Synergistic preparation method and system for ocean drilling mud

Abstract

The invention discloses a collaborative preparation method and system for ocean drilling mud. The method comprises the following steps: acquiring target geological data of a predetermined drilling site, and determining a drilling fluid system according to the target geological data; wherein the geological data comprises a geological structure, stratum parameters and adjacent well conditions; selecting a recommended drilling fluid formula from a drilling fluid expert system according to the drilling fluid system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to target geological data; inputting the recommended drilling fluid formula into a formula prediction model to obtain a target drilling fluid formula; the formula prediction model is obtained by optimizing a connection weight and a threshold value of a BP neural network through an improved particle swarm algorithm. According to the embodiment of the invention, the problems of long period, low efficiency and the like of formula design through expert experience and experiments can be relieved. The embodiment of the invention is beneficial to improving the drilling fluid formula design efficiency. The method can be widely applied to the technical field of intelligent ocean drilling.

Description

Technical Field

[0001] The present invention relates to the field of intelligent ocean drilling technology, and in particular to a method and system for collaborative preparation of ocean drilling mud. Background Art

[0002] During ocean drilling, drilling fluid can effectively stabilize the formation, improve drilling efficiency, reduce non-operating time, reduce the incidence of underground accidents, and effectively protect the reservoir, which has a huge impact on the trial production of natural gas hydrates in the sea area under low-temperature weak colloid formation environment. Compared with land drilling operations, offshore drilling operations have great differences in drilling formations, construction conditions and construction processes. Ocean drilling coring is generally full-hole coring, which has the characteristics of high coring quality requirements, small diameter, high rotation speed, and small annular space gap. Therefore, ocean drilling coring cannot completely copy the mud formula of land oil drilling. The preparation of ocean drilling coring mud is more challenging and requires experienced experts to be determined after a large number of repeated experiments. The design of drilling fluid has many contents and a large span. The design results vary greatly according to the design concepts of different designers. At the same time, since a large number of experiments are required to verify the design results during the design process, the design cycle is generally long, which affects the efficiency of offshore drilling operations. Summary of the invention

[0003] The purpose of the present invention is to solve one of the technical problems existing in the prior art to at least a certain extent.

[0004] Therefore, the object of the present invention is to provide an efficient method and system for co-preparing ocean drilling mud.

[0005] In order to achieve the above technical objectives, one aspect of an embodiment of the present invention provides a method for collaborative preparation of ocean drilling mud, comprising the following steps: obtaining target geological data of a predetermined drilling site, and determining a drilling fluid system according to the target geological data; wherein the geological data includes geological structure, formation parameters, and adjacent well conditions; selecting a recommended drilling fluid formula from a drilling fluid expert system according to the drilling fluid system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to the target geological data; inputting the recommended drilling fluid formula into a formula prediction model to obtain a target drilling fluid formula; the formula prediction model is obtained by optimizing the connection weights and thresholds of a BP neural network through an improved particle swarm algorithm. The embodiment of the present application recommends a drilling fluid formula, and predicts the required drilling fluid formula based on an improved particle swarm algorithm combined with a BP neural network, so as to achieve accurate prediction of the drilling fluid formula; and is conducive to alleviating the problems of long cycle and low efficiency of formula design through expert experience and experiments. The embodiment of the present application is conducive to improving the efficiency of drilling fluid formula design.

[0006] In some embodiments, in the method for coordinating the preparation of ocean drilling mud according to the embodiment of the present invention, the formulation prediction model is trained by the following steps:

[0007] Get samples of recommended drilling fluid formulations;

[0008] Determine the initial parameters of the BP neural network according to the geological data; the initial parameters include initial connection weights and initial thresholds;

[0009] Initializing the dimension and initial value of each particle in the particle swarm algorithm according to the initial parameters; and determining the current optimal parameters based on the improved particle swarm algorithm; the current optimal parameters include the current optimal connection weight and the current optimal threshold;

[0010] The BP neural network based on the current optimal parameters is trained through the recommended drilling fluid formula sample to obtain the trained BP neural network.

[0011] In some embodiments, in one embodiment of the present invention, the BP neural network based on the current optimal parameters is trained by the recommended drilling fluid formula sample to obtain the trained BP neural network, including:

[0012] Inputting the recommended drilling fluid formula sample into the BP neural network based on the current optimal parameters to obtain a prediction result; and determining an error value according to the difference between the prediction result and the target result;

[0013] If the error value is less than or equal to a preset error threshold, determining that the trained BP neural network is the BP neural network based on the current optimal parameters;

[0014] Alternatively, if the error value is greater than the preset error threshold, the error value is processed by a first function to determine the fitness value of the particle; based on the fitness value, the particle group is updated, the first optimal parameter is determined, and the first optimal parameter is used as the new current optimal parameter; return to execute the step of inputting the recommended drilling fluid formula sample into the BP neural network based on the current optimal parameter to obtain a prediction result.

