Digitization-based drilling machine intelligent remote control method and control system

Abstract

The invention relates to the field of intelligent remote control, in particular to a digitization-based intelligent remote control method and a digitization-based intelligent remote control system for a drilling machine. The system monitors and collects drilling parameter data in real time through a sensor network, and transmits the drilling parameter data to a remote control server through a wireless communication network. In a remote control server, a data processing and analysis algorithm based on artificial intelligence and deep learning is used to carry out time sequence analysis and collaborative association feature description on the drilling parameter data, so as to capture context semantics among time sequence features of each drilling parameter. The bit pressure time sequence semantics are used as query features to query and match the optimal bit pressure recommendation decoding value from the other drilling parameter time sequence collaborative association features, and the bit pressure recommendation decoding value is compared with the bit pressure value at the current time point to generate a bit pressure remote regulation and control instruction, so that the bit pressure remote regulation and control instruction is quickly adapted to the change of geological conditions in the drilling process; and remote intelligent control of the drilling machine is realized.

Description

Technical Field

[0001] The present application relates to the field of intelligent remote control, and in particular to a digital-based drilling rig intelligent remote control method and control system. Background Art

[0002] Drilling is a core technical activity in the oil and gas industry, which involves drilling holes in underground rock layers to explore and extract oil and gas resources. Drilling engineering is an extremely complex process involving a variety of technologies, equipment and materials. In order to ensure the effectiveness and safety of drilling operations, prevent accidents such as blowouts and stuck drill bits, and improve the safety of workers, remote and sophisticated real-time control of the drilling process is required.

[0003] However, traditional drilling rig control mainly relies on the experience and real-time decision-making of on-site operators, which may lead to inefficiency and safety issues caused by human factors. Specifically, differences in operator skill levels and experience will affect drilling results and may lead to misjudgment or slow response. In addition, when encountering complex geological conditions during the drilling process, the traditional drilling rig control solution has a low level of intelligence, often requiring frequent adjustments to drilling parameters, and is difficult to quickly adapt to changes in geological conditions. Not only that, traditional control methods have limited ability to handle emergencies, and are prone to safety accidents such as blowouts and well collapses, which also pose safety risks to rig operators.

[0004] Therefore, an optimized drilling rig control solution is needed to solve the above technical problems. Summary of the invention

[0005] This application is made in consideration of the above problems. One object of this application is to provide a digital-based intelligent remote control method and control system for a drilling rig.

[0006] The embodiment of the present application provides a digitized intelligent remote control method for a drilling rig, which includes: Collecting a time queue of drilling pressure value, a time queue of torque value, a time queue of pump pressure value, a time queue of mud flow, a time queue of mud density, a time queue of drilling depth, a time queue of temperature and a time queue of humidity through a sensor network; Transmitting the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity to a remote control server through a wireless communication module; On the remote control server, respectively, for the remote control server, the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature, and the time queue of the humidity are time-series encoded to obtain a drilling pressure time-series associated implicit feature vector, a torque time-series associated implicit feature vector, a pump pressure time-series associated implicit feature vector, a mud flow time-series associated implicit feature vector, a mud density time-series associated implicit feature vector, a drilling depth time-series associated implicit feature vector, a temperature time-series associated implicit feature vector, and a temperature time-series associated implicit feature vector; In the remote control server, the drilling pressure time series associated implicit feature vector is used as a query feature vector and a feature matrix consisting of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector is used as an initial key matrix, and the query feature vector and the initial key matrix are input into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector; On the remote control server, an optimal drilling pressure recommended decoding value is generated based on the drilling pressure cross-domain query matching feature vector, and a drilling pressure remote control instruction is generated based on the optimal drilling pressure recommended decoding value and the drilling pressure value at the current time point.

[0007] For example, according to the digitized intelligent remote control method for drilling rigs of an embodiment of the present application, in which, on the remote control server, the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity are time-series encoded for the remote control server respectively to obtain a drilling pressure time-series associated implicit feature vector, a torque time-series associated implicit feature vector, a pump pressure time-series associated implicit feature vector, a mud flow time-series associated implicit feature vector, a mud density time-series associated implicit feature vector, a drilling depth time-series associated implicit feature vector, a temperature time-series associated implicit feature vector and a temperature time-series associated implicit feature vector, including: The time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity are respectively input into the sequence encoder based on the Bi-LSTM model to obtain the drilling pressure time series associated implicit feature vector, the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector.

