A method, system and storage medium for zero-discharge treatment of produced water from oil and gas fields

By preprocessing and clustering the historical data of the produced water in oil and gas fields, establishing an association model and using neural networks to adjust the equipment parameters, the problem of different processing effects in different regions is solved, and zero emissions and working convenience of the produced water are achieved.

CN118445649BActive Publication Date: 2024-11-19SICHUAN XINGAO ENVIRONMENTAL TECH SERVICE CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410616187.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-17
Publication Date
2024-11-19
Estimated Expiration
2044-05-17

AI Technical Summary

Technical Problem

The parameters of the existing zero-emission treatment methods for producing water from oil and gas fields have a large range of adjustments, resulting in different treatment effects in different regions and oil field locations, increasing the workload of personnel and prone to errors.

Method used

By collecting historical data for preprocessing and clustering, an association model between the derived water, equipment and discharge water parameters is established, and the neural network model is used to automatically adjust the device parameters to achieve the best settings.

Benefits of technology

It realizes automatic adjustment of equipment parameters according to the actual situation of the produced water, reduces the work burden of relevant personnel, and ensures the zero-emission treatment effect of the produced water.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118445649B_ABST
    Figure CN118445649B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, system and storage medium for zero-discharge treatment of produced water in oil and gas fields, belonging to the field of sewage treatment, and the method comprises: pre-processing and integrating historical data to obtain the existence duration of a first parameter sequence, a second parameter sequence and a third parameter sequence; clustering the first parameter sequence to obtain a standard clustering result, analyzing the standard clustering result to obtain the first data; locating the existence duration corresponding to the first data, obtaining the corresponding second parameter sequence and third parameter sequence, which are defined as the second data and the third data; analyzing the first data, the second data and the third data to obtain an association model between produced water parameters, equipment parameters and discharged water parameters; inputting the currently measured produced water parameters and the set discharged water parameters into the association model to obtain the current optimal setting parameters of the water purification equipment. The present invention can automatically adjust the equipment parameters according to the actual situation of the produced water, so that various indicators in the discharged water meet the standards.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of sewage treatment, and in particular relates to a zero-discharge treatment method, system and storage medium for oil and gas field produced water. Background Art

[0002] With the increase of oil and gas field production activities, the treatment of produced water has become increasingly prominent. Developing an effective zero-discharge treatment method for oil and gas field produced water has become an urgent need in the industry.

[0003] To this end, the following methods have been proposed in the prior art to achieve the above-mentioned purpose. For example, Chinese patent document CN107311375A discloses a zero-emission treatment method for low-concentration oil and gas field produced water. The method adopts comprehensive treatment of electrocoagulation, ordinary filtration, electrodialysis and evaporation system to improve the cleanliness of produced water. In the electrodialysis treatment, the first and second stages are circulated separately to improve the purification efficiency while reducing emissions. At the same time, the evaporation system treats the mother liquor, fresh water and solid salt separately to improve the utilization of water and salt.

[0004] The above scheme has a clear treatment process and guiding parameters for the treatment of produced water, but the parameters have a large adjustment range. For example, current between 500 and 2000A can be treated. However, different parameters may have different treatment effects due to the different oil field locations, reinjection water quality, geological conditions and oil production methods in different regions. Therefore, specific parameters are usually determined by experienced personnel, but frequent adjustment of equipment parameters will greatly increase the workload of relevant personnel, and adjustments made incorrectly will also affect the treatment effect of produced water. Summary of the invention

[0005] To solve the above problems, the present invention provides a zero-emission treatment method, system and storage medium for oil and gas field produced water. Through the technical solution of the present invention, the equipment parameters can be automatically adjusted according to the actual situation of the produced water, which provides convenience for the work of relevant personnel and ensures the treatment effect of the produced water.

[0006] In order to achieve the above-mentioned purpose of the invention, the present invention proposes a zero-discharge treatment method for produced water in oil and gas fields, comprising:

[0007] Collecting historical data, the historical data including a first parameter sequence, a second parameter sequence and a third parameter sequence measured at multiple time points, the first parameter sequence, the second parameter sequence and the third parameter sequence respectively corresponding to produced water parameters, equipment parameters and discharged water parameters, and preprocessing the historical data;

[0008] Integrate the preprocessed historical data to obtain the existence duration of each of the first parameter sequence, the second parameter sequence, and the third parameter sequence;

[0009] Clustering the first parameter sequence to obtain a standard clustering result, analyzing the standard clustering result, and obtaining first data from the discharged water parameter;

[0010] Locate the existence duration corresponding to the first data, and obtain the second parameter sequence and the third parameter sequence corresponding to the existence duration, which are defined as second data and third data respectively;

[0011] analyzing the first data, the second data, and the third data to obtain a correlation model among the produced water parameter, the equipment parameter, and the discharged water parameter;

[0012] The currently measured produced water parameters and the set discharged water parameters are input into the correlation model to obtain the current optimal setting parameters of the water purification equipment.

