SHALE GAS WELL TYPE CLASSIFICATION METHOD

Description

CROSS-REFERENCE TO RELATED APPLICATION

This application claims priority to Chinese Patent Application No. 202411174236.7, filed on Aug. 26, 2024, which is herein incorporated by reference in its entirety.

TECHNICAL FIELD

The disclosure relates to the field of shale gas exploitation technologies, and more particularly to a shale gas well type classification method.

BACKGROUND

Shale gas is a hot topic in global oil and gas exploration. With the continuous promotion of production, a number of shale gas wells is increasing day by day. However, there is a significant difference in production capacities of different shale gas wells. For example, in the Luzhou Block of the Sichuan Basin, post-fracturing test production rates for wells L203, Y101H1-2, and Y101H3-4 are 137.9×10⁴ cubic meters per day (m³/d), 46.8×10⁴ m³/d, and 0.8×10⁴ m³/d, respectively. Therefore, it is essential to classify the shale gas wells reasonably and study production patterns of different types of the shale gas wells to guide the production of similar gas wells.

Based on static geological parameters (such as an effective thickness and a formation pressure) or dynamic production parameters (such as an open flow rate and a gas production per unit casing pressure drop), the shale gas wells are typically classified into three categories (Class I, Class II, and Class III) in the related art. This classification method has the following issues. First, static geological evaluation or dynamic production evaluation is used in isolation without combination, ignoring a trend of “geological-engineering integration”. Second, almost all current classification schemes classify the shale gas wells into three categories, without adjusting or refining gas well types according to actual production needs, which is not conducive to refinement of evaluation of single well production capacity. Third, traditional classification method is a hard classification, which is only applicable when each data point clearly belongs to a specific category; and it is not applicable when a category of a data point is unclear or there are overlapping features. For example, evaluation criteria for the shale gas wells in a certain block are as follows. If an effective thickness of a shale of a shale gas well is greater than 45 meters (m), a gas saturation of the shale gas well is greater than 50%, and a formation pressure coefficient of the shale gas well is greater than 1.2, the shale gas well is classified as the Class I. If the effective thickness of the shale of the shale gas well is between 30 m and 45 m, the gas saturation of the shale gas well is between 40% and 50%, and the formation pressure coefficient of the shale gas well is between 0.8 and 1.2, the shale gas well is classified as the Class II. However, if the effective thickness of the shale of the shale gas well is 60 m, the gas saturation of the shale gas well is 30%, and the formation pressure coefficient of the shale gas well is 0.9, the shale gas well cannot be directly classified as it has characteristics of both the Class I and the Class II.

Therefore, existing shale gas well type evaluation methods are unable to accurately and meticulously assess the shale gas wells, resulting in deviations in shale gas well classification.

SUMMARY

In response to a problem that existing shale gas well type evaluation methods are unable to accurately and meticulously assess shale gas wells, resulting in deviations in shale gas well classification, the disclosure provides a shale gas well type classification method.

A shale gas well type classification method provided by the disclosure includes the following steps S1 through S7.

In the step S1, raw data of evaluation parameters for stable production wells and trial production wells are acquired. The evaluation parameters include six parameters: an effective thickness, a gas saturation, a formation pressure, an open flow rate, a gas production per unit casing pressure drop, and a peak daily gas production. The raw data for the effective thickness, the gas saturation and the formation pressure of the evaluation parameters are acquired based on geological data, and the raw data for the open flow rate, the gas production per unit casing pressure drop and the peak daily gas production of the evaluation parameters are acquired based on production data.

In an embodiment, for a number of the stable production wells is at least 10.

In the step S2, to eliminate influence of dimensionality, standardization is performed on the raw data of the evaluation parameters to obtain standardized data of the evaluation parameters for the stable production wells and the trial production wells. The standardization is performed as per the following formula:

Z ⁢ x i ⁢ j = x i ⁢ j - min [ x i ⁢ j ] max [ x i ⁢ j ] - min [ x i ⁢ j ]

where Zx_ij represents a standardized value at an i-th row and a j-th column, X_ij represents a parameter value at the i-th row and the j-th column in the raw data, max [x_ij] represents a maximum value in the raw data, and min [x_ij] represents a minimum value in the raw data.

In the step S3, based on the standardized data of the evaluation parameters for the stable production wells, a distance coefficient d_ij between every two stable production wells of the stable production wells is calculated as per the following formula:

d i ⁢ j = ∑ k = 1 u ( x i ⁢ k - x j ⁢ k ) 2 , i , j = 1 , 2 , 3 , … , m

where x_ik represents a parameter value at the i-th row and a k-th column in the standardized data for the evaluation parameters of the stable production wells, x_jk represents a parameter value at a j-th row and the k-th column in the standardized data for the evaluation parameters of the stable production wells, m represents the number of the stable production wells, and u represents a number of the evaluation parameters.