[0008] For example, according to the digitized intelligent remote control method for drilling rigs according to an embodiment of the present application, in the remote control server, the drilling pressure time series associated implicit feature vector is used as a query feature vector and a feature matrix composed of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector is used as an initial key matrix, and the query feature vector and the initial key matrix are input into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector, including: After linearly transforming the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector respectively, the linearly transformed time series associated implicit feature vectors are matrixed to obtain the initial key matrix; Calculating the characteristic distribution difference energy coefficients between the WOB time series associated implicit characteristic vector and each row vector in the initial key matrix to obtain a set of drilling parameter time series semantic characteristic distribution difference energy coefficients; Comparing each drilling parameter time series semantic feature distribution difference energy coefficient in the set of drilling parameter time series semantic feature distribution difference energy coefficients with a predetermined threshold, and in response to the drilling parameter time series semantic feature distribution difference energy coefficient being less than the predetermined threshold, taking the mean vector between the row vector in the initial key matrix corresponding to the drilling parameter time series semantic feature distribution difference energy coefficient and the implicit feature vector associated with the drilling pressure time series as the updated row vector to obtain an updated key matrix; Using a value embedding matrix and a value embedding vector to process the drilling pressure time series associated implicit feature vector to obtain a drilling pressure time series semantic value vector; Taking the WOB time series associated implicit feature vector as a query vector, each updated row vector in the update key matrix as a key vector and the WOB time series semantic value vector as a value vector, the query vector, the key vector and the value vector are input into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of drilling parameter time series semantic cross-domain query attention vectors; The position-based mean vector of the sequence of the drilling parameter time series semantic cross-domain query focus vector is calculated to obtain the drilling pressure cross-domain query matching feature vector.

[0009] For example, according to the digitized intelligent remote control method for drilling rigs according to an embodiment of the present application, the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector are linearly transformed, and the linearly transformed time series associated implicit feature vectors are matrixed to obtain the initial key matrix, including: The torque time-series associated implicit feature vector, the pump pressure time-series associated implicit feature vector, the mud flow time-series associated implicit feature vector, the mud density time-series associated implicit feature vector, the drilling depth time-series associated implicit feature vector, the temperature time-series associated implicit feature vector and each of the time-series associated implicit feature vectors are multiplied by the corresponding linear weight matrix respectively and then added with the linear bias vector by position to obtain a plurality of linearly transformed time-series associated implicit feature vectors; The multiple linearly transformed temporally associated implicit feature vectors are arranged in a matrix to obtain the initial key matrix.

[0010] For example, according to the digitized intelligent remote control method for drilling rigs according to an embodiment of the present application, the characteristic distribution difference energy coefficient between the implicit characteristic vector associated with the drilling pressure time series and each row vector in the initial key matrix is ​​calculated to obtain a set of drilling parameter time series semantic characteristic distribution difference energy coefficients, including: Respectively calculating the positional difference, positional dot product and positional addition between the WOB timing-associated implicit feature vector and the row vector in the initial key matrix to obtain a WOB timing semantics-drilling parameter timing semantics differential feature vector, a WOB timing semantics-drilling parameter timing semantics dot product feature vector and a WOB timing semantics-drilling parameter timing semantics sum feature vector; The drilling pressure time series semantics-drilling parameter time series semantics differential feature vector, the drilling pressure time series semantics-drilling parameter time series semantics point product feature vector and the drilling pressure time series semantics-drilling parameter time series semantics sum feature vector are cascaded and processed through a one-dimensional convolution layer, and the obtained feature vector is subjected to maximum pooling processing to obtain the drilling pressure time series semantics-drilling parameter time series semantics feature association representation vector; Extracting the maximum value of the feature in the drilling pressure time series semantics-drilling parameter time series semantics feature association representation vector to obtain the maximum representation value of the drilling pressure time series semantics-drilling parameter time series semantics association; Calculating the mean and variance of the row vectors in the initial key matrix to obtain the mean and variance of the drilling parameter time series semantic features; Adding the drilling parameter time series semantic feature variance to the hyperparameter to obtain a first feature distribution difference energy factor; Calculate the square of the difference between the maximum representation value of the drilling pressure time series semantics-drilling parameter time series semantics association and the mean value of the drilling parameter time series semantic features to obtain a semantic feature difference value; The drilling parameter time series semantic feature variance is multiplied by a constant of two and then added to the semantic feature difference value and the hyperparameter to obtain a second feature distribution difference energy factor; The first feature distribution difference energy factor is divided by the second feature distribution difference energy factor to obtain the drilling parameter time series semantic feature distribution difference energy coefficient.

[0011] For example, according to the digitization-based intelligent remote control method for drilling rigs of an embodiment of the present application, the drilling pressure timing-associated implicit feature vector is processed using a value embedding matrix and a value embedding vector to obtain a drilling pressure timing semantic value vector, including: calculating the matrix multiplication between the drilling pressure timing-associated implicit feature vector and the value embedding matrix, and then adding the vector to the value embedding vector by position to obtain the drilling pressure timing semantic value vector.