[0013] Further, clustering the first parameter sequence comprises the following steps:

[0014] Setting an initial number of clusters, setting a final number of clusters based on the number of the discharged water parameters in the first parameter sequence, clustering all the first parameter sequences starting from the initial number of clusters to obtain a first result, including clusters of the initial number of clusters, each of which includes a plurality of the first parameter sequences;

[0015] The first parameter sequence in the same cluster is defined as a same-cluster sequence, a comparison threshold is set for each type of the discharged water parameter, the absolute value of the difference between the same type of the discharged water parameters of two same-cluster sequences is compared, the discharged water parameter whose absolute value is greater than the comparison threshold is defined as a target parameter, the first number of the target parameters in the two same-cluster sequences is counted, this step is repeated to complete the comparison between the two same-cluster sequences, an average value of the first number is calculated, and if the average value is greater than the first threshold, the cluster is defined as a defect target;

[0016] On the basis of the initial number of clusters, the number of clusters is increased to re-cluster the first parameter sequence to obtain a second result, and this step is repeated until the final number of clusters is reached, and the result with the smallest number of defect targets and less than or equal to a second threshold is selected as the standard clustering result.

[0017] Furthermore, if after reaching the final number of clusters, there is no result in which the defect target number is less than or equal to the second threshold, the value of the first threshold is increased, and the first parameter sequence is clustered again starting with the initial number of clusters.

[0018] Further, obtaining the first data comprises the following steps:

[0019] In the standard clustering result, if the first number of the target parameters between a plurality of the same cluster sequences is less than the first threshold, two of the same cluster sequences are defined as the first data.

[0020] Further, analyzing the first data, the second data and the third data comprises the following steps:

[0021] Acquire the second data corresponding to the first data, and if a first influencing parameter exists in the plurality of second data, define the second data containing the first influencing parameter as transit data, wherein the first influencing parameter is the device parameter in the plurality of second data, and a difference between the device parameters of the same type is greater than a third threshold;

[0022] Obtain the third data corresponding to the transit data, obtain the second influencing parameter in the third data, and establish a neural network model according to the target parameter, the first influencing parameter and the second influencing parameter to analyze the first data, the second data and the third data.

[0023] Furthermore, establishing a neural network model includes the following steps:

[0024] The target parameter, the first influencing parameter and the second influencing parameter form a training set, the number of input neurons in the neural network model is the same as the sum of the first influencing parameter and the second influencing parameter, the number of output neurons is the same as the number of the target parameters, and the training set is used to train and generate a neural network model.

[0025] Furthermore, obtaining the optimal setting parameters comprises the following steps:

[0026] The association model includes a preprocessing model and a selection model. After receiving the produced water parameter and the discharged water parameter that are actually input, the preprocessing model integrates the two into a first actual sequence and a second actual sequence respectively, and inputs the first actual sequence and the second actual sequence into the selection model;

[0027] The selection model is based on an existing neural network model, processes the first actual sequence and the second actual sequence into multiple subsequences, and inputs the subsequences into the neural network model of corresponding dimensions to obtain the optimal setting parameters.

[0028] Furthermore, preprocessing the historical data includes the following steps:

[0029] After acquiring the historical data, the produced water parameters, the equipment parameters and the discharged water parameters are standardized to obtain standardized data, water quality emission indicators are set, and the standardized data and the water quality emission indicators are compared to delete the discharged water parameters that do not meet the water quality emission indicators.