[0015] In some embodiments, in one embodiment of the present invention, the BP neural network includes an input layer, a hidden layer and an output layer; the initial parameters of the BP neural network are determined according to the geological data, including:

[0016] Determining the number of the input layers as a first number according to the geological data and the drilling data;

[0017] Determining the drilling fluid components according to the drilling fluid system, and then determining the number of the output layers to be a second number;

[0018] The initial parameters are determined according to the first number, the second number, and the number of hidden layers.

[0019] In some embodiments, in one embodiment of the present invention, the method further comprises:

[0020] If the number of hidden layers is 4, the initial connection weight is determined to be 5(m+n) 2 , the initial threshold is 4(m+n); wherein m represents the first number and n represents the second number.

[0021] In some embodiments, in one embodiment of the present invention, the drilling fluid includes a first component and a second component; the prediction result is: a first predicted proportion corresponding to the first component and a second predicted proportion corresponding to the second component; the target result is: a first actual proportion corresponding to the first component and a second actual proportion corresponding to the second component; the error value is determined by the following steps:

[0022] Taking an absolute value of a difference between the first actual proportion and the first predicted proportion to determine a first error;

[0023] Taking an absolute value of a difference between the second actual proportion and the second predicted proportion to determine a second error;

[0024] The sum of the first error and the second error is determined as the error value.

[0025] In some embodiments, in one embodiment of the present invention, the method further comprises:

[0026] According to the target drilling fluid formula, a mud configuration process is executed, and the configuration process is remotely interactive;

[0027] Alternatively, the target drilling fluid formula is stored in the drilling fluid expert system.

[0028] On the other hand, an embodiment of the present invention provides a system for coordinating preparation of ocean drilling mud, comprising:

[0029] The first module is used to obtain the target geological data of the predetermined drilling site and determine the drilling fluid system according to the target geological data; wherein the geological data includes geological structure, formation parameters, and adjacent well conditions;

[0030] The second module is used to select a recommended drilling fluid formula from a drilling fluid expert system according to the drilling fluid system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to the target geological data;

[0031] The third module is used to input the recommended drilling fluid formula into a formula prediction model to obtain a target drilling fluid formula; the formula prediction model is obtained by optimizing the connection weights and thresholds of the BP neural network through an improved particle swarm algorithm.

[0032] On the other hand, an embodiment of the present invention provides a device for coordinating preparation of ocean drilling mud, comprising:

[0033] at least one processor;

[0034] at least one memory for storing at least one program;

[0035] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned ocean drilling mud collaborative preparation method.

[0036] On the other hand, an embodiment of the present invention provides a storage medium storing a program executable by a processor, wherein the program executable by the processor is used to implement the above-mentioned method for coordinated preparation of ocean drilling mud when executed by the processor.

[0037] The embodiments of the present application include at least the following beneficial effects: The method provided by the embodiments of the present invention includes: obtaining the target geological data of the predetermined drilling site, and determining the drilling fluid system according to the target geological data; wherein the geological data include geological structure, formation parameters, and adjacent well conditions; according to the drilling fluid system, selecting a recommended drilling fluid formula from the drilling fluid expert system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to the target geological data; inputting the recommended drilling fluid formula into the formula prediction model to obtain the target drilling fluid formula; the formula prediction model is obtained by optimizing the connection weights and thresholds of the BP neural network through the improved particle swarm algorithm. The embodiments of the present application can achieve accurate prediction of the drilling fluid formula by recommending the drilling fluid formula, and predicting the required drilling fluid formula based on the improved particle swarm algorithm combined with the BP neural network; it is conducive to alleviating the problems of long cycle and low efficiency of formula design through expert experience and experiments. The embodiments of the present application are conducive to improving the efficiency of drilling fluid formula design. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the embodiments of the present invention or the drawings of related technical solutions in the prior art are introduced below. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1A schematic flow chart of an embodiment of the method for co-preparing ocean drilling mud provided by the present invention;

[0040] Figure 2 A schematic flow chart of another embodiment of the method for co-preparing ocean drilling mud provided by the present invention;

[0041] Figure 3 A schematic flow chart of an embodiment of a training process of a recipe prediction model provided by the present invention;

[0042] Figure 4 A schematic flow chart of another embodiment of the training process of the recipe prediction model provided by the present invention;

[0043] Figure 5 A schematic diagram of a flow chart of an embodiment of a process for determining initial parameters of a BP neural network provided by the present invention;

[0044] Figure 6 A schematic diagram of a flow chart of an embodiment of a process for determining an error value provided by the present invention;

[0045] Figure 7 A flowchart of an embodiment of a remote interaction process provided by the present invention;

[0046] Figure 8 A schematic diagram of a flow chart of an embodiment of a remote auxiliary recipe design process provided by the present invention;

[0047] Fig. 9 A schematic structural diagram of an embodiment of the ocean drilling mud collaborative preparation system provided by the present invention;

[0048] Fig.10 A schematic structural diagram of an embodiment of an ocean drilling mud collaborative preparation device provided by the present invention. DETAILED DESCRIPTION