[0012] For example, according to the digitized intelligent remote control method for drilling rigs according to an embodiment of the present application, the drilling pressure time series associated implicit feature vector is used as a query vector, each updated row vector in the update key matrix is ​​used as a key vector, and the drilling pressure time series semantic value vector is used as a value vector. The query vector, the key vector, and the value vector are input into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of drilling parameter time series semantic cross-domain query attention vectors, including: After calculating the product between the query vector and the transposed vector of the key vector, dividing the obtained query-key feature association value by the square root of the length of the key vector to obtain a query-key feature association coefficient; Inputting the query-key feature association coefficient into a softmax function to obtain a query-key feature association weight value; The query-key feature association weight value is used as a weight, and the transposed vector of the value vector is weighted to obtain a drilling parameter temporal semantic cross-domain query attention vector.

[0013] For example, according to the digitized intelligent remote control method for drilling rigs according to an embodiment of the present application, the remote control server generates an optimal recommended decoding value of drilling pressure based on the drilling pressure cross-domain query matching feature vector, and generates a remote control instruction of drilling pressure based on the optimal recommended decoding value of drilling pressure and the drilling pressure value at the current time point, including: At the remote control server, the drilling pressure cross-domain query matching feature vector is input into an optimal drilling pressure recommender based on a decoder to obtain an optimal drilling pressure recommended decoded value; The remote control server generates a drilling pressure remote control instruction based on the comparison between the optimal drilling pressure recommended decoded value and the drilling pressure value at the current time point.

[0014] For example, according to the digital-based intelligent remote control method for drilling rigs according to an embodiment of the present application, in response to the optimal drilling pressure recommended decoded value being greater than the drilling pressure value at the current time point, the drilling pressure remote control instruction is to reduce the drilling pressure value.

[0015] The embodiment of the present application also provides a digital-based intelligent remote control system for drilling rigs, which includes: A data acquisition module, used for collecting a time queue of drilling pressure value, a time queue of torque value, a time queue of pump pressure value, a time queue of mud flow, a time queue of mud density, a time queue of drilling depth, a time queue of temperature and a time queue of humidity through a sensor network; a transmission module, used for transmitting the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity to a remote control server through a wireless communication module; A timing encoding module is used for performing timing encoding on the remote control server respectively, on the remote control server, of the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity to obtain a timing-related implicit feature vector of the drilling pressure, a timing-related implicit feature vector of the torque, a timing-related implicit feature vector of the pump pressure, a timing-related implicit feature vector of the mud flow rate, a timing-related implicit feature vector of the mud density, a timing-related implicit feature vector of the drilling depth, a timing-related implicit feature vector of the temperature and a timing-related implicit feature vector of the temperature; A cross-domain attention gate query encoding module is used for, on the remote control server, using the drilling pressure time series associated implicit feature vector as a query feature vector and a feature matrix consisting of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector as an initial key matrix, and inputting the query feature vector and the initial key matrix into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector; The remote control instruction generation module is used to generate an optimal drilling pressure recommended decoding value based on the drilling pressure cross-domain query matching feature vector on the remote control server, and generate a drilling pressure remote control instruction based on the optimal drilling pressure recommended decoding value and the drilling pressure value at the current time point.

[0016] This application has at least the following technical effects: Compared with the prior art, the digitized drilling rig intelligent remote control method and control system of the embodiment of the present application monitors and collects drilling parameter data in real time through a sensor network, and transmits these drilling parameter data to a remote control server through a wireless communication network, so as to use data processing and analysis algorithms based on artificial intelligence and deep learning in the remote control server to perform time series analysis and collaborative correlation feature characterization on these drilling parameter data, so as to capture the contextual semantics between the time series features of each drilling parameter, and use the drilling pressure time series semantics as a query feature to query and match the optimal drilling pressure recommended decoding value from the remaining drilling parameter time series collaborative correlation features, and compare it with the drilling pressure value at the current time point to generate a drilling pressure remote control instruction, which helps to quickly adapt to changes in geological conditions during the drilling process and realize remote intelligent control of the drilling rig. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0018] Figure 1 A schematic diagram of the application architecture of the digitized drilling rig intelligent remote control method in an embodiment of the present application is shown; Figure 2 A flowchart of a drilling rig intelligent remote control method based on digitalization in an embodiment of the present application is shown; Figure 3 A flowchart of sub-step S540 of the digital-based intelligent remote control method for drilling rigs in an embodiment of the present application is shown; Figure 4 A structural schematic diagram of a digital-based drilling rig intelligent remote control system in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0020] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0021] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0022] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0023] Figure 1A schematic diagram of the application architecture of a digital drilling rig intelligent remote control method in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0024] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network. In some embodiments, the data exchanged through the network can be represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0025] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0026] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the program of the digitized drilling rig intelligent remote control method in the embodiment of the present application.