[0030] The present invention also provides a zero-discharge treatment system for oil and gas field produced water, which is used to implement the above-mentioned zero-discharge treatment method for oil and gas field produced water, and the system comprises:

[0031] A collection module, used to collect historical data, the historical data including a first parameter sequence, a second parameter sequence and a third parameter sequence measured at multiple time points, the first parameter sequence, the second parameter sequence and the third parameter sequence respectively corresponding to produced water parameters, equipment parameters and discharged water parameters, preprocessing the historical data, integrating the preprocessed historical data, and obtaining the existence time of each of the first parameter sequence, the second parameter sequence and the third parameter sequence;

[0032] A clustering module, clustering the first parameter sequence to obtain a standard clustering result, analyzing the standard clustering result, and obtaining first data from the discharged water parameter;

[0033] an analysis module, locating the existence duration corresponding to the first data, obtaining the second parameter sequence and the third parameter sequence corresponding to the existence duration, defining them as second data and third data respectively, and analyzing the first data, the second data and the third data to obtain a correlation model between the produced water parameter, the equipment parameter and the discharged water parameter;

[0034] The input module inputs the currently measured produced water parameters and the set discharged water parameters into the association model to obtain the current optimal setting parameters of the water purification equipment.

[0035] The present invention also discloses a computer storage medium, wherein the computer storage medium stores program instructions, wherein when the program instructions are executed, the device where the computer storage medium is located is controlled to execute the above-mentioned method.

[0036] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0037] The present invention first obtains a large amount of historical produced water treatment data, and then pre-processes and integrates these data to obtain standardized data; then, time labels are assigned to produced water parameters, equipment parameters and discharge water parameters according to the time point of collection, that is, the existence duration; after completing the above steps, the discharge water data is first clustered to obtain the time period when the discharge water parameters change, and according to the produced water parameters and equipment parameters corresponding to the time period and time labels, a neural network model is established to analyze the three parameters, and the change relationship between the changed discharge water parameters and the other two parameters can be obtained. In this way, when relevant personnel determine the equipment parameters, they only need to input the currently measured produced water parameters and the set emission indicators into the system, and the optimal equipment parameters can be automatically calculated through the change relationship, which not only ensures zero discharge of produced water, but also reduces the workload of relevant personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a flow chart of the steps of a zero-discharge treatment method for produced water from an oil and gas field according to the present invention;

[0039] Figure 2 The present invention is a structural schematic diagram of a zero-discharge treatment system for produced water from an oil and gas field. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0041] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0042] like Figure 1 As shown, a zero-discharge treatment method for produced water from an oil and gas field comprises:

[0043] Step S1: Collect historical data, which include first parameter sequence, second parameter sequence and third parameter sequence measured at multiple time points, the first parameter sequence, second parameter sequence and third parameter sequence correspond to produced water parameters, equipment parameters and discharged water parameters respectively, and pre-process the historical data.

[0044] Specifically, historical data includes data collected in different regions and locations, including produced water parameters obtained by various oil production methods. The first parameter sequence includes suspended matter concentration, particulate matter concentration, etc. Similarly, the second parameter sequence includes sedimentation time, current and voltage of the electrolysis device, addition ratio of hydrogen peroxide and sodium hydroxide, operating pressure of high-pressure and low-pressure reverse osmosis devices, and capacity of treatment pools at all levels. The third parameter sequence includes suspended matter concentration, organic matter content, pH value, COD concentration, ammonia nitrogen concentration, heavy metals and other concentrations of various harmful substances.

[0045] The preprocessing of historical data includes data standardization and filtering. Standardization is to integrate the obtained data into the same unit and order of magnitude, which is convenient for subsequent clustering. Filtering is to filter out the discharge water parameters that do not meet the discharge standards. Such parameters have no reference significance and are therefore deleted.

[0046] Step S2: Integrate the preprocessed historical data to obtain the existence duration of each first parameter sequence, second parameter sequence and third parameter sequence.

[0047] For example, for the same device in the same region, the first parameter sequence 1 was measured on March 1, the first parameter sequence 2 was measured on March 15, and the first parameter sequence 3 was measured on March 31. Then the existence duration of the first parameter sequence 1 is from March 1 to 14, the existence duration of the first parameter sequence 2 is from March 15 to 30, and the existence duration of the first parameter sequence 3 is from March 31 to the present. The second and third parameter sequences are obtained in the same way and will not be introduced here.

[0048] Step S3: clustering the first parameter sequence to obtain a standard clustering result, analyzing the standard clustering result, and obtaining first data from the discharged water parameters.

[0049] For example, the first parameter sequence has 30 types of discharge water parameters. Through clustering, similar first parameter sequences are clustered into one category. For example, a standard clustering result is obtained through clustering, in which there are 10 clusters. In cluster 1, there are 42 first parameter sequences. Among these 42 first parameter sequences, only discharge water parameters 15-18 are different from each other. These 42 first parameter sequences are set as first data. The specific method of obtaining the first data will be introduced in detail later.

[0050] Step S4: locate the existence duration corresponding to the first data, obtain the second parameter sequence and the third parameter sequence corresponding to the existence duration, and define them as the second data and the third data respectively.