[0049] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limitations of the present invention. For the step numbers in the following embodiments, they are only provided for the convenience of explanation, and the order between the steps is not limited in any way, and the execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0050] During ocean drilling, drilling fluid can effectively stabilize the formation, improve drilling efficiency, reduce non-operating time, reduce the incidence of underground accidents, and effectively protect the reservoir, which has a huge impact on the trial production of natural gas hydrates in the sea area under low-temperature weak colloid formation environment. Compared with land drilling operations, offshore drilling operations have great differences in drilling formations, construction conditions and construction processes. Ocean drilling coring is generally full-hole coring, which has the characteristics of high coring quality requirements, small diameter, high rotation speed, and small annular space gap. Therefore, ocean drilling coring cannot completely copy the mud formula of land oil drilling. The preparation of ocean drilling coring mud is more challenging and requires experienced experts to be determined after a large number of repeated experiments. The design of drilling fluid has many contents and a large span. The design results vary greatly according to the design concepts of different designers. At the same time, since a large number of experiments are required to verify the design results during the design process, the design cycle is generally long, which affects the efficiency of offshore drilling operations. When designing drilling fluid, it is necessary to consider not only factors such as lithology, geostress, shale physical and chemical properties, fluid properties in the formation, three pressure profiles, and geothermal gradient, but also relevant requirements for reservoir protection. At the same time, it is also necessary to refer to the complex downhole conditions and drilling fluid usage that have occurred in the wells that have been drilled in this block and the wells that have been drilled in adjacent blocks. Therefore, the design decision problem of drilling fluid belongs to a nonlinear optimization problem under multiple variables.

[0051] Drilling fluid plays an increasingly important role in drilling projects. Many complex problems in drilling projects are solved by adjusting the drilling fluid system or performance. Especially in the process of offshore natural gas hydrate trial production, the characteristics of drilling fluid have a significant impact on the decomposition and recovery rate of hydrates. Therefore, designing a scientific and reasonable drilling fluid system and formula in offshore drilling operations is of great significance to reduce formation damage and reduce the occurrence of drilling accidents.

[0052] Drilling fluid design is an interdisciplinary and complex system engineering, involving theoretical knowledge of many disciplines, and requires a large number of on-site indoor tests. It is difficult to achieve the preparation of drilling fluid under complex conditions by relying solely on shipboard experts and instruments and equipment. The embodiment of the present application adopts an algorithm based on chaotic search particle swarm to improve the global and local search capabilities of the algorithm, ensuring the correctness and efficiency of the preparation of ocean drilling mud under complex working conditions.

[0053] The ocean drilling mud collaborative preparation method and system proposed in the embodiment of the present invention will be described in detail below with reference to the accompanying drawings. First, the ocean drilling mud collaborative preparation method proposed in the embodiment of the present invention will be described with reference to the accompanying drawings.

[0054] Reference Figure 1In an embodiment of the present invention, a method for collaborative preparation of ocean drilling mud is provided. The method for collaborative preparation of ocean drilling mud in an embodiment of the present invention can be applied to a terminal, a server, or software running in a terminal or a server. The terminal can be a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The method for collaborative preparation of ocean drilling mud in an embodiment of the present invention mainly includes the following steps:

[0055] S100: Acquire target geological data of a predetermined drilling location, and determine a drilling fluid system according to the target geological data; wherein the geological data includes geological structure, formation parameters, and adjacent well conditions;

[0056] S200: selecting a recommended drilling fluid formula from a drilling fluid expert system according to the drilling fluid system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to the target geological data;

[0057] S300: inputting the recommended drilling fluid formula into a formula prediction model to obtain a target drilling fluid formula; the formula prediction model is obtained by optimizing the connection weights and thresholds of the BP neural network through an improved particle swarm algorithm.

[0058] It is understandable that the design of drilling fluid in the process of ocean drilling involves many contents and a large span, and the design schemes vary greatly according to the design concepts of different designers. At the same time, since a large number of experiments are required to verify the design results during the design process, the design cycle is generally long, which affects the efficiency of offshore drilling operations. When designing drilling fluid, it is necessary not only to consider factors such as lithology, geostress, physical and chemical properties of shale, fluid properties in the formation, three pressure profiles, and geothermal gradients, but also to consider relevant requirements for reservoir protection. At the same time, it is also necessary to refer to the complex downhole conditions and drilling fluid usage that have occurred in the wells that have been drilled in this block and the wells that have been drilled in adjacent blocks. The ocean drilling mud collaborative preparation method and system of the embodiment of the present application can realize the land-sea coordination and intelligent preparation of the drilling fluid preparation process, and improve the scientificity and timeliness of the drilling fluid preparation in the ocean drilling process.

[0059] Specifically, the collaborative configuration method in the embodiment of the present application includes:

[0060] Step S11, determining the drilling fluid system according to data such as geological structure, formation parameters, and adjacent well conditions;

[0061] Step S12, selecting one or more groups of drilling fluid formulas recommended for selection from the drilling fluid expert system according to the drilling fluid system, type and performance determined in the first step;

[0062] Step S13, prepare multiple groups of drilling fluids, obtain performance parameters of the prepared drilling fluids, and conduct laboratory verification work.