In the step S4, the stable production wells are classified based on a principle that a correlation between two wells will be stronger when a distance coefficient between the two wells becomes small, including the following steps S41 through S43.

In the step S41, the distance coefficient between every two stable production wells of the stable production wells except for 0 are sorted in an ascending order to obtain sorted distance coefficients. Two stable production wells of the stable production wells corresponding to a distance coefficient ranked first in the sorted distance coefficients are merged into one class to form a first new entity to thereby obtain first-merged stable production wells comprising the first new entity and remaining stable production wells of the stable production wells for subsequent merging and classifying.

In the step S42, two of the first-merged stable production wells corresponding to a distance coefficient ranked second in the sorted coefficients are merged into one class to form a second new entity to thereby obtain second-merged stable production wells comprising the second new entity and remaining stable production wells of the first merged stable production wells for subsequent merging and classifying.

In the step S43, according to the above steps, a merging and classification operation is performed on two of the second-merged stable production wells corresponding to a distance coefficient ranked third in the sorted distance coefficients. The merging and classification operation is repeated until a class number of the stable production wells is q as required for production, thereby stopping the merging and classification operation to obtain q classes of the stable production wells, where 1≤q≤m, and q and m are positive integers.

In an embodiment, in the above steps S41 through S43, during the merging and classification operation, for the distance coefficient d_ij, one of the stable production wells corresponding to the i-th row is merged with another one of the stable production wells corresponding to the j-th column into a same class, and a row and a column corresponding to a larger number between i and j are removed. Then the merging and classification operation is repeated on remaining data.

In an embodiment, in the step S43, the merging and classification operation is stopped when a distance coefficient ranked in the sorted distance coefficients reaches a maximum value satisfying d_ij≤0.100 to thereby obtain the q classes of the stable production wells.

In the step S5, after determining the class number of the stable production wells, comprehensive evaluation parameters S of the stable production wells are calculated based on the standardized data of the evaluation parameters for the stable production wells. Each of the comprehensive evaluation parameters S is a sum of standardized values of the six evaluation parameters and is expressed as that: S=a standardized effective thickness value+a standardized gas saturation value+a standardized formation pressure value+a standardized open flow rate value-a standardized gas production value per unit casing pressure drop+a standardized peak daily gas production value. Subsequently, an average value of the comprehensive evaluation parameters S of the stable production wells in each of the q classes is calculated. The q classes of the stable production wells are designated sequentially as Class I gas wells, Class II gas wells, Class III gas wells, . . . , Class q gas wells in a descending order of the average value of the comprehensive evaluation parameters S.

In the step S6, for each of the q classes of the stable production wells, average values of the standardized data of the evaluation parameters for the stable production wells are calculated to construct a standard fuzzy set A, thereby obtaining standard fuzzy sets A₁, A₂, A₃, . . . ,A_q, where A=(an average effective thickness value, an average gas saturation value, an average formation pressure value, an average open flow rate value, an average gas production value per unit casing pressure drop, an average peak daily gas production value). For each of the trial production wells, average values of the standardized data of the evaluation parameters for the trial production wells are calculated to construct a to-be-identified fuzzy set B, thereby obtaining to-be-identified fuzzy sets B₁, B₂, B₃, . . . ,B_n, where n represents a number of the trial production wells, and B=(an average effective thickness value, an average gas saturation value, an average formation pressure value, an average open flow rate value, an average gas production value per unit casing pressure drop, an average peak daily gas production value).

In the step S7, a closeness degree between one of the to-be-identified fuzzy sets and each of the standard fuzzy sets is calculated. Based on a principle that a similarity between the standard fuzzy set and the to-be-identified fuzzy set will be higher when the closeness degree becomes greater, one of the trial production wells corresponding to the one of the to-be-identified fuzzy sets is classified into one of the q classes of the stable production wells corresponding to a highest closeness degree. A formula for calculating the closeness degree is expressed as follows:

σ ⁡ ( A , B ) = def ∑ k = 1 z [ A ⁡ ( x c ) ∧ B ⁡ ( x c ) ] ∑ k = 1 z [ A ⁡ ( x c ) ∨ B ⁡ ( x c ) ]

where σ(A,B) represents the closeness degree between A and B, A represents the standard fuzzy set, B represents the to-be-identified fuzzy set, z represents one of a total number of elements in the standard fuzzy set and a total number of elements in the to-be-identified fuzzy set, x_c represents a c-th element of the elements in one of the standard fuzzy set and the to-be-identified fuzzy set, Λ represents taking an infimum (i.e., a minimum value) between two elements respectively in the standard fuzzy set and the to-be-identified fuzzy set, and V represents taking a supremum (i.e., a maximum value) between the two elements respectively in the standard fuzzy set and the to-be-identified fuzzy set.