[0051] Based on the introduction of step S2, each of the first parameter sequence, the second parameter sequence and the third parameter sequence corresponds to an existence duration, so first obtain the existence duration of the first data, such as the existence duration of the first data 1 is March 1-14, and then continue to obtain the second parameter sequence and the third parameter sequence under March 1-14; in particular, if the existence duration of the first parameter sequence 1 is March 1-5, and the existence duration of the first parameter sequence 2 is March 6 to March 14, then split the first data 1 into March 1-5 and March 6 to March 14, and the first parameter sequence 1 and the first parameter sequence 2 correspond to the split first data respectively, which can facilitate the subsequent training of the neural network model.

[0052] Step S5: Analyze the first data, the second data, and the third data to obtain a correlation model between produced water parameters, equipment parameters, and discharged water parameters.

[0053] In this embodiment, since various indicators among produced water parameters, equipment parameters and discharge water parameters are usually nonlinearly related, and a single equipment parameter may also affect multiple discharge water parameters, such as removing hypochlorite ions by hydrogen peroxide, but this will increase the concentration of hydrogen peroxide in the water, and sodium oxide will be needed to remove excess hydrogen peroxide later, and neural network technology is very suitable for processing nonlinear data, the neural network model is trained by the first data, the second data and the third data, and the obtained association model can be used to set the equipment parameters.

[0054] Step S6: Input the currently measured produced water parameters and the set discharged water parameters into the correlation model to obtain the current optimal setting parameters of the water purification equipment.

[0055] After the association model is established, when determining the equipment parameters, relevant personnel only need to input the currently measured produced water parameters and the set emission indicators into the system to obtain the optimal equipment parameters. In this way, even less experienced personnel can set reasonable parameters so that the various indicators in the discharged water meet the standards, thereby achieving zero emission treatment of produced water.

[0056] The present invention first obtains a large amount of historical produced water treatment data, and then pre-processes and integrates these data to obtain standardized data; then, time labels are assigned to produced water parameters, equipment parameters and discharge water parameters according to the time point of collection, that is, the existence duration; after completing the above steps, the discharge water data is first clustered to obtain the time period when the discharge water parameters change, and according to the produced water parameters and equipment parameters corresponding to the time period and time labels, a neural network model is established to analyze the three parameters, and the change relationship between the changed discharge water parameters and the other two parameters can be obtained. In this way, when relevant personnel determine the equipment parameters, they only need to input the currently measured produced water parameters and the set emission indicators into the system, and the optimal equipment parameters can be automatically calculated through the change relationship, which not only ensures zero discharge of produced water, but also reduces the workload of relevant personnel.

[0057] It is particularly noteworthy that, through the technical solution of the present invention, equipment parameters can be automatically adjusted according to the actual situation of produced water, providing convenience for the work of relevant personnel, thereby significantly reducing the workload.

[0058] In this embodiment, clustering the first parameter sequence includes the following steps:

[0059] An initial number of clusters is set, and a final number of clusters is set based on the number of discharged water parameters in the first parameter sequence. All first parameter sequences are clustered starting with the initial number of clusters to obtain a first result, which includes the initial number of clusters, and each cluster includes multiple first parameter sequences.

[0060] Specifically, the initial number of clusters is set to 5. In this embodiment, 42 types of drainage water parameters are set, so the generated first parameter sequence is drainage water parameter 1 ~ drainage water parameter 2 ~ ... ~ drainage water parameter 42. Here, the final number of clusters is set to 2 / 3 of the number of drainage water parameters, that is, the final number of clusters is 28. In this embodiment, the K-means algorithm is used for clustering. First, the number of clusters is set to 5. If there are 500 first parameter sequences, then there will be 5 clusters in the first result obtained after clustering, and each cluster will include multiple first parameter sequences. The specific process of the clustering algorithm is a prior art and will not be introduced here.

[0061] The first parameter sequence in the same cluster is defined as a same-cluster sequence, a comparison threshold is set for each type of discharged water parameter, the absolute value of the difference between the same type of discharged water parameters of two same-cluster sequences is compared, and the discharged water parameter whose absolute value is greater than the comparison threshold is defined as the target parameter. The first quantity of the target parameters in the two same-cluster sequences is counted, and this step is repeated to complete the comparison between the two same-cluster sequences, and the average value of the first quantity obtained is calculated. If the average value is greater than the first threshold, the cluster is defined as a defect target.