[0027] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the digital-based intelligent remote control methods for drilling rigs in the embodiments of the present application according to the obtained program instructions.

[0028] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0029] The following is a non-restrictive description of the digital-based intelligent remote control method for drilling rigs provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0030] The rapid development of digital technology has enabled various sensors and wireless communication technologies to be widely used in drilling operations. By deploying a variety of sensors on the drilling rig, key parameters in the drilling process can be monitored in real time, such as drilling pressure, torque, pump pressure, mud flow, mud density, drilling depth, temperature and humidity. The real-time collection and collaborative analysis of these data can not only help engineers understand the drilling status in a timely manner, but also provide data support for subsequent decision-making.

[0031] Based on this, Figure 2 The flowchart of the digitized drilling rig intelligent remote control method in the embodiment of the present application is shown. For example, the digitized drilling rig intelligent remote control method can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 2As shown, according to the embodiment of the present application, the intelligent remote control method for drilling rigs based on digitization includes the following steps: S510, collecting a time queue of drilling pressure values, a time queue of torque values, a time queue of pump pressure values, a time queue of mud flow, a time queue of mud density, a time queue of drilling depth, a time queue of temperature, and a time queue of humidity through a sensor network; S520, transmitting the time queue of drilling pressure values, the time queue of torque values, the time queue of pump pressure values, the time queue of mud flow, the time queue of mud density, the time queue of drilling depth, the time queue of temperature, and the time queue of humidity to a remote control server through a wireless communication module; S530, at the remote control server, respectively, time-series encoding the time queue of drilling pressure values, the time queue of torque values, the time queue of pump pressure values, the time queue of mud flow, the time queue of mud density, the time queue of drilling depth, the time queue of temperature, and the time queue of humidity on the remote control server to obtain a drilling pressure time-series associated implicit feature vector, a torque time-series associated implicit feature vector, and a time-series associated implicit feature vector. S540, in the remote control server, using the drilling pressure time series associated implicit feature vector as a query feature vector and the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector; S541, in the remote control server, using the drilling pressure time series associated implicit feature vector as a query feature vector and the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector A feature matrix consisting of a feature vector, an implicit feature vector associated with the drilling depth time series, an implicit feature vector associated with the temperature time series, and an implicit feature vector associated with the temperature time series is used as an initial key matrix, and the query feature vector and the initial key matrix are input into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector; S550, at the remote control server, based on the drilling pressure cross-domain query matching feature vector, an optimal drilling pressure recommended decoding value is generated, and based on the optimal drilling pressure recommended decoding value and the drilling pressure value at the current time point, a drilling pressure remote control instruction is generated.

[0032] Specifically, in the technical solution of the present application, a digital-based intelligent remote control method for a drilling rig is proposed, which can monitor and collect drilling parameter data in real time through a sensor network, including drilling pressure, torque, pump pressure, mud flow, mud density, drilling depth, temperature and humidity, and transmit these drilling parameter data to a remote control server through a wireless communication network, so that the remote control server can use data processing and analysis algorithms based on artificial intelligence and deep learning to perform time series analysis and collaborative correlation feature characterization on these drilling parameter data, so as to capture the contextual semantics between the time series features of each drilling parameter, and use the drilling pressure time series semantics as a query feature to query and match the optimal drilling pressure recommended decoding value from the remaining drilling parameter time series collaborative correlation features, and compare it with the drilling pressure value at the current time point to generate a drilling pressure remote control instruction, which can automatically control the drilling pressure parameters of the drilling rig based on the actual drilling status, thereby improving the intelligence level of remote drilling rig control, and helping to quickly adapt to changes in geological conditions during the drilling process and reduce safety accidents.

[0033] Specifically, in the technical solution of the present application, first, a time queue of drilling pressure values, a time queue of torque values, a time queue of pump pressure values, a time queue of mud flow, a time queue of mud density, a time queue of drilling depth, a time queue of temperature and a time queue of humidity are collected through a sensor network, and the time queue of drilling pressure values, the time queue of torque values, the time queue of pump pressure values, the time queue of mud flow, the time queue of mud density, the time queue of drilling depth, the time queue of temperature and the time queue of humidity are transmitted to a remote control server through a wireless communication module, thereby providing a basis for remote control of the drilling rig.