[0063] It is understandable that one or more drilling fluid formulas (i.e., recommended drilling fluid formulas) are selected based on human experience or artificial intelligence according to information such as geological structure, formation parameters, and adjacent well conditions. This formula is the starting point for solving the optimal formula later. In some embodiments, the selection process of the recommended drilling fluid formula prioritizes matching the drilling fluid system, then considers the matching type, and finally considers the matching performance.

[0064] The output layer outputs the proportion of each component of the drilling fluid. First, the drilling fluid system is determined according to the working conditions, such as the improved PEC drilling fluid system (applicable to the upper and middle formations of marine drilling), the PEM drilling fluid system (applicable to the deep formations and large displacement and horizontal wells of marine drilling), the EZFLOW drilling fluid system (applicable to large displacement wells, horizontal wells and oil layer protection in marine drilling), and the invisible acid completion fluid system (suitable for completion operations in marine drilling). The above are four commonly used drilling fluid systems in marine drilling. After determining the drilling fluid system, the dimension of the neural network output layer is determined according to the common components in the system. At the same time, according to the drilling fluid system, a similar recommended drilling fluid formula is selected, and this formula is used as the starting point for optimization.

[0065] In other embodiments, reference Figure 2 The specific process of the remote assisted design of the mud formula (i.e., the working process of the formula prediction model) provided in the embodiment of the present application is as follows:

[0066] Step S21, initializing the BP neural network structure to determine the number of neurons in each layer;

[0067] Step S22, initializing the dimension and initial value of each particle in the ECPSO algorithm (i.e., improved particle swarm algorithm), and randomly selecting training samples to input into the BP neural network for training;

[0068] Step S23, running the ECPSO algorithm, calculating the current fitness value of each individual particle, and obtaining the optimal connection weight and threshold;

[0069] Step S24, running the BP neural network to output the predicted value of the training sample and calculating the training error of the current training sample;

[0070] Step S25, determine whether all training samples have completed training, if completed, proceed to step S26; otherwise, use the current optimal connection weight and threshold as the ECPSO global optimal particle position, and return to step S23 to continue training;

[0071] Step S26, calculating the total training error of all training samples, and judging whether the total error satisfies the training termination condition, until the termination condition is met.

[0072] In short, since the preparation of drilling fluid is closely related to factors such as field conditions, formation characteristics and expert experience, it is difficult to achieve scientific and rapid preparation of drilling fluid under complex working conditions relying solely on field personnel and expert systems supported by a database. It is necessary to comprehensively discuss and determine the field information and the rich experience of shore-based experts. To this end, the embodiment of the present application provides a remote collaborative interaction platform for drilling fluid based on an expert knowledge base and remote audio and video interaction, which improves the accuracy of spare parts prediction. In response to the premature convergence problem of the particle swarm algorithm in the process of ocean drilling mud reasoning, the embodiment of the present application proposes an inertia weight calculation method based on the number of iterations and fitness value, which improves the global search ability and local search ability of the algorithm, and uses the chaotic idea to perform chaotic search on the global optimal solution to improve the convergence accuracy of the algorithm.

[0073] Optionally, refer to Figure 3 In one embodiment of the present invention, the recipe prediction model is trained by the following steps:

[0074] Step S400, obtaining a recommended drilling fluid formula sample;

[0075] Step S500, determining the initial parameters of the BP neural network according to the geological data; the initial parameters include initial connection weights and initial thresholds;

[0076] Step S600, initializing the dimension and initial value of each particle in the particle swarm algorithm according to the initial parameters; and determining the current optimal parameters based on the improved particle swarm algorithm; the current optimal parameters include the current optimal connection weight and the current optimal threshold;

[0077] Step S700, training the BP neural network based on the current optimal parameters by recommending drilling fluid formula samples to obtain a trained BP neural network.

[0078] In some possible implementations, the dimension of the particle in the embodiment of the present application is determined according to the structure of the neural network, and the number of dimensions of the particle is the sum of the weight and threshold in the neural network. It can be understood as the number of parameters in the neural network. The initial value of the particle is generated by chaotic mapping, that is, the value of the weight and threshold in the neural network. The chaotic system is also called a chaotic sequence. As a nonlinear dynamic system, it has the characteristics of ergodicity, unpredictability, and non-periodicity, and is usually used to improve the performance of heuristic algorithms. The idea of ​​a chaotic sequence is to generate chaotic variables in the neighborhood space of the target to be optimized through a specific chaotic mapping method, and the entire neighborhood space can be traversed without repetition through rapid iteration of chaotic variables, so as to solve the problem of stagnation of the optimal solution in the heuristic algorithm, thereby improving the convergence speed of the algorithm.

[0079] It can be understood that the embodiment of the present application determines the current optimal parameters through the improved particle swarm algorithm, and then optimizes the BP neural network according to the current optimal parameters, and then predicts the recipe through the optimized BP neural network.