[0062] Here, the first parameter sequence 1 and the first parameter sequence 2 are located in cluster 1, so the first parameter sequence 1 and the first parameter sequence 2 are set as the same cluster sequence. Before that, a comparison threshold is set for each type of discharge water parameter, such as setting a judgment threshold of appropriate size for suspended solids concentration, organic matter content, pH value, etc. Then, the discharge water parameter of type 1 of the first parameter sequence 1 is subtracted from the discharge water parameter of type 1 of the first parameter sequence 2 and the absolute value is obtained. If the absolute value is greater than the judgment threshold, it indicates that the numerical difference between the two discharge water parameters is large, and this parameter is set as the target parameter. When there is a target parameter greater than the first threshold between the first parameter sequence 1 and the first parameter sequence 2, the first threshold is set to 3, which indicates that the similarity between the two parameter sequences is low, that is, there are many different parameters.

[0063] Then compare the first parameter sequences 1 and 3, the first parameter sequences 2 and 3, and so on, until the pairwise comparisons of the parameter sequences in the same cluster are completed. Finally, calculate and obtain the average values ​​of all the first quantities. When the average value is greater than the first threshold, it indicates that the clustering effect of the cluster is poor, and the cluster is located as a defect target because a large number of dissimilar parameter sequences are classified into one category.

[0064] On the basis of the initial number of clusters, the number of clusters is increased to re-cluster the first parameter sequence to obtain a second result, and this step is repeated until the final number of clusters is reached, and the result with the smallest number of defect targets and less than or equal to the second threshold is selected as the standard clustering result.

[0065] If the number of defective targets in the first result is greater than the second threshold, it indicates that the clustering result is poor, and it also indicates that setting the clustering target to 5 cannot distinguish the current first parameter sequence well. Then the number of clusters is increased, such as increasing by 1 each time. In the next time, the first parameter sequence is re-clustered with the cluster number 6 as the target to obtain the second result, and then the number of defective targets in the second result is counted again. That is, in this embodiment, the number of defective targets is used as the evaluation criterion for evaluating the clustering results. When the number of defects is less than or equal to the second threshold, it indicates that the clustering effect is good this time. The result with the smallest number of defective targets is used as the standard clustering result, and the clusters included therein are used for subsequent data analysis.

[0066] In this embodiment, if after reaching the final number of clusters, there is no result where the number of defective targets is less than or equal to the second threshold, the value of the first threshold is increased, and the first parameter sequence is clustered again starting with the initial number of clusters.

[0067] After clustering with 28 as the clustering target, the standard clustering result cannot be obtained, indicating that the first threshold is set too low. At this time, the first threshold is increased. For example, when there are 3 target parameters between the two first parameter sequences, the similarity between the two is considered to be low. After the increase, when there are 6 target parameters between the two first parameter sequences, the similarity between the two is considered to be low. That is, the evaluation criteria for clustering results are relaxed.

[0068] In this embodiment, obtaining the first data includes the following steps:

[0069] In the standard clustering result, if the first number of target parameters between a plurality of sequences in the same cluster is less than a first threshold, both of the two sequences in the same cluster are defined as first data.

[0070] After obtaining the standard clustering results, if the first parameter sequence 1, the first parameter sequence 2 and the first parameter sequence 3 belong to cluster 1, and the first parameter sequence 1, the first parameter sequence 2 and the first parameter sequence 3 only differ in the discharge water parameters 15-18, which are less than the first threshold, then the first parameter sequence 1, the first parameter sequence 2 and the first parameter sequence 3 are set as the first data.

[0071] In this embodiment, analyzing the first data, the second data, and the third data includes the following steps:

[0072] Obtain second data corresponding to the first data. If a first influencing parameter exists in the plurality of second data, define the second data with the first influencing parameter as transit data, wherein the first influencing parameter is a device parameter in the plurality of second data, and a difference between the parameters of the same type of devices is greater than a third threshold.

[0073] As before, the first parameter sequence 1, the first parameter sequence 2 and the first parameter sequence 3 are the first data, and the corresponding existence durations are March 1-14, March 15-30, and March 31 to the present, then the corresponding second data are the second parameter sequence 1 with an existence duration of March 1-14, the second parameter sequence 2 with an existence duration of March 15-30, and the second parameter sequence 3 with an existence duration of March 31 to the present; the third threshold is the same as the comparison threshold, and both are determined according to different device parameters; if the difference between device parameter 1 in the second parameter sequence 1 and the second parameter sequence 2 is greater than the third threshold, the difference between device parameters 2 is greater than the third threshold, the difference between device parameter 1 in the second parameter sequence 2 and the second parameter sequence 3 is greater than the third threshold, the difference between device parameters 2 is greater than the third threshold, the difference between device parameter 1 in the second parameter sequence 1 and the second parameter sequence 3 is greater than the third threshold, and the difference between device parameters 2 is greater than the third threshold; then device parameter 1 and device parameter 2 are the first influencing parameters, and the second parameter sequences 1-3 are set as transit data.