In an embodiment, the shale gas well type classification method further includes: classifying the trial production wells according to step S7 to obtain classification results of the trial production wells, and exploiting shale gas from the trial production wells based on the classification results of the trial production wells. For example, when the distance coefficient ranked in the sorted distance coefficients reaches the maximum value satisfying d_ij≤0.100, the class number of the stable production wells is 4, i.e., 4 classes of the stable production wells are obtained. Then, based on the comprehensive evaluation parameter, where production capacity of a stable production well will be higher when the comprehensive evaluation parameter becomes greater, the 4 classes of the stable production wells are designated sequentially as Class I gas wells (optimal), Class II gas wells (sub-optimal), Class III gas wells (medium), and Class IV gas wells (poor). Among them, an average value of the comprehensive evaluation parameters for Class I gas wells reaches 1.086, significantly higher than an average value of the comprehensive evaluation parameters 0.367 for Class IV gas wells, directly guiding the development of resources to prioritize high potential wells. For the trial production wells, by calculating their closeness degrees to the 4 classes of the stable production wells, each of the trial production wells is classified into a most closely matching class in the 4 classes of the stable production wells based on a nearest principle. In the subsequent development of the trial production wells, allocation of fracturing investment and exploitation resources strictly follows a priority order of Class I gas wells>Class II gas wells>Class III gas wells>Class IV gas wells, thereby improving the efficiency of shale gas field development. This method provided by the disclosure solves problems of classification isolation, category rigidity, and overlapping features that traditional methods cannot handle through specific mathematical transformation steps such as multi parameter standardization, distance coefficient matrix construction, dynamic fuzzy clustering, and proximity matching. It achieves dual improvement in resource optimization allocation and single well productivity evaluation accuracy.

Compared to the related art, the disclosure may achieve the following beneficial effects.

• • (1) The shale gas well type classification method provided by the disclosure employs a mathematical approach combining fuzzy clustering with fuzzy recognition. Initially, both geological static parameters and production dynamic parameters are considered to classify the stable production wells, and a number of classifications can be flexibly determined based on production needs. Subsequently, the trial production wells are classified into these classes. This method provided by the disclosure can better assess development potential of the trial production wells, ensuring that more technology and funding are allocated to high-potential trial production wells, thereby increasing overall return on investment. It is of significant importance for the efficient development of shale gas fields.
  • (2) Conventional methods consider only geological factors or production factors in isolation. In contrast, the disclosure integrates both the geological factors and the production factors, aligning with a trend of “geology-engineering integration”. This method provided by the disclosure facilitates a comprehensive evaluation of the potential of gas wells and minimizing risks to the greatest extent.
  • (3) The conventional methods classify the gas wells into three classes. In contrast, the disclosure allows for flexible determination of the class number based on the production needs and site-specific conditions, thereby enabling more precise evaluation of single well productivity.
  • (4) Conventional classification methods are hard classifications, that's to say, these conventional classification methods are only applicable when each data point distinctly belongs to a specific class. In contrast, the classification method provided by the disclosure remains applicable even when gas well data points exhibit ambiguous or overlapping characteristics between classes.

Other advantages, purposes, and features of the disclosure will be partially demonstrated through the following description, and partially will be understood by those skilled in the art through research and practice of the disclosure.

DETAILED DESCRIPTION OF EMBODIMENTS

A specific embodiment of the disclosure is described below, and it should be understood that the specific embodiment described herein is only for illustrating and explaining the disclosure, and is not intended to limit the disclosure.

A shale gas well type classification method of the disclosure is applied to a specific case and includes the following steps S1 through S7.

In the step S1, for a certain shale gas block, static evaluation parameters for a batch of stable production wells and trial production wells are acquired based on geological data. The static evaluation parameters include an effective thickness, a gas saturation and a formation pressure. Dynamic evaluation parameters for the batch of stable production wells and trial production wells are acquired based on production data. The dynamic evaluation parameters include an open flow rate, a gas production per unit casing pressure drop and a peak daily gas production. The greater the effective thickness, the more organic matter and pore space can be provided, and the higher the gas production. The gas saturation refers to a proportion of gas in rock pores. The higher the gas saturation in a shale reservoir, the stronger the gas storage capacity of shale pores, and the higher the gas production. As the formation pressure increases, a gas volume is compressed, and an amount of free gas stored in the same space increases, resulting in higher gas production. The open flow rate indicates an ideal flow rate of gas in the absence of any flow resistance. The higher the open flow rate, the more effectively gas can flow towards the wellbore in the subsurface reservoir, resulting in higher gas production. The unit casing pressure drop refers to the pressure drop required to maintain a certain gas production. When the unit casing pressure drop increases, it indicates that a greater pressure needs to be applied to maintain gas flow. A higher pressure drop makes gas flow more difficult, thereby reducing gas production. Therefore, under same conditions, the higher the gas production per unit casing pressure drop, the lower the gas production. A high peak daily gas production indicates strong production capacity of a gas well in an early stage of exploitation, which can contribute more gas production in a shorter time. Therefore, the gas well with a higher peak daily gas production often have a higher total gas production. Therefore, selection of the effective thickness, the gas saturation, the formation pressure, the open flow rate, the gas production per unit casing pressure drop, and the peak daily gas production as evaluation parameters in the disclosure can well reflect the gas production capacity of a single well. In this embodiment, 15 stable production wells and 10 trial production wells are selected. The raw data of the evaluation parameters for the stable production wells are shown in Table 1, and the raw data of the evaluation parameters for the trial production wells are shown in Table 2.