[0080] Optionally, refer to Figure 4 In one embodiment of the present invention, by recommending a drilling fluid formula sample, a BP neural network based on the current optimal parameters is trained to obtain a trained BP neural network, including:

[0081] Step S710, inputting the recommended drilling fluid formula sample into the BP neural network based on the current optimal parameters to obtain a prediction result; and determining an error value according to the difference between the prediction result and the target result;

[0082] Step S720, if the error value is less than or equal to the preset error threshold, determine that the trained BP neural network is a BP neural network based on the current optimal parameters;

[0083] Alternatively, in step S730, if the error value is greater than a preset error threshold, the error value is processed by a first function to determine the fitness value of the particle; based on the fitness value, the particle swarm is updated to determine the first optimal parameter, and the first optimal parameter is used as the new current optimal parameter; and the step of returning to execute inputting the recommended drilling fluid formula sample into the BP neural network based on the current optimal parameter to obtain the prediction result is performed.

[0084] In some possible implementations, the training process of the recipe prediction model may be:

[0085] Step S31, initializing the BP neural network structure and determining the number of neurons in each layer;

[0086] Step S32, initializing the dimension and initial value of each particle in the ECPSO algorithm according to the number and initial value of the connection weights and thresholds of the BP neural network, and randomly selecting training samples to input into the BP neural network for training;

[0087] Step S33, running the ECPSO algorithm, calculating the current fitness value of each individual particle, and obtaining the optimal connection weight and threshold;

[0088] Step S34, running the BP neural network using the optimal connection weight and threshold, outputting the predicted value of the training sample, and calculating the training error of the current training sample;

[0089] Step S35, determine whether all training samples have completed training, if completed, proceed to step S36; otherwise, use the current optimal connection weight and threshold as the ECPSO global optimal particle position, and return to step S33 to continue training;

[0090] Step S36, calculate the total training error of all training samples, and determine whether the total error meets the training termination condition. If so, the training is completed; otherwise, return to step S33 to continue training until the termination condition is met.

[0091] Optionally, refer to Figure 5 In one embodiment of the present invention, the BP neural network includes an input layer, a hidden layer and an output layer; according to the geological data, the initial parameters of the BP neural network are determined, including:

[0092] Step S510, determining the number of input layers as a first number according to geological data and drilling data;

[0093] Step S520, determining the drilling fluid components according to the drilling fluid system, and further determining the number of output layers to be a second number;

[0094] Step S530: determining initial parameters according to the first quantity, the second quantity, and the number of hidden layers.

[0095] Optionally, in one embodiment of the present invention, the method further comprises:

[0096] If the number of hidden layers is 4, the initial connection weight is determined to be 5 (m+n) 2 , the initial threshold is 4(m+n); wherein m represents the first number and n represents the second number.

[0097] Optionally, refer to Figure 6In one embodiment of the present invention, the drilling fluid includes a first component and a second component; the prediction result is: a first predicted proportion corresponding to the first component and a second predicted proportion corresponding to the second component; the target result is: a first actual proportion corresponding to the first component and a second actual proportion corresponding to the second component; the error value is determined by the following steps:

[0098] Step S711, taking an absolute value of a difference between the first actual proportion and the first predicted proportion to determine a first error;

[0099] Step S712, taking an absolute value of the difference between the second actual proportion and the second predicted proportion to determine a second error;

[0100] Step S713: determine the sum of the first error and the second error as the error value.

[0101] In some possible implementations, the first component and the second component in the embodiment of the present application may represent several components in the drilling fluid, and the number of components in the drilling fluid is not limited. For example, the drilling fluid contains different components, and the absolute value of the error of each component is summed. For example, the target content is 0.7 water, 0.2 dispersant, 0.1 polymer, and the model output value is 0.6 water, 0.25 dispersant, 0.15 polymer. Then the error is |0.7-0.6|+|0.2-0.25|+|0.1-0.15|=0.2.

[0102] Optionally, refer to Figure 7 In one embodiment of the present invention, the method further comprises:

[0103] Step S800, executing the mud configuration process according to the target drilling fluid formula, and making the configuration process remotely interactive;

[0104] Alternatively, in step S900, the target drilling fluid formula is stored in the drilling fluid expert system.

[0105] In some possible implementations, the present application also provides the following functions:

[0106] 1) Remote interaction of audio and video information during mud preparation process;

[0107] 2) Remote auxiliary design of mud formula;

[0108] 3) Remote query and sharing of information;

[0109] 4) Remote assisted computing;

[0110] 5) Remote knowledge maintenance function.

[0111] Specifically:

[0112] 1) Remote interaction of audio and video information during mud preparation process;

[0113] During the preparation of mud, experts are required to visually observe and judge the morphology and rheological properties of the prepared test mud. Therefore, the remote video sharing function of the mud remote support system can realize the real-time sharing of the state of the test mud prepared on the ship and shore. At the same time, it provides a more efficient and friendly remote communication platform for ship and shore experts, so that ship and shore experts can share parameters and information while conducting efficient voice and video communication.

[0114] 2) Remote auxiliary design of mud formula;

[0115] Shore-based experts or shipboard users can input relevant information about the drilled strata and obtain multiple successful construction cases in such strata. In the formula details of these successful cases, they can find the corresponding formula, mud performance reference and detailed data of the mud materials used in the formula, so as to assist on-site mud technicians to prepare suitable mud formulas more quickly.