[0074] The third data corresponding to the transit data is obtained, the second influencing parameter existing in the third data is obtained, and a neural network model is established according to the target parameter, the first influencing parameter and the second influencing parameter to analyze the first data, the second data and the third data.

[0075] The principle of obtaining the second influencing parameter in the third data is the same as the principle of obtaining the first influencing parameter in the second data; after obtaining the second influencing parameter, it can be inferred that the first data only has changes in the target parameter, and the corresponding second data has changes in the first influencing parameter, and the third data has changes in the third influencing parameter, then the target data may be affected by the first influencing parameter and the second influencing parameter, then only the target parameter in the first data, the second influencing parameter in the second data and the third influencing parameter in the third data are used to establish a neural network model, so that only these parameters are analyzed, which greatly reduces the complexity of the neural network model.

[0076] This embodiment establishes the neural network model including the following steps:

[0077] The target parameter, the first influencing parameter and the second influencing parameter form a training set, the number of input neurons in the neural network model is the same as the sum of the first influencing parameter and the second influencing parameter, the number of output neurons is the same as the number of target parameters, and the training set is used to train and generate the neural network model.

[0078] In this embodiment, a neural network is generated for each cluster training. For example, if cluster 1 includes 500 first parameter sequences, then there are 500 corresponding second parameter sequences and third parameter sequences. After processing, a combination of 500 target data, first influencing parameters and second influencing parameters can be obtained, and this combination can be used as a training set; wherein, if there are 5 target parameters, the number of output neurons of neural network model 1 is set to 5. If the number of first influencing parameters is 3, the number of second influencing parameters is 5, and the sum of the two is 8, then the number of input neurons of neural network model 1 is set to 8. If there are 2 target parameters in cluster 2, the number of output neurons of neural network model 2 is set to 2. If the number of first influencing parameters is 2, the number of second influencing parameters is 4, and the sum of the two is 6, then the number of input neurons of neural network model 2 is set to 6.

[0079] The association model includes a preprocessing model and a selection model. After receiving the actual input produced water parameters and discharged water parameters, the preprocessing model integrates the two into a first actual sequence and a second actual sequence respectively, and inputs the first actual sequence and the second actual sequence into the selection model.

[0080] The relevant personnel re-input 30 actual produced water parameters and 42 actual required discharged water parameters into the system. The preprocessing model will then generate a first actual sequence for the produced water parameters and a second actual sequence for the discharged water parameters, and then input the first actual sequence and the second actual sequence into the selection model.

[0081] The selection model is based on an existing neural network model, processes the first actual sequence and the second actual sequence into multiple subsequences, and inputs the subsequences into the neural network model of corresponding dimensions to obtain optimal setting parameters.

[0082] As before, there are multiple neural network models in the system. For example, the parameters that need to be input for neural network model 1 are: produced water parameters 1-3, discharged water parameters 2-4, and the parameters that need to be input for neural network model 2 are: produced water parameters 4-5, discharged water parameters 1, 5, 6. Then, after obtaining the first actual sequence, the first actual sequence includes produced water parameters 1-5 and discharged water parameters 1-6, so the first actual sequence is split into produced water parameters 1-3, produced water parameters 4-5, and the second actual sequence is split into discharged water parameters 1, 5, 6, and discharged water parameters 2-4, and then subsequence 1 is obtained as produced water parameters 1-3, discharged water parameters 2-4, and subsequence 2 is produced water parameters 4-5, discharged water parameters 1, 5, 6. Finally, subsequence 1 is input as a one-dimensional matrix into neural network model 1 to obtain a part of the optimal setting parameters, and subsequence 2 is input as a one-dimensional matrix into neural network model 2 to obtain another part of the optimal setting parameters.

[0083] In this embodiment, the historical data is preprocessed including the following steps:

[0084] After acquiring historical data, the produced water parameters, equipment parameters and discharged water parameters are standardized to obtain standardized data, set water quality emission indicators, and compare the standardized data with the water quality emission indicators to delete the discharged water parameters that do not meet the water quality emission indicators.

