A drilling fluid waste composition classification optimization method based on a drilling scenario

Abstract

The application discloses a drilling fluid waste composition classification optimization method based on a drilling scene, and comprises the following steps: a waste composition determination model is constructed by utilizing drilling scene features, drilling fluid category features and waste composition; a pollutant composition determination model is constructed by utilizing drilling scene features, drilling fluid category features, waste composition and pollutant composition; the waste composition determination model is applied to a target drilling to determine the waste composition of drilling fluid waste in the target drilling as a waste composition target composition; and the pollutant composition determination model is applied to the target drilling to determine a pollutant composition target composition. The application provides a detection direction for pollutant composition detection, avoids a redundant composition detection method for a pollutant composition that does not exist in drilling waste liquid, and realizes directional detection of drilling fluid waste composition to achieve efficiency optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of drilling fluid waste classification, and particularly relates to a drilling fluid waste composition classification optimization method based on a drilling scene. BACKGROUND

[0002] Drilling waste liquid has complex components and is generally stored in a mud pit during drilling. The drilling waste liquid generated during drilling has high pollution load, low utilization rate of cuttings and waste liquid, and causes great difficulty in drilling waste liquid treatment. In addition, the drilling operation process has the characteristics of fluidity and dispersity, which brings serious environmental safety hazards to drilling work.

[0003] According to the pollution components in the drilling waste liquid, the drilling waste liquid is divided into COD, petroleum, phenol, heavy metal and inorganic sulfide. The pollution components in the drilling waste liquid are classified to provide analysis basis for treating the main pollution components in the drilling waste liquid. In the prior art, unknown components of the drilling waste liquid are analyzed by applying various component detection methods without distinction and direction, a large number of component detection methods need to be applied, and a large number of detection steps need to be performed. However, the component detection method for components not in the drilling waste liquid is invalid and redundant detection. Therefore, the non-directional component detection process is redundant and complex, which leads to low component detection efficiency. SUMMARY

[0004] The present application aims to provide a drilling fluid waste composition classification optimization method based on a drilling scene to solve the technical problem of redundant and complex non-directional component detection process in the prior art, which leads to low component detection efficiency.

[0005] To solve the above technical problem, the present application specifically provides the following technical scheme:

[0006] A drilling fluid waste composition classification optimization method based on a drilling scene, comprising the following steps:

[0007] Step S1, drilling fluid waste under multiple drilling scenes is obtained, and a non-discriminatory full detection of the drilling fluid waste is performed by using a pollutant component detection method set to determine the waste component composition and the pollutant component composition in the drilling fluid waste. The pollutant component detection method set contains all pollutant component detection methods that can be used for drilling fluid waste pollution detection;

[0008] Step S2, drilling scene features and drilling fluid category features are extracted from the drilling scene and the drilling fluid waste, and a waste component determination model is constructed by using the drilling scene features, the drilling fluid category features and the waste component composition. A pollutant component determination model is constructed by using the drilling scene features, the drilling fluid category features, the waste component composition and the pollutant component composition;

[0009] Step S3, applying the waste composition determination model to the target drilling well to determine the waste composition of the drilling fluid waste in the target drilling well as the waste composition target composition, and applying the pollutant composition determination model to the target drilling well to determine the pollutant composition of the drilling fluid waste in the target drilling well as the pollutant composition target composition, and screening the pollutant composition target detection method from the pollutant composition detection method set according to the pollutant composition target composition of the target drilling well, and detecting the pollutant composition target composition by using the pollutant composition target detection method to obtain the component concentration of the pollutant composition target composition of the target drilling well, so as to realize the directional detection of the drilling fluid waste composition and achieve efficiency optimization.