[0116] 3) Remote query and sharing of information;

[0117] Shore-based experts can query in real time the information on the on-site formation and the parameters of mud preparation and testing stored and recorded by the ship-borne terminal of the ship-borne mud remote support system. At the same time, the ship-borne mud remote support system can query the shore-based database of the shore-based mud remote support system for information such as mature mud formulas, engineering examples, mud formula systems and mud materials.

[0118] 4) Remote assisted computing;

[0119] In the experimental process of mud formula design, it is necessary to infer and calculate various parameters of the mud. The mud decision-making using knowledge base and intelligent algorithm requires a lot of calculations and numerical iterations. Relying on the ship-borne computing system is time-consuming and inefficient. Through the mud remote support system, the field parameters are transmitted back to the shore, and the shore-based computing center implements the model training and reasoning calculation. After the corresponding formula is obtained in the experiment, the engineering amount of each mud material must be calculated according to the formula addition. The above calculation process can be completed by the land-based remote computing center, thereby reducing the requirements for ship-borne computing equipment and improving computing efficiency.

[0120] 5) Remote knowledge maintenance function;

[0121] Onboard users can add, modify and delete information in each database within the system, which can conveniently and effectively save successful experiences. At the same time, they can continuously improve the successful cases and knowledge elements within the system, thereby continuously enriching and improving the effectiveness of the inference engine.

[0122] The prediction process provided by the embodiment of the present application is described in detail below with a specific embodiment:

[0123] The first step is to initialize the BP neural network structure. The input layer input feature vector mainly contains geological information: rock type, formation porosity, density, permeability, gas content, water content, formation pressure, pH, stability. Drilling information: well depth, well diameter, drilling pressure, footage, pump pressure, pump displacement, riser pressure, casing pressure and other multi-dimensional information. The number of layers and the number of neurons in the hidden layer are determined based on experience according to the complexity of the task; the output layer determines the drilling fluid system based on data such as geological structure, formation parameters, and adjacent well conditions, and determines the main components of the drilling fluid based on the drilling fluid system to determine the number of neurons in the output layer. It can be understood that the determination of the number of hidden layers and the number of neurons in each layer of the neural network can be set according to the corresponding empirical formulas in different embodiments, and this application does not make specific limitations.

[0124] In the second step, if the number of input layers and output layers of the neural network are m and n respectively, it is determined based on experience that 4 hidden layers have a better training effect. The number of hidden layers in each layer is m+n, and the number of neural network connection weights can be determined to be 5(m+n). 2 , the number of thresholds is 4(m+n). The initial values ​​of the neural network connection weights and thresholds are randomly initialized in the range of [-1,1]. According to the number and initial values ​​of the BP neural network connection weights and thresholds, the dimension of each particle in the ECPSO algorithm is initialized (5(m+n) 2 +4(m+n)) and initial value (which can be generated according to chaotic mapping), randomly select training samples and input them into BP neural network for training.

[0125] The third step is to run the ECPSO algorithm to calculate the current fitness value of each particle and obtain the optimal connection weight and threshold;

[0126] The fourth step is to run the BP neural network using the optimal connection weights and thresholds, output the predicted value of the training sample, and calculate the training error of the current training sample. Where n is the number of drilling fluid components, x i is the predicted value of the proportion of the i-th component of the drilling fluid (i.e., the first predicted proportion or the second predicted proportion), s i is the set value of the proportion of the i-th component of the drilling fluid (i.e., the first actual proportion or the second actual proportion).

[0127] Step 5: Determine whether all training samples have completed training. If so, proceed to step 6. Otherwise, use the current optimal connection weight and threshold as the global optimal particle position of ECPSO and return to step 3 to continue training.

[0128] The sixth step is to calculate the total training error of all training samples and determine whether the total error meets the training termination condition. If so, the training is completed; otherwise, return to the third step to continue training until the termination condition is met.

[0129] It is understandable that, referring to Figure 8 As shown, the recipe remote assisted design process in the embodiment of the present application may also include the following steps:

[0130] The training process of BP neural network based on particle swarm is as follows:

[0131] Step S81, initialization. Specifically includes:

[0132] Step S811, initialize the weights and thresholds of the BP neural network, and determine the structure of the network, including the number of nodes in the input layer, hidden layer, and output layer.

[0133] Step S812, initialize the particle swarm, each particle represents a set of weights and thresholds of the BP neural network. The size of the particle swarm is determined according to the actual situation.

[0134] Step S82, fitness evaluation.

[0135] Step S821: for each particle, substitute its corresponding weight and threshold into the BP neural network.

[0136] Step S822, using the training data set to train the BP neural network and calculate the output error of the network.

[0137] Step S823: A certain function of the output error (such as the inverse of the mean square error) is used as the fitness value of the particle.

[0138] Step S83, particle swarm update.

[0139] Step S831, speed update.

[0140] Step S832, location update.

[0141] Step S84: BP neural network training.

[0142] Step S841, using the weights and thresholds obtained by particle swarm optimization, the BP neural network is trained again.