TABLE 1
raw data of evaluation parameters for stable production wells

					Gas
					production
				Open	per unit
	Effective	Gas	Formation	flow	casing	Peak daily gas
Well	thickness	saturation	pressure	rate	pressure drop	production
name	(m)	(%)	(MPa)	(104m3/d)	(104/MPa)	(104m3/d)

A1	38	56	102	770	240	15
A2	65	75	125	950	80	21
A3	20	35	70	690	370	9
A4	47	45	81	680	500	8
A5	14	24	77	710	380	12
A6	58	68	130	1000	150	17
A7	23	33	75	720	340	15
A8	35	51	95	810	290	17
AS	41	49	101	800	260	16
A10	60	82	118	860	120	24
A11	28	31	75	740	400	14
A12	44	41	89	650	440	7
A13	40	57	97	790	270	18
A14	49	39	80	700	470	5
A15	17	27	69	740	360	10

TABLE 2
raw data of evaluation parameters for trial production wells

				Open	Gas production
	Effective	Gas	Formation	flow	per unit casing	Peak daily gas
Well	thickness	saturation	pressure	rate	pressure drop	production
name	(m)	(%)	(MPa)	(104m3/d)	(104/MPa)	(104m3/d)

B1	45	56	69	670	450	13
B2	37	25	76	970	170	24
B3	65	63	107	820	340	30
B4	43	16	87	760	280	23
B5	15	32	95	780	300	17
B6	23	65	110	830	410	8
B7	34	44	76	910	420	15
B8	29	35	83	860	230	20
B9	51	58	94	790	350	18
B10	46	43	71	950	410	13

In the step S2, to eliminate influence of dimensionality, standardization is performed on the raw data of the evaluation parameters to obtain standardized data of the evaluation parameters for the stable production wells and the trial production wells. The standardization is performed as per the following formula:

Z ⁢ x i ⁢ j = x i ⁢ j - min [ x i ⁢ j ] max [ x i ⁢ j ] - min [ x i ⁢ j ]

where Zx_ij represents a standardized value at an i-th row and a j-th column, X_ij represents a parameter value at the i-th row and the j-th column in the raw data, max [x_ij] represents a maximum value in the raw data, and min [x_ij] represents a minimum value in the raw data.

The standardized data of the evaluation parameters for the stable production wells and the trial production wells are shown in Table 3 and Table 4, respectively.

TABLE 3
standardized data of evaluation parameters for stable production wells

				Open	Gas production	Peak daily
	Effective	Gas	Formation	flow	per unit casing	gas
Well	thickness	saturation	pressure	rate	pressure drop	production
name	(m)	(%)	(MPa)	(104m3/d)	(104/MPa)	(104m3/d)

A1	0.033	0.051	0.097	0.769	0.236	0.010
A2	0.060	0.070	0.121	0.950	0.075	0.016
A3	0.015	0.030	0.065	0.688	0.367	0.004
A4	0.042	0.040	0.076	0.678	0.497	0.003
A5	0.009	0.019	0.072	0.709	0.377	0.007
A6	0.053	0.063	0.126	1.000	0.146	0.012
A7	0.018	0.028	0.070	0.719	0.337	0.010
A8	0.030	0.046	0.090	0.809	0.286	0.012
A9	0.036	0.044	0.096	0.799	0.256	0.011
A10	0.055	0.077	0.114	0.859	0.116	0.019
A11	0.023	0.026	0.070	0.739	0.397	0.009
A12	0.039	0.036	0.084	0.648	0.437	0.002
A13	0.035	0.052	0.092	0.789	0.266	0.013
A14	0.044	0.034	0.075	0.698	0.467	0.000
A15	0.012	0.022	0.064	0.739	0.357	0.005

TABLE 4
standardized data of evaluation parameters for trial production wells

				Open	Gas production
	Effective	Gas	Formation	flow	per unit casing	Peak daily gas
Well	thickness	saturation	pressure	rate	pressure drop	production
name	(m)	(%)	(MPa)	(104m3/d)	(104/MPa)	(104m3/d)