[0010] As a preferred scheme of the present application, the pollutant composition detection method set comprises spectrophotometry, atomic fluorescence spectrophotometry, diphenyl carbonyl dihydrazine colorimetry, fluorine reagent colorimetry, dichromate method, dilution and inoculation method, weight method, dilution method, glass click method, petroleum ether extraction weight method, luminescent bacteria method;

[0011] The spectrophotometry is used for component detection of lead and cadmium in the waste composition;

[0012] The atomic fluorescence spectrophotometry is used for component detection of mercury and arsenic in the waste composition;

[0013] The diphenyl carbonyl dihydrazine colorimetry is used for component detection of total chromium and hexavalent chromium in the waste composition;

[0014] The fluorine reagent colorimetry is used for component detection of fluoride in the waste composition;

[0015] The dichromate method is used for component detection of chemical oxygen demand in the waste composition;

[0016] The dilution and inoculation method is used for component detection of biological oxygen demand in the waste composition;

[0017] The glass click method is used for component detection of suspended solids in the waste composition;

[0018] The petroleum ether extraction weight method is used for component detection of petroleum in the waste composition;

[0019] The luminescent bacteria method is used for component detection of biological toxicity in the waste composition;

[0020] The methylene blue colorimetry is used for component detection of sulfide in the waste composition;

[0021] The aminoantipyrine photometry is used for component detection of phenol in the waste composition.

[0022] As a preferred scheme of the present application, the waste component composition of the drilling fluid waste is determined by indiscriminately and fully detecting the drilling fluid waste by using the set of pollutant component detection methods, and the set of pollutant component detection methods comprises:

[0023] The spectrophotometry, atomic fluorescence spectrophotometry, diphenyl carbonyl dihydrazine colorimetry, fluorine reagent colorimetry, dichromate method, dilution and inoculation method, weight method, dilution method, glass tapping method, petroleum ether extraction weight method and luminous bacteria method in the set of pollutant component detection methods are used to detect the components of lead, cadmium, mercury, arsenic, total chromium, hexavalent chromium, fluoride, chemical oxygen demand, biological oxygen demand, suspended solids, petroleum components, sulfide components and phenols in sequence, wherein,

[0024] When the component concentration of lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand / biological oxygen demand / suspended solids / petroleum / sulfide / phenol is higher than 0, the lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand / biological oxygen demand / suspended solids / petroleum / sulfide / phenol is taken as the waste component of the drilling fluid waste;

[0025] When the component concentration of lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand / biological oxygen demand / suspended solids / petroleum / sulfide / phenol is higher than the standard threshold value, the lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand / biological oxygen demand / suspended solids / petroleum / sulfide / phenol is taken as the pollutant component of the drilling fluid waste.

[0026] As a preferred scheme of the present application, the drilling scene features and drilling fluid category features are extracted from the drilling scene and the drilling fluid waste, and the extraction comprises:

[0027] The drilling scene features are extracted from each drilling scene, and the drilling fluid category features are extracted from the drilling fluid waste in each drilling scene;

[0028] The drilling scene features comprise drilling geological features and drilling structural features.

[0029] The drilling fluid category features comprise component composition of the drilling fluid and component concentration of the component composition of the drilling fluid.

[0030] As a preferred scheme of the present application, the waste component determination model is constructed by using the drilling scene features and the drilling fluid category features and the waste component composition, and the construction comprises:

[0031] The drilling scene features and the drilling fluid category features are taken as input items of the BP neural network, the waste component composition is taken as an output item of the BP neural network, and the waste component determination model is obtained by network training of the input items and the output items by using the BP neural network;

[0032] The model expression of the waste composition determination model is:

[0033] [E] m = BP ([Scene_Features] k , [Element_Features] r ) ;

[0034] In the formula, [E] m is a waste composition sequence composed of m waste components, [Scene_Features] k is a drilling scene feature vector composed of k drilling scene feature components, [Element_Features] r is a drilling fluid category feature vector composed of r drilling fluid category feature components, and BP is a BP neural network.

[0035] As a preferred scheme of the present application, the waste composition determination model is constructed by using drilling scene features, drilling fluid category features, and waste composition and pollutant composition, and includes the following steps.

[0036] The drilling scene features, drilling fluid category features, and waste composition are taken as the second input items of the BP neural network, the pollutant composition is taken as the second output item of the BP neural network, and the waste composition determination model is obtained by network training of the second input items and the second output item by using the BP neural network.