[0143] Step S842, continuously adjust the weights and thresholds so that the output error of the network gradually decreases.

[0144] Step S85, repeat steps S82 to S84.

[0145] Step S851, repeat the process of fitness evaluation, particle swarm update and BP neural network training until the stop condition is met.

[0146] Step S852, the stopping condition may be that the maximum number of iterations is reached, the output error is less than a certain threshold, etc.

[0147] Step S86, output the result.

[0148] Step S861, after the training is completed, the weights and thresholds of the optimized BP neural network are output.

[0149] Step S862: Use the test data set to test the trained network and evaluate its performance.

[0150] In summary, the method provided by the embodiment of the present application includes: obtaining the target geological data of the predetermined drilling site, and determining the drilling fluid system according to the target geological data; wherein the geological data includes geological structure, formation parameters, and adjacent well conditions; according to the drilling fluid system, selecting a recommended drilling fluid formula from the drilling fluid expert system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to the target geological data; inputting the recommended drilling fluid formula into the formula prediction model to obtain the target drilling fluid formula; the formula prediction model is obtained by optimizing the connection weights and thresholds of the BP neural network through the improved particle swarm algorithm. The embodiment of the present application recommends the drilling fluid formula, and predicts the required drilling fluid formula based on the improved particle swarm algorithm combined with the BP neural network, so as to achieve accurate prediction of the drilling fluid formula; it is helpful to alleviate the problems of long cycle and low efficiency of formula design through expert experience and experiments. The embodiment of the present application is conducive to improving the efficiency of drilling fluid formula design.

[0151] Secondly, refer to the attached Fig. 9 A system for coordinating preparation of ocean drilling mud according to an embodiment of the present invention is described.

[0152] Fig. 9 1 is a schematic diagram of the structure of an ocean drilling mud collaborative preparation system according to an embodiment of the present invention. The system specifically includes:

[0153] The first module 910 is used to obtain target geological data of a predetermined drilling location and determine a drilling fluid system according to the target geological data; wherein the geological data includes geological structure, formation parameters, and adjacent well conditions;

[0154] The second module 920 is used to select a recommended drilling fluid formula from the drilling fluid expert system according to the drilling fluid system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to the target geological data;

[0155] The third module 930 is used to input the recommended drilling fluid formula into the formula prediction model to obtain the target drilling fluid formula; the formula prediction model is obtained by optimizing the connection weights and thresholds of the BP neural network through the improved particle swarm algorithm.

[0156] Optionally, in one embodiment of the present invention, the system further comprises a fourth module for obtaining a sample of a recommended drilling fluid formula;

[0157] According to the geological data, the initial parameters of the BP neural network are determined; the initial parameters include the initial connection weight and the initial threshold;

[0158] According to the initial parameters, the dimension and initial value of each particle in the particle swarm algorithm are initialized; and based on the improved particle swarm algorithm, the current optimal parameters are determined; the current optimal parameters include the current optimal connection weight and the current optimal threshold;

[0159] By recommending drilling fluid formula samples, the BP neural network based on the current optimal parameters is trained to obtain a trained BP neural network.

[0160] Optionally, in one embodiment of the present invention, the system further comprises a fifth module, which is used to input the recommended drilling fluid formula sample into the BP neural network based on the current optimal parameters to obtain a prediction result; and determine an error value according to the difference between the prediction result and the target result;

[0161] If the error value is less than or equal to the preset error threshold, it is determined that the trained BP neural network is a BP neural network based on the current optimal parameters;

[0162] Alternatively, if the error value is greater than a preset error threshold, the error value is processed by the first function to determine the fitness value of the particle; based on the fitness value, the particle group is updated to determine the first optimal parameter, and the first optimal parameter is used as the new current optimal parameter; return to execute the step of inputting the recommended drilling fluid formula sample into the BP neural network based on the current optimal parameter to obtain the prediction result.

[0163] Optionally, in one embodiment of the present invention, the system further comprises a sixth module, configured to determine the number of input layers as a first number based on the geological data and the drilling data;

[0164] Determine the drilling fluid components according to the drilling fluid system, and then determine the number of output layers as a second number;

[0165] An initial parameter is determined according to the first number, the second number, and the number of hidden layers.

[0166] Optionally, in one embodiment of the present invention, the system further comprises a seventh module for executing a mud configuration process according to a target drilling fluid formulation and performing remote interaction on the configuration process;

[0167] Alternatively, the target drilling fluid formula is stored in a drilling fluid expert system.

[0168] It can be seen that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0169] Reference Fig.10 The embodiment of the present invention provides a device for coordinating preparation of ocean drilling mud, comprising:

[0170] at least one processor 110;

[0171] At least one memory 120, used to store at least one program;

[0172] When the at least one program is executed by the at least one processor 110, the at least one processor 110 implements the method for coordinated preparation of ocean drilling mud.

[0173] Similarly, the contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0174] An embodiment of the present invention further provides a computer-readable storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to execute the above-mentioned method for coordinated preparation of ocean drilling mud.