[0085] By standardizing the data, calculation errors due to different units can be avoided, such as converting 2kV to 2000V; then the water quality emission indicators for each type of discharge water parameter are set. For example, COD concentration below 40mg / L meets the emission standard. If the obtained discharge water parameter is COD concentration, its value is 100mg / L, which exceeds the water quality evaluation standard. Therefore, the discharge water parameter is deleted. In addition, since the parameter is deleted, the discharge water parameter under the corresponding existence time will be missing. Therefore, in order to facilitate subsequent comparison, all discharge water parameters under this existence time are deleted.

[0086] like Figure 2As shown, the present invention also provides a zero-discharge treatment system for oil and gas field produced water, which is used to implement the above-mentioned zero-discharge treatment method for oil and gas field produced water, and the system includes:

[0087] A collection module is used to collect historical data, the historical data includes a first parameter sequence, a second parameter sequence and a third parameter sequence measured at multiple time points, the first parameter sequence, the second parameter sequence and the third parameter sequence correspond to produced water parameters, equipment parameters and discharged water parameters respectively, pre-process the historical data, integrate the pre-processed historical data, and obtain the existence time of each first parameter sequence, second parameter sequence and third parameter sequence;

[0088] A clustering module, clustering the first parameter sequence to obtain a standard clustering result, analyzing the standard clustering result, and obtaining first data from the discharged water parameter;

[0089] an analysis module, locating the existence time corresponding to the first data, obtaining a second parameter sequence and a third parameter sequence corresponding to the existence time, defining them as second data and third data respectively, and analyzing the first data, the second data and the third data to obtain a correlation model among produced water parameters, equipment parameters and discharged water parameters;

[0090] The input module inputs the currently measured produced water parameters and the set discharged water parameters into the correlation model to obtain the current optimal setting parameters of the water purification equipment.

[0091] The present invention also discloses a computer storage medium, wherein the computer storage medium stores program instructions, wherein when the program instructions are executed, the device where the computer storage medium is located is controlled to execute the above method.

[0092] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0093] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0094] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A zero-discharge treatment method for produced water from oil and gas fields, characterized in that: include: Collecting historical data, the historical data including a first parameter sequence, a second parameter sequence and a third parameter sequence measured at multiple time points, the first parameter sequence, the second parameter sequence and the third parameter sequence respectively corresponding to a discharge water parameter, an equipment parameter and a produced water parameter, and preprocessing the historical data; Integrate the preprocessed historical data to obtain the existence duration of each of the first parameter sequence, the second parameter sequence, and the third parameter sequence; Clustering the first parameter sequence to obtain a standard clustering result, analyzing the standard clustering result, and obtaining first data from the discharged water parameter; Locate the existence duration corresponding to the first data, and obtain the second parameter sequence and the third parameter sequence corresponding to the existence duration, which are defined as second data and third data respectively; analyzing the first data, the second data, and the third data to obtain a correlation model among the produced water parameter, the equipment parameter, and the discharged water parameter; Inputting the currently measured produced water parameters and the set discharged water parameters into the correlation model to obtain the current optimal setting parameters of the water purification equipment; Clustering the first parameter sequence comprises the following steps: Setting an initial number of clusters, setting a final number of clusters based on the number of the discharged water parameters in the first parameter sequence, clustering all the first parameter sequences starting from the initial number of clusters to obtain a first result, including clusters of the initial number of clusters, each of which includes a plurality of the first parameter sequences; The first parameter sequence in the same cluster is defined as a same-cluster sequence, a comparison threshold is set for each type of the discharged water parameter, the absolute value of the difference between the same type of the discharged water parameters of two same-cluster sequences is compared, the discharged water parameter whose absolute value is greater than the comparison threshold is defined as a target parameter, the first number of the target parameters in the two same-cluster sequences is counted, this step is repeated to complete the comparison between the two same-cluster sequences, an average value of the first number is calculated, and if the average value is greater than the first threshold, the cluster is defined as a defect target; On the basis of the initial number of clusters, the number of clusters is increased to re-cluster the first parameter sequence to obtain a second result, and this step is repeated until the final number of clusters is reached, and the result with the smallest number of defect targets and less than or equal to a second threshold is selected as the standard clustering result; Obtaining the first data comprises the following steps: In the standard clustering result, if the first number of the target parameters between a plurality of the same cluster sequences is less than the first threshold, two of the same cluster sequences are defined as the first data.