[0034] However, in the remote control server, the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity are respectively input into the sequence encoder based on the Bi-LSTM model for encoding, so as to respectively extract the temporal context long-distance dependence implicit correlation feature information of the drilling pressure, torque, pump pressure, mud flow, mud density, drilling depth, temperature and humidity in the time dimension, so as to obtain the drilling pressure time series correlation implicit feature vector, torque time series correlation implicit feature vector, pump pressure time series correlation implicit feature vector, mud flow time series correlation implicit feature vector, mud density time series correlation implicit feature vector, drilling depth time series correlation implicit feature vector, temperature time series correlation implicit feature vector and temperature time series correlation implicit feature vector.

[0035] Correspondingly, in step S530, the remote control server performs time series encoding on the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature, and the time queue of the humidity, respectively, to obtain a drilling pressure time series associated implicit feature vector, a torque time series associated implicit feature vector, a pump pressure time series associated implicit feature vector, a mud flow time series associated implicit feature vector, a mud density time series associated implicit feature vector, a drilling depth time series associated implicit feature vector, a temperature time series associated implicit feature vector, and a temperature time series associated implicit feature vector. The method comprises: inputting the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity into a sequence encoder based on a Bi-LSTM model respectively to obtain the drilling pressure time series associated implicit feature vector, the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector.

[0036] Further, considering that each of the time-series-associated implicit feature vectors of the drilling pressure time-series-associated implicit feature vector, the torque time-series-associated implicit feature vector, the pump pressure time-series-associated implicit feature vector, the mud flow time-series-associated implicit feature vector, the mud density time-series-associated implicit feature vector, the drilling depth time-series-associated implicit feature vector, the temperature time-series-associated implicit feature vector and the temperature time-series-associated implicit feature vector respectively contains the time-series implicit association features of different drilling parameter data in the time dimension, these different drilling parameter time-series features have semantic connections and potential influences of correlation. Therefore, in order to use the drilling pressure time-series feature to query and match the relevant semantics from other drilling parameter time-series collaborative features, so as to perform subsequent drilling pressure recommendation control, it is necessary to explore the collaborative association relationship between the other drilling parameter time-series features and the contextual implicit matching semantics between the drilling pressure time-series features. However, the traditional context encoding method can usually only focus on the association relationship between the query vector and the key vector, while ignoring the potential association between the other time-series-associated implicit feature vectors as key vectors. If the implicit correlation information contained in the key matrix composed of these time-series-related implicit feature vectors can be utilized, the accuracy of the cross-domain query matching coding can be significantly improved. Based on this, in the remote control server, the drilling pressure time-series-related implicit feature vector is further used as the query feature vector and the feature matrix composed of the torque time-series-related implicit feature vector, the pump pressure time-series-related implicit feature vector, the mud flow time-series-related implicit feature vector, the mud density time-series-related implicit feature vector, the drilling depth time-series-related implicit feature vector, the temperature time-series-related implicit feature vector and the temperature time-series-related implicit feature vector is used as the initial key matrix, and the query feature vector and the initial key matrix are input into the cross-domain attention query encoder to obtain the drilling pressure cross-domain query matching feature vector.

[0037] Specifically, the cross-domain focus query encoder first performs a linear transformation on each of the time-series-associated implicit feature vectors, including the torque time-series-associated implicit feature vector, the pump pressure time-series-associated implicit feature vector, the mud flow time-series-associated implicit feature vector, the mud density time-series-associated implicit feature vector, the drilling depth time-series-associated implicit feature vector, the temperature time-series-associated implicit feature vector, and the temperature time-series-associated implicit feature vector, to adjust their feature representation dimensions so as to align them with the feature dimensions of the query feature vector, so as to ensure that other time-series-associated implicit feature vectors and the drilling pressure time-series-associated implicit feature vector can be compared in the same feature space.

[0038] Next, the feature distribution difference between the query feature vector and each row vector (key vector) in the initial key matrix is ​​calculated to obtain a set of difference energy coefficients. That is, the implicit correlation feature of the drilling pressure time series is used as a soft anchor reference point and the feature distribution energy difference coefficient is used to measure the feature distribution difference between the query feature vector and each key vector in the initial key matrix to obtain a set of difference energy coefficients. Specifically, considering that the query feature vector is a common soft anchor reference point, the set of distribution energy difference coefficients as a whole also contains the local correlation information between the semantic encoding features of other drilling parameter time series. Therefore, these obtained difference energy coefficients reflect the similarity between the query vector and the key vector, that is, the synergistic relationship between the other drilling parameter time series semantics and the drilling pressure time series semantics in terms of drilling status, and at the same time contain the local correlation information between the key vectors.