[0037] The model expression of the waste composition determination model is:

[0038] [W] n = BP ([Scene_Features] k , [Element_Features] r , [E] m ) ;

[0039] In the formula, [W] n is a waste composition sequence composed of n pollutant components, [Scene_Features] k is a drilling scene feature vector composed of k drilling scene feature components, [Element_Features] r is a drilling fluid category feature vector composed of r drilling fluid category feature components, and [E] mThe waste component composition sequence is composed of m waste components, BP is a BP neural network, n, m, k, and r are total amounts of pollutant components, total amounts of waste components, total amounts of drilling scene feature components, and total amounts of drilling fluid category feature components, respectively.

[0040] As a preferred scheme of the present application, the waste component composition of the drilling fluid waste in the target drilling is determined as a target waste component composition by applying the waste component determination model to the target drilling.

[0041] The drilling scene feature of the target drilling and the drilling fluid category feature of the drilling fluid used in the target drilling are input into the waste component determination model, and the waste component composition of the drilling fluid waste in the target drilling is output by the waste component determination model.

[0042] As a preferred scheme of the present application, the pollutant component composition of the drilling fluid waste in the target drilling is determined as a target pollutant component composition by applying the pollutant component determination model to the target drilling.

[0043] The drilling scene feature of the target drilling, the drilling fluid category feature of the drilling fluid used in the target drilling, and the target waste component composition are input into the pollutant component determination model, and the pollutant component composition of the drilling fluid waste in the target drilling is output by the pollutant component determination model.

[0044] As a preferred scheme of the present application, the pollutant component target detection method is screened out from the pollutant component detection method set according to the target pollutant component composition of the target drilling.

[0045] The pollutant component target detection method for detecting each pollutant component is matched from the pollutant component detection method set according to the pollutant component in the target pollutant component composition.

[0046] The pollutant component and the pollutant component target detection method are one-to-one corresponding.

[0047] As a preferred scheme of the present application, the drilling scene feature and the drilling fluid category feature are respectively subjected to feature normalization processing before model operation.

[0048] Compared with the prior art, the present application has the following beneficial effects:

[0049] This invention extracts drilling scene features and drilling fluid category features from drilling scenarios and drilling fluid waste. It then constructs a waste composition determination model using these features and waste composition, and a contaminant composition determination model using these features and contaminant composition. These waste composition determination models are applied to target wells to determine the waste composition of drilling fluid waste as the target waste composition, and the contaminant composition determination model is applied to target wells to determine the contaminant composition of drilling fluid waste as the target contaminant composition. This provides directionality for contaminant detection, avoiding redundant component detection methods for contaminants not present in drilling waste fluid, thus achieving targeted detection of drilling fluid waste composition for efficiency optimization. Attached Figure Description

[0050] To more clearly illustrate the embodiments of the present invention or the technical solutions in 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 merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0051] Figure 1 A flowchart illustrating the drilling fluid waste composition classification and optimization method provided in this embodiment of the invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] like Figure 1 As shown, this invention provides a method for classifying and optimizing the composition of drilling fluid waste based on drilling scenarios, including the following steps:

[0054] Step S1: Obtain drilling fluid waste from multiple drilling scenarios, and use a set of pollutant component detection methods to perform non-discriminatory full detection on the drilling fluid waste to determine the composition of waste components and pollutant components in the drilling fluid waste. The set of pollutant component detection methods includes all pollutant component detection methods that can be used for drilling fluid waste pollution detection. Waste components are the various components in drilling fluid waste, and pollutant components refer to the components in the waste components that exceed the emission standards and need to be treated.

