolution stage monitoring of oil shale due to key technical defects. On the one hand, although the gas sensor arrays can detect the multi-component gas in real-time, they have a low accuracy and a high detection limit for single specific gases. On the other hand, although the optical gas detection technology provides more accurate detection results for the single specific gases, its real-time detection of the multi-component gas requires the use of a tunable light sources capable of covering a absorption wavelength of t