[0039] Then, the characteristic distribution difference energy coefficient between the drilling pressure timing characteristics and other drilling parameter timing characteristics is compared with the predetermined threshold. For the case where the value is less than the threshold, the row vector in the key matrix is ​​updated using the mean vector of the query feature vector and the corresponding row vector. The update operation here strengthens the drilling parameter timing semantics that are more associated with the query feature vector. Through the local association enhancement mechanism, the pertinence and key matrix are improved, thereby providing a basis for the subsequent return drilling pressure control task.

[0040] Furthermore, the drilling pressure time series associated implicit feature vector is used as the query vector, each updated row vector in the update key matrix is ​​used as the key vector, and the drilling pressure time series semantic value vector is used as the value vector. The cross-domain attention gate query module based on the Transformer structure is used to combine the query vector, key vector and value vector to perform feature self-attention calculation to achieve cross-domain information integration. The effect of the cross-domain attention mechanism is to allow the model to flow and integrate information between different semantic feature domains, improve the richness of the semantic association expression between the drilling pressure time series features and other drilling parameter time series features, and enhance the ability to express features with higher correlation with the drilling pressure time series features in the collaborative semantics of other drilling parameter time series. Finally, the final drilling pressure cross-domain query matching feature vector is obtained by calculating the positional mean of the sequence of drilling parameter temporal semantic cross-domain query attention vectors. Its purpose is to comprehensively consider all cross-domain query feature matching information related to the drilling pressure temporal semantic expression in the temporal collaborative correlation features between drilling pressure, torque, pump pressure, mud flow, mud density, drilling depth, temperature and humidity, so as to provide a basis for the subsequent drilling pressure recommendation adjustment and the generation of drilling pressure remote control instructions.

[0041] Accordingly, in step S540, if Figure 3As shown, in the remote control server, the drilling pressure time series associated implicit feature vector is used as a query feature vector and a feature matrix composed of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector is used as an initial key matrix, and the query feature vector and the initial key matrix are input into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector, including: S54 1, respectively, after linear transformation of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector, the linearly transformed time series associated implicit feature vectors are matrixed to obtain the initial key matrix; S542, calculating the characteristic distribution difference energy coefficient between the drilling pressure time series associated implicit feature vector and each row vector in the initial key matrix to obtain the drilling parameter S543, comparing each drilling parameter time series semantic feature distribution difference energy coefficient in the set of drilling parameter time series semantic feature distribution difference energy coefficients with a predetermined threshold, and in response to the drilling parameter time series semantic feature distribution difference energy coefficient being less than the predetermined threshold, using the mean vector between the row vector in the initial key matrix corresponding to the drilling parameter time series semantic feature distribution difference energy coefficient and the drilling pressure time series associated implicit feature vector as the updated row vector to obtain an updated key matrix; S544, using the value embedding matrix and the value embedding vector to embed the drilling pressure time series associated implicit feature vector The implicit feature vector is processed to obtain a drilling pressure time series semantic value vector; S545, the drilling pressure time series associated implicit feature vector is used as a query vector, each updated row vector in the update key matrix is ​​used as a key vector and the drilling pressure time series semantic value vector is used as a value vector, and the query vector, the key vector and the value vector are input into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of drilling parameter time series semantic cross-domain query attention vectors; S546, the positional mean vector of the sequence of drilling parameter time series semantic cross-domain query attention vectors is calculated to obtain the drilling pressure cross-domain query matching feature vector.

[0042] Wherein, in step S541, the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector are linearly transformed, and the linearly transformed time series associated implicit feature vectors are matrixed to obtain the initial key matrix, including: ... The time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and each time series associated implicit feature vector in the temperature time series associated implicit feature vector are multiplied by the corresponding linear weight matrix respectively and then added with the linear bias vector by position to obtain a plurality of time series associated implicit feature vectors after the linear transformation; the plurality of time series associated implicit feature vectors after the linear transformation are matrixed to obtain the initial key matrix.