[0055] The pollutant component detection method set includes spectrophotometry, atomic fluorescence spectrophotometry, diphenyl carbonyl dihydrazine colorimetry, fluorine reagent colorimetry, dichromate method, dilution and inoculation method, gravimetric method, dilution method, glass click method, petroleum ether extraction gravimetric method, luminescent bacteria method;

[0056] The spectrophotometry is used for component detection of lead and cadmium in waste components;

[0057] The atomic fluorescence spectrophotometry is used for component detection of mercury and arsenic in waste components;

[0058] The diphenyl carbonyl dihydrazine colorimetry is used for component detection of total chromium and hexavalent chromium in waste components;

[0059] The fluorine reagent colorimetry is used for component detection of fluoride in waste components;

[0060] The dichromate method is used for component detection of chemical oxygen demand (COD) in waste components;

[0061] The dilution and inoculation method is used for component detection of biological oxygen demand (BOD) in waste components;

[0062] The glass click method is used for component detection of suspended solids in waste components;

[0063] The petroleum ether extraction gravimetric method is used for component detection of petroleum in waste components;

[0064] The luminescent bacteria method is used for component detection of biological toxicity components in waste components;

[0065] The methylene blue colorimetry is used for component detection of sulfide in waste components;

[0066] The aminoantipyrine photometry is used for component detection of phenols in waste components

[0067] The pollutant component detection method set is used for indiscriminate full detection of drilling fluid waste to determine the composition of waste components in the drilling fluid waste, including:

[0068] The spectrophotometry, atomic fluorescence spectrophotometry, diphenyl carbonyl dihydrazine colorimetry, fluorine reagent colorimetry, dichromate method, dilution and inoculation method, gravimetric method, dilution method, glass click method, petroleum ether extraction gravimetric method, and luminescent bacteria method in the pollutant component detection method set are used for component detection of lead, cadmium, mercury, arsenic, total chromium, hexavalent chromium, fluoride, chemical oxygen demand (COD), biological oxygen demand (BOD), suspended solids, petroleum components, sulfide components, and phenols, respectively, wherein,

[0069] when the constituent concentration of lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand (COD) / biological oxygen demand (BOD) / suspended matter / oil / sulfide / phenol is higher than 0, the lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand (COD) / biological oxygen demand (BOD) / suspended matter / oil / sulfide / phenol is used as a waste constituent of the drilling fluid waste;

[0070] when the constituent concentration of lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand (COD) / biological oxygen demand (BOD) / suspended matter / oil / sulfide / phenol is higher than the standard threshold, the lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand (COD) / biological oxygen demand (BOD) / suspended matter / oil / sulfide / phenol is used as a pollutant constituent of the drilling fluid waste.

[0071] when the constituent concentration of lead is higher than 10 mg / L, the lead is used as a pollutant constituent of the drilling fluid waste;

[0072] when the constituent concentration of cadmium is higher than 0.1 mg / L, the constituent of cadmium is used as a pollutant constituent of the drilling fluid waste;

[0073] when the constituent concentration of mercury is higher than 0.05 mg / L, the constituent of mercury is used as a pollutant constituent of the drilling fluid waste;

[0074] when the constituent concentration of arsenic is higher than 0.5 mg / L, the constituent concentration of arsenic is used as a pollutant constituent of the drilling fluid waste;

[0075] when the constituent concentration of total chromium is higher than 1.5 mg / L, the constituent concentration of total chromium is used as a pollutant constituent of the drilling fluid waste;

[0076] when the constituent concentration of hexavalent chromium is higher than 0.5 mg / L, the constituent concentration of hexavalent chromium is used as a pollutant constituent of the drilling fluid waste;

[0077] when the constituent concentration of fluoride is higher than 10 mg / L, the constituent concentration of fluoride is used as a pollutant constituent of the drilling fluid waste;

[0078] when the constituent concentration of chemical oxygen demand is higher than 150 mg / L, the constituent concentration of chemical oxygen demand is used as a pollutant constituent of the drilling fluid waste;

[0079] when the constituent concentration of biological oxygen demand is higher than 30 mg / L, the constituent concentration of biological oxygen demand is used as a pollutant constituent of the drilling fluid waste;

[0080] when the constituent concentration of suspended matter is higher than 150 mg / L, the constituent concentration of suspended matter is used as a pollutant constituent of the drilling fluid waste;