B1	0.038	0.050	0.063	0.688	0.459	0.005
B2	0.030	0.018	0.071	1.000	0.168	0.017
B3	0.059	0.057	0.103	0.844	0.345	0.023
B4	0.036	0.008	0.082	0.782	0.283	0.016
B5	0.007	0.025	0.090	0.802	0.304	0.009
B6	0.016	0.059	0.106	0.854	0.418	0.000
B7	0.027	0.037	0.071	0.938	0.428	0.007
B8	0.022	0.028	0.078	0.886	0.231	0.012
B9	0.045	0.052	0.089	0.813	0.356	0.010
B10	0.040	0.036	0.065	0.979	0.418	0.005

In the step S3, based on the standardized data of the evaluation parameters for the stable production wells, a distance coefficient d_ij between every two stable production wells is calculated as per the following formula:

d i ⁢ j = ∑ k = 1 u ( x i ⁢ k - x j ⁢ k ) 2 , i , j = 1 , 2 , 3 , … , m

where x_ik represents a parameter value at the i-th row and a k-th column in the standardized data for the evaluation parameters of the stable production wells, x_jk represents a parameter value at a j-th row and the k-th column in the standardized data for the evaluation parameters of the stable production wells, m represents a number of the stable production wells, u represents a number of the evaluation parameters, and k represents a k-th column in the standardized data of the evaluation parameters for the stable production wells as a matrix.

The distance coefficient d_ij between every two stable production wells is calculated as shown in Table 5.

TABLE 5
distance coefficient between every two stable production wells

Well name

A1

A2

A3

A4

A5

A6

A7

A8

A9

A10

A11

A12

A13

A14

A15

A1	0	0.245	0.159	0.278	0.160	0.251	0.119	0.065	0.037	0.156	0.168	0.235	0.337	0.244	0.134
A2	0.245	0	0.400	0.505	0.396	0.087	0.358	0.258	0.239	0.100	0.392	0.474	0.253	0.470	0.363
A3	0.159	0.400	0	0.135	0.027	0.390	0.044	0.149	0.162	0.314	0.060	0.087	0.148	0.106	0.052
A4	0.278	0.505	0.135	0	0.131	0.480	0.168	0.249	0.271	0.426	0.120	0.068	0.257	0.037	0.158
A5	0.160	0.396	0.027	0.131	0	0.381	0.044	0.141	0.157	0.314	0.040	0.093	0.145	0.099	0.037
A6	0.251	0.087	0.390	0.480	0.381	0	0.348	0.242	0.233	0.145	0.370	0.460	0.246	0.445	0.346
A7	0.119	0.358	0.044	0.168	0.044	0.348	0	0.108	0.119	0.273	0.064	0.126	0.106	0.135	0.031
A8	0.065	0.258	0.149	0.249	0.141	0.242	0.108	0	0.033	0.184	0.134	0.221	0.030	0.214	0.107
A9	0.037	0.239	0.162	0.271	0.157	0.233	0.119	0.033	0	0.159	0.157	0.236	0.017	0.235	0.126
A10	0.156	0.100	0.314	0.426	0.314	0.145	0.273	0.184	0.159	0	0.315	0.389	0.171	0.392	0.283
A11	0.168	0.392	0.060	0.120	0.040	0.370	0.064	0.134	0.157	0.315	0	0.102	0.145	0.085	0.043
A12	0.235	0.474	0.087	0.068	0.093	0.460	0.126	0.221	0.236	0.389	0.102	0	0.222	0.060	0.126
A13	0.037	0.253	0.148	0.257	0.145	0.246	0.106	0.030	0.017	0.171	0.145	0.222	0	0.222	0.114
A14	0.244	0.470	0.106	0.037	0.099	0.445	0.135	0.214	0.235	0.392	0.085	0.060	0.222	0	0.123
A15	0.134	0.363	0.052	0.158	0.037	0.346	0.031	0.107	0.126	0.283	0.043	0.126	0.114	0.123	0

In the step S4, for any two gas wells, a correlation between the two wells will be stronger when a distance coefficient between the two wells becomes small. Based on this principle, the stable production wells are merged and classified. Specific steps for merging and classification are as follows.