[0043] Wherein, in step S542, the feature distribution difference energy coefficient between the drilling pressure time series associated implicit feature vector and each row vector in the initial key matrix is ​​calculated to obtain a set of drilling parameter time series semantic feature distribution difference energy coefficients, including: respectively calculating the position difference, position point multiplication and position addition between the drilling pressure time series associated implicit feature vector and the row vector in the initial key matrix to obtain the drilling pressure time series semantics-drilling parameter time series semantics difference feature vector, the drilling pressure time series semantics-drilling parameter time series semantics point multiplication feature vector and the drilling pressure time series semantics-drilling parameter time series semantics addition feature vector; cascading the drilling pressure time series semantics-drilling parameter time series semantics difference feature vector, the drilling pressure time series semantics-drilling parameter time series semantics point multiplication feature vector and the drilling pressure time series semantics-drilling parameter time series semantics addition feature vector, and processing them through a one-dimensional convolution layer, and performing maximum pooling processing on the obtained feature vector to obtain the drilling pressure time series semantics-drilling parameter time series semantics. The invention relates to a semantic feature association representation vector; extracts the maximum feature value in the drilling pressure time series semantics-drilling parameter time series semantic feature association representation vector to obtain the drilling pressure time series semantics-drilling parameter time series semantic association maximum representation value; calculates the mean and variance of the row vectors in the initial key matrix to obtain the drilling parameter time series semantic feature mean and the drilling parameter time series semantic feature variance; adds the drilling parameter time series semantic feature variance to the hyperparameter to obtain the first feature distribution difference energy factor; calculates the square of the difference between the drilling pressure time series semantics-drilling parameter time series semantic association maximum representation value and the drilling parameter time series semantic feature mean to obtain the semantic feature difference value; multiplies the drilling parameter time series semantic feature variance by a constant two and then adds it to the semantic feature difference value and the hyperparameter to obtain the second feature distribution difference energy factor; divides the first feature distribution difference energy factor by the second feature distribution difference energy factor to obtain the drilling parameter time series semantic feature distribution difference energy coefficient.

[0044] Among them, in step S544, the drilling pressure timing-associated implicit feature vector is processed using a value embedding matrix and a value embedding vector to obtain a drilling pressure timing semantic value vector, including: calculating the matrix multiplication between the drilling pressure timing-associated implicit feature vector and the value embedding matrix, and then adding the vector to the value embedding vector by position to obtain the drilling pressure timing semantic value vector.

[0045] Among them, in step S545, the drilling pressure time series associated implicit feature vector is used as the query vector, each updated row vector in the update key matrix is ​​used as the key vector, and the drilling pressure time series semantic value vector is used as the value vector. The query vector, the key vector and the value vector are input into a cross-domain attention gate query module based on the Transformer structure to obtain a sequence of drilling parameter time series semantic cross-domain query attention vectors, including: after calculating the product between the query vector and the transposed vector of the key vector, the obtained query-key feature association value is divided by the square root of the length of the key vector to obtain a query-key feature association coefficient; the query-key feature association coefficient is input into a softmax function to obtain a query-key feature association weight value; the query-key feature association weight value is used as a weight to weight the transposed vector of the value vector to obtain a drilling parameter time series semantic cross-domain query attention vector.

[0046] In one example, in the remote control server, the drilling pressure time series associated implicit feature vector is used as a query feature vector and a feature matrix composed of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector is used as an initial key matrix, and the query feature vector and the initial key matrix are input into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector, including: using the drilling pressure time series associated implicit feature vector as a query feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector as a query feature vector. The associated implicit feature vector is used as a query feature vector and a feature matrix consisting of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector is used as an initial key matrix, and the query feature vector and the initial key matrix are input into the cross-domain attention query encoder to be processed with the following cross-domain attention gate query formula to obtain the drilling pressure cross-domain query matching feature vector; wherein, the cross-domain attention gate query formula is: in, are respectively the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and each of the time series associated implicit feature vectors, is a sequence of temporally associated implicit eigenvectors, is the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector. time series associated implicit feature vectors, is the linear weight matrix, is the linear bias vector, is a linear transformation, For the said The linearly transformed temporal correlation implicit feature vector corresponding to the temporal correlation implicit feature vector is are the time series associated implicit feature vectors after linear transformation in the sequence of time series associated implicit feature vectors after linear transformation, is the initial bond matrix, is the query feature vector, , and represent positional subtraction, positional dot multiplication, and positional addition, respectively. Indicates cascade processing, is a one-dimensional convolution operation, is the maximum pooling operation, For the said The feature association representation vector corresponding to the temporal association implicit feature vector, Indicates the maximum value operation. and They are respectively The feature mean and feature variance corresponding to the implicit feature vector of the time series association, is a hyperparameter, For the said The characteristic distribution difference energy coefficient corresponding to the time series associated implicit characteristic vector, is the predetermined threshold, For the said The updated temporal correlation latent feature vector corresponding to the temporal correlation latent feature vector, The updated key matrix is ​​obtained by matrixing the implicit eigenvectors associated with each updated time series. is the value embedding matrix, is the value embedding vector, is the value vector corresponding to the query feature vector, for function, For transposition, is the length of each updated temporal association implicit feature vector, For the said The cross-domain query attention vector corresponding to the temporal correlation implicit feature vector, is the number of vectors in the sequence of the temporal association implicit feature vectors, A feature vector is matched for the drilling pressure cross-domain query.