[0081] When the concentration of the color component is higher than 80 mg / L, the concentration of the color component is taken as the pollutant component of the drilling fluid waste;

[0082] When the concentration of the PH value component is higher than 9 mg / L or lower than 6 mg / L, the concentration of the PH value component is taken as the pollutant component of the drilling fluid waste;

[0083] When the concentration of the oil component is higher than 10 mg / L, the concentration of the oil component is taken as the pollutant component of the drilling fluid waste;

[0084] When the concentration of the sulfide / phenolic component is higher than 7 mg / L, the concentration of the sulfide / phenolic component is taken as the pollutant component of the drilling fluid waste.

[0085] In the selected drilling scenarios and drilling fluid waste, all pollutant component detection methods are applied for full coverage monitoring, so as to identify various waste components in the drilling waste liquid waste under each drilling scenario, and based on the waste component concentration, the pollutant components that do not meet the discharge standard and need to be treated are determined, so as to provide training data for subsequent waste composition determination model and pollutant composition determination model.

[0086] Step S2, drilling scene features and drilling fluid category features are extracted from the drilling scene and the drilling fluid waste, and a waste composition determination model is constructed by using the drilling scene features and the drilling fluid category features and the waste component composition, and a pollutant composition determination model is constructed by using the drilling scene features, the drilling fluid category features and the waste component composition and the pollutant component composition;

[0087] The drilling scene features and the drilling fluid category features are extracted from the drilling scene and the drilling fluid waste, including:

[0088] The drilling scene features are extracted in each drilling scene, and the drilling fluid category features are extracted in the drilling fluid waste in each drilling scene;

[0089] The drilling scene features include drilling geological features and drilling structural features.

[0090] The drilling fluid category features include the component composition of the drilling fluid and the component concentration of the drilling fluid component composition.

[0091] The waste composition determination model is constructed by using the drilling scene features and the drilling fluid category features and the waste component composition, including:

[0092] The drilling scene features and the drilling fluid category features are taken as input items of the BP neural network, the waste component composition is taken as an output item of the BP neural network, and the input items and the output item are trained by using the BP neural network to obtain the waste component determination model;

[0093] The model expression of the waste component determination model is:

[0094] [E] m = BP ([Scene_Features] k , [Element_Features] r ) ;

[0095] In the formula, [E] m is a waste component composition sequence composed of m waste components, [Scene_Features] k is a drilling scene feature vector composed of k drilling scene feature components, [Element_Features] r is a drilling fluid category feature vector composed of r drilling fluid category feature components, BP is a BP neural network, and m, k and r are respectively total amounts of waste components, drilling scene feature components and drilling fluid category feature components.

[0096] A pollutant component determination model is constructed by using the drilling scene features, the drilling fluid category features, the waste component composition and the pollutant component composition, and the pollutant component determination model comprises the following steps:

[0097] The drilling scene features, the drilling fluid category features and the waste component composition are taken as second input items of the BP neural network, the pollutant component composition is taken as a second output item of the BP neural network, and the second input items and the second output item are trained by using the BP neural network to obtain the waste component determination model;

[0098] The model expression of the pollutant component determination model is:

[0099] [W] n = BP ([Scene_Features] k , [Element_Features] r , [E] m ) ;

[0100] In the formula, [W] n is a waste component composition sequence composed of n pollutant components, [Scene_Features] k is a drilling scene feature vector composed of k drilling scene feature components, [Element_Features] ra drilling fluid category feature vector composed of r drilling fluid category feature components, [E] m a waste component composition sequence composed of m waste components, BP is a BP neural network, n, m, k, and r are respectively the total amount of pollutant components, the total amount of waste components, the total amount of drilling scene feature components, and the total amount of drilling fluid category feature components.