• • (1) Prior to any merging, the stable production wells are classified into 15 classes/types, with each of the stable production wells constituting a separate class.
  • (2) In a first merge, a minimum distance coefficient d_13,9=0.017 (excluding zero) is found. At this point, the stable production wells are classified into 14 classes, where wells A13 and A9 are merged into one class to form a first new entity, each remaining well is maintained as a separate class, and a 13-th row and a 13-th column are removed.
  • (3) In a second merge, a minimum distance coefficient d_5,3=0.027 (excluding zero) in remaining data is found. The stable production wells are now classified into 13 classes, where wells A13 and A9 are merged into one class to form a first new entity, wells A5 and A₃ are merged into one class to form a second new entity, each remaining well is maintained as a separate class, and a 5-th row and a 5-th column are removed.
  • (4) In a third merge, a minimum distance coefficient d_15,7=0.031 (excluding zero) in remaining data is found. The stable production wells are now classified into 12 classes, where wells A13 and A9 are merged into one class to form the first new entity, wells A5 and A₃ are merged into one class to form the second new entity, wells A15 and A7 are merged into still another class, each remaining well is maintained as a separate class, and a 15-th row and a 15-th column are removed.
  • (5) In a fourth merge, a minimum distance coefficient d_8,9=0.033 (excluding zero) in remaining data is found. The stable production wells are now classified into 11 classes, where wells A13, A9 and A8 are merged into one class to form the first new entity, wells A5 and A₃ are merged into one class to form the second new entity, wells A15 and A7 are merged into one class to form a third new entity, each remaining well is maintained as a separate class, and a 9-th row and a 9-th column are removed.
  • (6) In a fifth merge, a minimum distance coefficient d_14,4=0.037 (excluding zero) in remaining data is found. The stable production wells are now classified into 10 classes, where wells A13, A9 and A8 are merged into one class to from the first new entity, wells A5 and A₃ are merged into one class to form the second new entity, wells A15 and A7 are merged into one class to form the third new entity, wells A14 and A4 are merged into one class to form a fourth new entity, each remaining well is maintained as a separate class, and a 14-th row and a 14-th column are removed.
  • (7) In a sixth merge, a minimum distance coefficient d_3,7=0.044 (excluding zero) in remaining data is found. The stable production wells are now classified into 9 classes, where wells A13, A9 and A8 are merged into one class to form the first new entity, wells A5, A₃, A15 and A7 are merged into one class to form the second new entity, wells A14 and A4 are merged into one class to form the third new entity, each remaining well is maintained as a separate class, and a 7-th row and a 7-th column are removed.
  • (8) In a seventh merge, a minimum distance coefficient d_11,3=0.060 (excluding zero) in remaining data is found. The stable production wells are now classified into 8 classes, where wells A13, A9 and A8 are merged into one class to form the first new entity, wells A5, A₃, A15, A7 and A11 are merged into one class to form the second new entity, wells A14 and A4 are merged into one class to form the third new entity, each remaining well is maintained as a separate class, and a 11-th row and a 11-th column are removed.
  • (9) In an eighth merge, a minimum distance coefficient d_8,1=0.065 (excluding zero) in remaining data is found. The stable production wells are now classified into 7 classes, where wells A13, A9, A8 and A₁ are merged into one class to form the first new entity, wells A5, A₃, A15, A7 and A11 are merged into one class to from the second new entity, wells A14 and A4 are merged into one class to form the third new entity, each remaining well is maintained as a separate class, and an 8-th row and an 8-th column are removed.
  • (10) In a ninth merge, a minimum distance coefficient d_12,4=0.068 (excluding zero) in remaining data is found. The stable production wells are now classified into 6 classes, where wells A13, A9, A8 and A₁ are merged into one class to form the first new entity, wells A5, A₃, A15, A7 and A11 are merged into one class to form the second new entity, wells A14, A4 and A12 are merged into one class to form the third new entity, each remaining well is maintained as a separate class, and a 12-th row and a 12-th column are removed.
  • (11) In a tenth merge, a minimum distance coefficient d_2,6=0.087 (excluding zero) in remaining data is found. The stable production wells are now classified into 5 classes, where wells A13, A9, A8 and A₁ are merged into one class to form the first new entity, wells A5, A3, A15, A7 and A11 are merged into one class to form the second new entity, wells A14, A4 and A12 are merged into one class to form the third new entity, wells A2 and A6 are merged into one class to form the fourth new entity, well A10 remains as a separate class and a 6-th row and a 6-th column are removed.
  • (12) In an eleventh merge, a minimum distance coefficient d_10,2=0.100 (excluding zero) in remaining data is found. The stable production wells are now classified into 4 classes, where wells A13, A9, A8 and A₁ are merged into one class to form the first new entity, wells A5, A₃, A15, A7 and A11 are merged into one class to form the second new entity, wells A14, A4 and A12 are merged into one class to form the third new entity, wells A₂, A6 and A10 are merged into one class to form the fourth new entity, and a 10-th row and a 10-th column are removed.
  • (13) In a twelfth merge, a minimum distance coefficient d_4,3=0.135 (excluding zero) in remaining data is found. The stable production wells are now classified into 3 classes, where wells A13, A9, A8 and A₁ are merged into one class to from the first new entity, wells A5, A₃, A15, A7, All, A4, A14 and A12 are merged into one class to form the second new entity, wells A₂, A6 and A10 are merged into one class to form the third new entity, and a 4-th row and a 4-th column are removed.
  • (14) In a thirteenth merge, a minimum distance coefficient d_3,1=0.159 (excluding zero) in remaining data is found. The stable production wells are now classified into 2 classes, where wells A13, A9, A8, A₁, A5, A₃, A15, A7, A11, A4, A14 and A12 are merged into one class to form the first new entity, wells A₂, A6 and A10 are merged into one class to from the second new entity, and a 3-th row and a 3-th column are removed.
  • (15) In a fourteenth merge, a minimum distance coefficient d_2,1=0.245 (excluding zero) in remaining data is found. At this point, all of the stable production wells are merged into a single class.