[0175] Similarly, the contents of the above method embodiments are all applicable to the present storage medium embodiments. The functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0176] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0177] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0178] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several programs to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0179] The logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable programs for implementing the logical functions, and may be embodied in any computer-readable medium for use by a program execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch and execute a program from a program execution system, device or apparatus), or in conjunction with such program execution systems, devices or apparatuses. For purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by a program execution system, device or apparatus, or in conjunction with such program execution systems, devices or apparatuses.

[0180] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0181] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable program execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0182] In the above description of this specification, the description with reference to the terms "one embodiment / example", "another embodiment / example" or "certain embodiments / examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0183] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0184] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A method for co-preparing ocean drilling mud, characterized in that: The following steps are involved: Obtaining target geological data of a predetermined drilling site, and determining a drilling fluid system based on the target geological data; wherein the geological data includes geological structure, formation parameters, and adjacent well conditions; According to the drilling fluid system, selecting a recommended drilling fluid formula from a drilling fluid expert system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to the target geological data; The recommended drilling fluid formula is input into a formula prediction model to obtain a target drilling fluid formula; the formula prediction model is obtained by optimizing the connection weights and thresholds of a BP neural network through an improved particle swarm algorithm.

2. The method for coordinating preparation of ocean drilling mud according to claim 1, characterized in that: The recipe prediction model is trained by the following steps: Get samples of recommended drilling fluid formulations; Determine the initial parameters of the BP neural network according to the geological data; the initial parameters include initial connection weights and initial thresholds; Initializing the dimension and initial value of each particle in the particle swarm algorithm according to the initial parameters; and determining the current optimal parameters based on the improved particle swarm algorithm; the current optimal parameters include the current optimal connection weight and the current optimal threshold; The BP neural network based on the current optimal parameters is trained through the recommended drilling fluid formula sample to obtain the trained BP neural network.

3. The method for coordinating ocean drilling mud according to claim 2, characterized in that: The step of training the BP neural network based on the current optimal parameters by using the recommended drilling fluid formula sample to obtain the trained BP neural network includes: Inputting the recommended drilling fluid formula sample into the BP neural network based on the current optimal parameters to obtain a prediction result; and determining an error value according to the difference between the prediction result and the target result; If the error value is less than or equal to a preset error threshold, determining that the trained BP neural network is the BP neural network based on the current optimal parameters; Alternatively, if the error value is greater than the preset error threshold, the error value is processed by a first function to determine the fitness value of the particle; based on the fitness value, the particle group is updated, the first optimal parameter is determined, and the first optimal parameter is used as the new current optimal parameter; return to execute the step of inputting the recommended drilling fluid formula sample into the BP neural network based on the current optimal parameter to obtain a prediction result.

4. The method for coordinating preparation of ocean drilling mud according to claim 2, characterized in that: The BP neural network includes an input layer, a hidden layer and an output layer; Determining the initial parameters of the BP neural network according to the geological data includes: Determining the number of the input layers as a first number according to the geological data and the drilling data; Determining the drilling fluid components according to the drilling fluid system, and then determining the number of the output layers to be a second number; The initial parameters are determined according to the first number, the second number, and the number of hidden layers.

5. The method for coordinating ocean drilling mud according to claim 4, characterized in that: The method further comprises: If the number of hidden layers is 4, the initial connection weight is determined to be 5(m+n) 2 , the initial threshold is 4(m+n); wherein m represents the first number and n represents the second number.

6. The method for coordinating preparation of ocean drilling mud according to claim 3, characterized in that: The drilling fluid includes a first component and a second component; the prediction result is: a first predicted proportion corresponding to the first component and a second predicted proportion corresponding to the second component; the target result is: a first actual proportion corresponding to the first component and a second actual proportion corresponding to the second component; the error value is determined by the following steps: Taking an absolute value of a difference between the first actual proportion and the first predicted proportion to determine a first error; Taking an absolute value of a difference between the second actual proportion and the second predicted proportion to determine a second error; The sum of the first error and the second error is determined as the error value.

7. The method for coordinating preparation of ocean drilling mud according to claim 1, characterized in that: The method further comprises: According to the target drilling fluid formula, a mud configuration process is executed, and the configuration process is remotely interactive; Alternatively, the target drilling fluid formula is stored in the drilling fluid expert system.

8. A system for coordinating preparation of ocean drilling mud, characterized in that: include: The first module is used to obtain the target geological data of the predetermined drilling site and determine the drilling fluid system according to the target geological data; wherein the geological data includes geological structure, formation parameters, and adjacent well conditions; The second module is used to select a recommended drilling fluid formula from a drilling fluid expert system according to the drilling fluid system; wherein the geological data corresponding to the recommended drilling fluid formula is similar to the target geological data; The third module is used to input the recommended drilling fluid formula into a formula prediction model to obtain a target drilling fluid formula; the formula prediction model is obtained by optimizing the connection weights and thresholds of the BP neural network through an improved particle swarm algorithm.

9. A device for coordinating preparation of ocean drilling mud, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the ocean drilling mud coordinated preparation method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The processor-executable program is used to implement the ocean drilling mud coordinated preparation method according to any one of claims 1 to 7 when executed by the processor.

Images