[0101] The waste composition determination model and the pollutant composition determination model are constructed to represent the mapping relationship between the drilling scene feature and the drilling fluid category feature and the waste component composition, and the mapping relationship between the drilling scene feature, the drilling fluid category feature, the waste component composition, and the pollutant component composition, and then the waste component is obtained only according to the drilling scene feature and the drilling fluid category feature, and the pollutant component is obtained according to the drilling scene feature, the drilling fluid category feature, and the waste component composition, so that the detection range of the drilling fluid waste is reduced from the full component range to the pollutant component range, that is, the full component range is the component detection of lead, cadmium, mercury, arsenic, total chromium, hexavalent chromium, fluoride, chemical oxygen demand (COD), biological oxygen demand (BOD), suspended solids, petroleum, sulfide, and phenol, and the pollutant component is lead, cadmium, chemical oxygen demand (COD), petroleum, and sulfide, and only the component detection of lead, cadmium, chemical oxygen demand (COD), petroleum, and sulfide is needed, and the determination of the pollutant component determines the detection method used for detecting the pollutant component, and the detection direction of the drilling fluid waste is determined, that is, the component detection method of lead, cadmium, chemical oxygen demand (COD), petroleum, and sulfide is used, and in summary, the above model can realize directional compression of drilling fluid waste detection, and achieve efficiency optimization of component detection.

[0102] In step S3, the waste composition determination model is applied to the target drilling to determine the waste component composition of the drilling fluid waste in the target drilling as the waste component target composition, and the pollutant composition determination model is applied to the target drilling to determine the pollutant component composition of the drilling fluid waste in the target drilling as the pollutant component target composition, and the pollutant component target detection method is screened from the pollutant component detection method set according to the pollutant component target composition of the target drilling, and the component concentration of the pollutant component target composition is obtained by using the pollutant component target detection method to correspondingly detect the pollutant component target composition, so as to realize directional detection of the drilling fluid waste composition and achieve efficiency optimization.

[0103] The waste composition determination model is applied to the target drilling to determine the waste component composition of the drilling fluid waste in the target drilling as the waste component target composition, including:

[0104] The drilling scene feature of the target drilling well and the drilling fluid category feature of the drilling fluid used by the target drilling well are input into the waste composition determination model, and the waste composition of the drilling fluid waste in the target drilling well is output by the waste composition determination model.

[0105] The pollutant composition determination model is applied to the target drilling well to determine the pollutant composition of the drilling fluid waste in the target drilling well as the pollutant composition target composition, which comprises:

[0106] The drilling scene feature of the target drilling well, the drilling fluid category feature of the drilling fluid used by the target drilling well and the waste composition target composition are input into the pollutant composition determination model, and the pollutant composition of the drilling fluid waste in the target drilling well is output by the pollutant composition determination model.

[0107] The pollutant composition target detection method is screened out from the pollutant composition detection method set according to the pollutant composition target composition of the target drilling well, which comprises:

[0108] The pollutant composition target detection method for detecting each pollutant composition is matched from the pollutant composition detection method set according to the pollutant composition in the pollutant composition target composition;

[0109] The pollutant composition and the pollutant composition target detection method are one-to-one corresponding.

[0110] The drilling scene feature and the drilling fluid category feature are respectively subjected to feature normalization processing before model operation.

[0111] The drilling scene feature and the drilling fluid category feature are extracted from the drilling scene and the drilling fluid waste, and the waste composition determination model is constructed by using the drilling scene feature, the drilling fluid category feature and the waste composition, the pollutant composition determination model is constructed by using the drilling scene feature, the drilling fluid category feature, the waste composition and the pollutant composition, the waste composition determination model is applied to the target drilling well to determine the waste composition of the drilling fluid waste in the target drilling well as the waste composition target composition, and the pollutant composition determination model is applied to the target drilling well to determine the pollutant composition of the drilling fluid waste in the target drilling well as the pollutant composition target composition, which provides the detection direction for the pollutant composition detection, avoids the redundant composition detection method for the pollutant composition which does not exist in the drilling waste liquid, and realizes the directional detection of the drilling fluid waste composition to achieve the efficiency optimization.