In this embodiment, there are a total of 15 stable production wells, which can be classified into a maximum of 15 classes and a minimum of 1 class, as shown in Table 6. In the table, same numbers indicate that the stable production wells belong to a same class under the respective classification scheme.

TABLE 6
classification scheme

Well	15	14	13	12	11	10	9	8	7	6	5	4	3	2	1
name	types	types	types	types	types	types	types	types	types	types	types	types	types	types	type

A1	1	1	1	1	1	1	1	1	1	1	1	1	1	1	1
A2	2	2	2	2	2	2	2	2	2	2	2	2	2	2	1
A3	3	3	3	3	3	3	3	3	3	3	3	3	3	1	1
A4	4	4	4	4	4	4	4	4	4	4	4	3	3	1	1
A5	5	5	3	3	3	3	3	3	3	3	3	3	3	1	1
A6	6	6	5	5	5	5	5	5	5	5	2	2	2	2	1
A7	7	7	6	6	6	6	3	3	3	3	3	3	3	1	1
A8	8	8	7	7	7	7	6	1	1	1	1	1	1	1	1
A9	9	9	8	8	7	7	6	1	1	1	1	1	1	1	1
A10	10	10	9	9	8	8	7	6	6	6	5	4	2	2	1
A11	11	11	10	10	9	9	8	7	3	3	3	3	3	1	1
A12	12	12	11	11	10	10	9	8	7	4	4	3	3	1	1
A13	13	9	8	8	7	7	6	1	1	1	1	1	1	1	1
A14	14	13	12	12	11	4	4	4	4	4	4	3	3	1	1
A15	15	14	13	6	6	6	3	3	3	3	3	3	3	1	1

In this embodiment, the stable production wells can be classified into 1 to 15 classes. In practical applications, a class number of the stable production can be determined based on production needs. In this embodiment, a class number are determined based on the distance coefficient d_ij. When the distance coefficient d_ij≤0.100, it indicates a strong correlation between two samples, and they can be classified into a same class. The class number corresponding to a maximum value of the distance coefficient satisfying d_ij≤0.100 is selected as a final class number.

In this embodiment, during the eleventh merge, the distance coefficient d_10,2=0.100 meets the aforementioned condition. At this point, wells A13, A9, A8 and A1 are merged into Class 1; wells A5, A3, A15, A7 and A11 are merged into Class 2; wells A14, A4 and A12 are merged into Class 3; and wells A2, A6 and A10 are merged into Class 4, resulting in a total of 4 classes.

In the step S5, after determining the class number of the stable production wells, comprehensive evaluation parameters S of the stable production wells are calculated based on the standardized data of the evaluation parameters for the stable production wells. An average value of the comprehensive evaluation parameters S of the stable production wells in each of the q classes is calculated, and the 4 classes of the stable production wells are sorted by in descending order of the average value of the comprehensive evaluation parameters S to determine their relative quality. Each of the comprehensive evaluation parameters Sis calculated as per the following formula:

S=a standardized effective thickness value+a standardized gas saturation value+a standardized formation pressure value+a standardized open flow rate value−a standardized gas production value per unit casing pressure drop+a standardized peak daily gas production value.

Based on the formula for calculating the comprehensive evaluation parameter, average values of the comprehensive evaluation parameters for each class of the stable production wells are as follows: S₁=0.718 for Class 1 wells, S₂=0.468 for Class 2 wells, S₃=0.367 for Class 3 wells, and S₄=1.086 for Class 4 wells. Since S₄>S₁>S₂>S₃, wells A2, A6 and A10 are determined to be the best and are classified as Class I gas wells; wells A13, A9, A8 and A1 are determined to be relatively better and are classified as Class II gas wells; ells A5, A₃, A15, A7 and A11 are determined to be relatively poorer and are classified as Class III gas wells; and wells A14, A4 and A12 are determined to be the poorest and are classified as Class IV gas wells.

In the step S6, for each of the q classes of the production wells, average values of the standardized data of the evaluation parameters for the stable production wells are calculated to construct a standard fuzzy set A. In a same manner, for each of the trial production wells, average values of the standardized data of the evaluation parameters for the trial production wells are calculated to construct a to-be-identified fuzzy set B.