2. The method according to claim 1, characterized in that If there is no result that the defect target number is less than or equal to the second threshold after reaching the final cluster number, the value of the first threshold is increased, and the first parameter sequence is clustered again starting with the initial cluster number.

3. The method according to claim 1, characterized in that Analyzing the first data, the second data, and the third data comprises the following steps: Acquire the second data corresponding to the first data, and if a first influencing parameter exists in the plurality of second data, define the second data containing the first influencing parameter as transit data, wherein the first influencing parameter is the device parameter in the plurality of second data, and a difference between the device parameters of the same type is greater than a third threshold; Obtain the third data corresponding to the transit data, obtain the second influencing parameter in the third data, and establish a neural network model according to the target parameter, the first influencing parameter and the second influencing parameter to analyze the first data, the second data and the third data.

4. The method according to claim 3, characterized in that Building a neural network model involves the following steps: The target parameter, the first influencing parameter and the second influencing parameter form a training set, the number of input neurons in the neural network model is the same as the sum of the target parameter and the second influencing parameter, the number of output neurons is the same as the number of the first influencing parameter, and the training set is used to train and generate a neural network model.

5. The method according to claim 4, characterized in that Obtaining the optimal setting parameters includes the following steps: The association model includes a preprocessing model and a selection model. After receiving the produced water parameter and the discharged water parameter that are actually input, the preprocessing model integrates the two into a first actual sequence and a second actual sequence respectively, and inputs the first actual sequence and the second actual sequence into the selection model; The selection model is based on an existing neural network model, processes the first actual sequence and the second actual sequence into multiple subsequences, and inputs the subsequences into the neural network model of corresponding dimensions to obtain the optimal setting parameters.

6. The method according to claim 1, characterized in that Preprocessing the historical data includes the following steps: After acquiring the historical data, the produced water parameters, the equipment parameters and the discharged water parameters are standardized to obtain standardized data, water quality emission indicators are set, and the standardized data and the water quality emission indicators are compared to delete the discharged water parameters that do not meet the water quality emission indicators.

7. A zero-discharge treatment system for produced water from oil and gas fields, used to implement the method according to any one of claims 1 to 6, characterized in that: include: A collection module, used to collect historical data, the historical data including a first parameter sequence, a second parameter sequence and a third parameter sequence measured at multiple time points, the first parameter sequence, the second parameter sequence and the third parameter sequence respectively corresponding to a discharge water parameter, an equipment parameter and a produced water parameter, preprocessing the historical data, integrating the preprocessed historical data, and obtaining the existence time of each of the first parameter sequence, the second parameter sequence and the third parameter sequence; a clustering module, clustering the first parameter sequence to obtain a standard clustering result; during clustering, setting an initial cluster number, setting a final cluster number based on the number of the drainage water parameters in the first parameter sequence, clustering all the first parameter sequences starting with the initial cluster number to obtain a first result, including clusters of the initial cluster number, each cluster including a plurality of the first parameter sequences, defining the first parameter sequences in the same cluster as same-cluster sequences, setting a comparison threshold for each type of drainage water parameter, comparing the absolute value of the difference between the drainage water parameters of the same type between two sequences in the same cluster, defining the drainage water parameter whose absolute value is greater than the comparison threshold as a target parameter, and counting the number of the target parameters in the two sequences in the same cluster; The first number of parameters, repeating this step to complete the comparison between the two same-cluster sequences, calculating the average value of the first number, if the average value is greater than a first threshold, defining the cluster as a defect target, increasing the number of clusters on the basis of the initial number of clusters, re-clustering the first parameter sequence to obtain a second result, repeating this step until the final number of clusters is reached, screening the result with the smallest number of defect targets and less than or equal to the second threshold as the standard clustering result, analyzing the standard clustering result, and obtaining the first data from the discharged water parameter, wherein, in the standard clustering result, if the first number of the target parameters between multiple same-cluster sequences is less than the first threshold, both of the same-cluster sequences are defined as the first data; an analysis module, locating the existence duration corresponding to the first data, obtaining the second parameter sequence and the third parameter sequence corresponding to the existence duration, defining them as second data and third data respectively, and analyzing the first data, the second data and the third data to obtain a correlation model between the produced water parameter, the equipment parameter and the discharged water parameter; The input module inputs the currently measured produced water parameters and the set discharged water parameters into the association model to obtain the current optimal setting parameters of the water purification equipment.

8. A computer storage medium, characterized in that The computer storage medium stores program instructions, wherein when the program instructions are executed, the device where the computer storage medium is located is controlled to execute the method according to any one of claims 1 to 6.