[0112] The above examples are only exemplary embodiments of the present application, and are not used to limit the present application, and the protection scope of the present application is defined by the claims. Those skilled in the art can make various modifications or equivalent replacements to the present application within the spirit and protection scope of the present application, and such modifications or equivalent replacements are also regarded as falling within the protection scope of the present application.

Claims

1. A method for drilling fluid waste composition classification optimization based on drilling scenarios, the method comprising: Comprise the following steps: Step S1, obtain a plurality of drilling fluid waste under drilling scene, and use the pollutant component detection method set to detect the drilling fluid waste without distinction to determine the waste component composition and the pollutant component composition in the drilling fluid waste, the pollutant component detection method set contains all pollutant component detection methods that can be used for drilling fluid waste pollution detection; Step S2, extract drilling scene features and drilling fluid category features in the drilling scene and the drilling fluid waste, and use the drilling scene features and the drilling fluid category features and the waste component composition to construct the waste component determination model, and use the drilling scene features, the drilling fluid category features and the waste component composition and the pollutant component composition to construct the pollutant component determination model; Step S3, the waste component determination model is applied to the target drilling to determine the waste component composition of the drilling fluid waste in the target drilling as the waste component target composition, and the pollutant component determination model is applied to the target drilling to determine the pollutant component composition of the drilling fluid waste in the target drilling as the pollutant component target composition, and the pollutant component target detection method is screened out from the pollutant component detection method set according to the pollutant component target composition of the target drilling, the pollutant component target detection method is used to detect the pollutant component target composition to obtain the component concentration of the pollutant component target composition of the target drilling, so as to realize the directional detection of the drilling fluid waste composition and achieve the efficiency optimization.

2. The method of claim 1, wherein: The pollutant component detection method set comprises spectrophotometry, atomic fluorescence spectrophotometry, diphenyl carbonyl dihydrazine colorimetry, fluorine reagent colorimetry, dichromate method, dilution and inoculation method, weight method, dilution method, glass click method, petroleum ether extraction weight method, luminescent bacteria method; The spectrophotometry is used for component detection of lead and cadmium in waste components; The atomic fluorescence spectrophotometry is used for component detection of mercury and arsenic in waste components; The diphenyl carbonyl dihydrazine colorimetry is used for component detection of total chromium and hexavalent chromium in waste components; The fluorine reagent colorimetry is used for component detection of fluoride in waste components; The dichromate method is used for component detection of chemical oxygen demand in waste components; The dilution and inoculation method is used for component detection of biological oxygen demand in waste components; The glass click method is used for component detection of suspended solids in waste components; The petroleum ether extraction weight method is used for component detection of petroleum in waste components; The luminescent bacteria method is used for component detection of biological toxicity in waste components; The methylene blue colorimetry is used for component detection of sulfide in waste components; The aminoantipyrine photometry is used for component detection of phenol in waste components.

3. The method of claim 2, wherein: The use of the pollutant component detection method set to detect the drilling fluid waste without distinction to determine the waste component composition in the drilling fluid waste comprises: The component detection methods in the pollution component detection method set are used to detect the components of lead, cadmium, mercury, arsenic, total chromium, hexavalent chromium, fluoride, chemical oxygen demand, biological oxygen demand, suspended solids, petroleum components, sulfide components, and phenols in sequence by using spectrophotometry, atomic fluorescence spectrophotometry, diphenyl carbonyl dihydrazine colorimetry, fluorine reagent colorimetry, dichromate method, dilution and inoculation method, gravimetric method, dilution method, glass tapping method, petroleum ether extraction gravimetric method, and luminescent bacteria method. When the component concentration of lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand / biological oxygen demand / suspended solids / petroleum / sulfide / phenol is higher than 0, the lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand / biological oxygen demand / suspended solids / petroleum / sulfide / phenol is used as the waste component of the drilling fluid waste. When the component concentration of lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand / biological oxygen demand / suspended solids / petroleum / sulfide / phenol is higher than the standard threshold value, the lead / cadmium / mercury / arsenic / total chromium / hexavalent chromium / fluoride / chemical oxygen demand / biological oxygen demand / suspended solids / petroleum / sulfide / phenol is used as the pollution component of the drilling fluid waste.