In this embodiment, standard fuzzy sets established based on the stable-production wells are as follows:

• • the standard fuzzy set A₁ for the Class I gas wells as that: A₁=(0.056, 0.070, 0.120, 0.936, 0.112, 0.016);
  • the standard fuzzy set A₂ for the Class II gas wells as that: A₂=(0.034, 0.048, 0.094, 0.791. 0.261, 0.012);
  • the standard fuzzy set A₃ for the Class III gas wells as that: A₃=(0.015, 0.025, 0.069, 0.719. 0.367, 0.007); and
  • the standard fuzzy set A4 for the Class IV gas wells as that: A4=(0.042, 0.037, 0.079, 0.675. 0.467, 0.002).

To-be-identified fuzzy sets established based on the trial production wells are as follows:

• • the to-be-identified fuzzy set B₁ for well B₁ as that: B₁=(0.038, 0.050, 0.063, 0.688, 0.459, 0.005);
  • the to-be-identified fuzzy set B₂ for well B₂ as that: B₂=(0.030, 0.018, 0.071, 1.000, 0.168, 0.017);
  • the to-be-identified fuzzy set B₃ for well B₃ as that: B₃=(0.059, 0.057, 0.103, 0.844, 0.345, 0.023);
  • the to-be-identified fuzzy set B₄ for well B₄ as that: B₄=(0.036, 0.008, 0.082, 0.782, 0.283, 0.016);
  • the to-be-identified fuzzy set B₅ for well B₅ as that: B₅=(0.007, 0.025, 0.090, 0.802, 0.304, 0.009);
  • the to-be-identified fuzzy set B₆ for well B₆ as that: B₆=(0.016, 0.059, 0.106, 0.854, 0.418, 0.000);
  • the to-be-identified fuzzy set B₇ for well B₇ as that: B₇=(0.027, 0.037, 0.071, 0.938, 0.428, 0.007);
  • the to-be-identified fuzzy set B₈ for well B₈ as that: B₈=(0.022, 0.028, 0.078, 0.886, 0.231, 0.012);
  • the to-be-identified fuzzy set By for well B₉ as that: B₉=(0.045, 0.052, 0.089, 0.813, 0.356, 0.010); and
  • the to-be-identified fuzzy set B₁₀ for well B₁₀ as that: B₁₀=(0.040, 0.036, 0.065, 0.979, 0.418, 0.005).

In the step S7, a closeness degree between one of the to-be-identified fuzzy sets and each of the standard fuzzy sets is calculated, and a formula for calculating the closeness degree is expressed as follows:

σ ⁡ ( A , B ) = def ∑ k = 1 z [ A ⁡ ( x c ) ∧ B ⁡ ( x c ) ] ∑ k = 1 z [ A ⁡ ( x c ) ∨ B ⁡ ( x c ) ]

where σ(A,B) represents the closeness degree between A and B, A represents the standard fuzzy set, B represents the to-be-identified fuzzy set, z represents one of a total number of elements in the standard fuzzy set and a total number of elements in the to-be-identified fuzzy set, x represents a c-th element of the elements in one of the standard fuzzy set and the to-be-identified fuzzy set, A represents taking an infimum between two elements respectively in the standard fuzzy set and the to-be-identified fuzzy set, and V represents taking a supremum between the two elements respectively in the standard fuzzy set and the to-be-identified fuzzy set.

The closeness degrees are calculated as shown in Table 7. Based on a principle that a similarity between the standard fuzzy set and the to-be-identified fuzzy set will be higher when the closeness degree becomes greater, one of the trial production wells corresponding to the one of the to-be-identified fuzzy sets is classified into one of the q classes of the stable production wells corresponding to a highest closeness degree.

TABLE 7
closeness degrees σ between to-be-identified fuzzy sets and standard fuzzy sets

	B1	B2	B3	B4	B5	B6	B7	B8	B9	B10

A1	0.577	0.826	0.765	0.915	0.697	0.710	0.731	0.796	0.550	0.720
A2	0.761	0.749	0.867	0.746	0.916	0.817	0.769	0.870	0.735	0.750
A3	0.868	0.660	0.812	0.608	0.863	0.818	0.797	0.765	0.888	0.773
A4	0.963	0.587	0.759	0.571	0.753	0.800	0.789	0.680	0.935	0.768
	IV	I	II	I	II	III	III	II	IV	III

As shown in Table 7, among the trial production wells B1 to B10, wells B2 and B4 are classified as the Class I gas wells, wells B3, B5, and B8 are classified as the Class II gas wells, wells B6, B7, and B10 are classified as the Class III gas wells, and wells B1 and B9 are classified as the Class IV gas wells.

The above description is only the specific embodiment of the disclosure and does not limit the disclosure in any form. Although the disclosure has been described above with reference to the specific embodiment, it is not intended to limit the disclosure. Any those skilled in the art, without departing from the scope of a technical solution of the disclosure, may make certain modifications or adaptations to the technical content disclosed above, resulting in equivalent embodiments. Any simple modifications, equivalent variations, and embellishments made to the above embodiments, which do not deviate from the essence of the technical solution of the invention, still fall within the scope of the technical solution of the disclosure.