4. The method of claim 3, wherein: The drilling scene features and the drilling fluid category features are extracted from the drilling scene and the drilling fluid waste, and the drilling scene features and the drilling fluid category features are used to construct a waste component determination model. The drilling scene features and the drilling fluid category features are extracted from each drilling scene and each drilling fluid waste in the drilling scene. The drilling scene features include drilling geological features and drilling structural features. The drilling fluid category features include component compositions of the drilling fluid and component concentrations of the drilling fluid component compositions.

5. A method of drilling fluid waste composition classification optimization based on drilling scenario as claimed in claim 4, wherein: The waste component determination model is constructed by using the drilling scene features, the drilling fluid category features, and waste component compositions. The drilling scene features and the drilling fluid category features are used as input items of a BP neural network, the waste component compositions are used as output items of the BP neural network, and the waste component determination model is obtained by network training of the input items and the output items by using the BP neural network. The model expression of the waste component determination model is as follows. [E] m = BP([Scene_Features] k , [Element_Features] r ); In the formula, [E] m is a waste component composition sequence composed of m waste components, [Scene_Features] k is a drilling scene feature vector composed of k drilling scene feature components, [Element_Features] r is a drilling fluid category feature vector composed of r drilling fluid category feature components, BP is a BP neural network, and m, k, and r are respectively the total amount of waste components, the total amount of drilling scene feature components, and the total amount of drilling fluid category feature components.

6. The method of claim 5, wherein: The pollution component determination model is constructed by using the drilling scene features, the drilling fluid category features, the waste component compositions, and pollution component compositions. The drilling scene features, the drilling fluid category features, and the waste component compositions are used as second input items of a BP neural network, the pollution component compositions are used as second output items of the BP neural network, and the pollution component determination model is obtained by network training of the second input items and the second output items by using the BP neural network. The model expression of the pollution component determination model is as follows. [W] n = BP([Scene_Features] k , [Element_Features] r , [E] m ); In the formula, [W] n is a waste component composition sequence composed of n pollutant components, [Scene_Features] k is a drilling scene feature vector composed of k drilling scene feature components, [Element_Features] r is a drilling fluid category feature vector composed of r drilling fluid category feature components, [E] m is a waste component composition sequence composed of m waste components, BP is a BP neural network, and n, m, k, and r are respectively the total amount of pollutant components, the total amount of waste components, the total amount of drilling scene feature components, and the total amount of drilling fluid category feature components.

7. The method of claim 6, wherein: The waste component determination model is applied to a target drilling to determine the waste component compositions of the drilling fluid waste in the target drilling as waste component target compositions, and the waste component target compositions are used as the waste component target compositions of the drilling fluid waste in the target drilling. The drilling scene features of the target drilling and the drilling fluid category features of the drilling fluid used in the target drilling are input into the waste component determination model, and the waste component determination model outputs the waste component compositions of the drilling fluid waste in the target drilling.

8. The method of claim 7, wherein: The application of the pollutant composition determination model to the target drilling well determines a pollutant composition of the drilling fluid waste in the target drilling well as a pollutant composition target, and the pollutant composition target includes: The drilling scene feature of the target drilling well, the drilling fluid category feature of the drilling fluid used by the target drilling well, and the waste composition target are input into the pollutant composition determination model, and the pollutant composition of the drilling fluid waste in the target drilling well is output by the pollutant composition determination model.

9. The method of claim 8, wherein, The pollutant composition target is screened out from the pollutant composition detection method set according to the pollutant composition target of the target drilling well, and the pollutant composition target includes: According to the pollutant composition in the pollutant composition target, a pollutant composition target detection method for detecting each pollutant composition is matched in the pollutant composition detection method set; Wherein, the pollutant composition and the pollutant composition target detection method are one-to-one corresponding.

10. The method of claim 9, wherein, The drilling scene feature and the drilling fluid category feature are respectively subjected to feature normalization processing before model operation.

Images