[0047] In the remote control server, the drilling pressure cross-domain query matching feature vector is further input into the optimal drilling pressure recommender based on the decoder to obtain the optimal drilling pressure recommended decoded value. Specifically, the cross-domain query matching feature of the drilling pressure time series semantics is used to perform decoding regression, so as to decode the optimal drilling pressure recommended decoded value, and compare it with the drilling pressure value at the current time point to generate a drilling pressure remote control instruction. In particular, in response to the optimal drilling pressure recommended decoded value being greater than the drilling pressure value at the current time point, the drilling pressure remote control instruction is to reduce the drilling pressure value, which can automatically control the drilling pressure parameters of the drilling rig based on the actual drilling status, thereby improving the intelligence level of remote drilling rig control, helping to quickly adapt to changes in geological conditions during the drilling process and reduce safety accidents.

[0048] Accordingly, in step S550, the remote control server generates an optimal drilling pressure recommended decoded value based on the drilling pressure cross-domain query matching feature vector, and generates a drilling pressure remote control instruction based on the optimal drilling pressure recommended decoded value and the drilling pressure value at the current time point, including: at the remote control server, inputting the drilling pressure cross-domain query matching feature vector into the optimal drilling pressure recommender based on the decoder to obtain the optimal drilling pressure recommended decoded value; at the remote control server, generating a drilling pressure remote control instruction based on the comparison between the optimal drilling pressure recommended decoded value and the drilling pressure value at the current time point. Finally, in response to the optimal drilling pressure recommended decoded value being greater than the drilling pressure value at the current time point, the drilling pressure remote control instruction is to reduce the drilling pressure value.

[0049] In a preferred example, considering that the drilling pressure time series association implicit feature vector, torque time series association implicit feature vector, pump pressure time series association implicit feature vector, mud flow time series association implicit feature vector, mud density time series association implicit feature vector, drilling depth time series association implicit feature vector, temperature time series association feature vector and humidity time series association implicit feature vector respectively represent the short-range and long-range bidirectional time series association characteristics of the drilling pressure value, torque value, pump pressure value, mud flow value, mud density value, drilling depth value, temperature value and humidity value, therefore, when performing cross-domain attention gate query encoding based on key matrix local association enhancement, due to the difference in local association enhancement of the time series feature key matrix of heterogeneous data and the superposition of cross-domain attention gate query encoding differences, the drilling pressure cross-domain query matching feature vector will have the difference in the spatial structure of the query interaction distribution of the encoded semantic feature diversity, which affects the convergence consistency of the decoder, thereby affecting the accuracy of the optimal drilling pressure recommended decoding value obtained by decoding and regressing the optimal drilling pressure recommender based on the decoder of the drilling pressure cross-domain query matching feature vector input.

[0050] Based on this, in the preferred example, the drilling pressure cross-domain query matching feature vector is input into the optimal drilling pressure recommender based on the decoder to obtain the optimal drilling pressure recommended decoded value, which specifically includes: The absolute value sum and the square root of the square sum of all eigenvalues ​​of the drilling pressure cross-domain query matching eigenvector are calculated to obtain the first drilling pressure cross-domain query matching spatial structure value and the second drilling pressure cross-domain query matching spatial structure value, that is: in, The first one represents the drilling pressure cross-domain query matching feature vector eigenvalues, Indicates the first drilling pressure cross-domain query matching spatial structure value, Indicates the second drilling pressure cross-domain query matching spatial structure value; Determine the number of eigenvalues ​​of all eigenvalues ​​of the drilling pressure cross-domain query matching eigenvector , that is, the length of the drilling pressure cross-domain query matching feature vector For each eigenvalue of the drilling pressure cross-domain query matching eigenvalue vector, the first drilling pressure cross-domain query matching spatial structure value minus the product of the eigenvalue and the number of eigenvalues ​​is calculated to obtain the first drilling pressure cross-domain query matching long-range dependency value ,in, Indicates the first drilling pressure cross-domain query matching spatial structure value, The first one represents the drilling pressure cross-domain query matching feature vector eigenvalues, represents the number of eigenvalues ​​of all eigenvalues ​​of the drilling pressure cross-domain query matching eigenvector, Indicates the first drilling pressure cross-domain query matching long-range dependency value; Calculate the second drilling pressure cross-domain query matching long-range dependency value obtained by multiplying the square root of the number of eigenvalues ​​by the product of the eigenvalues ​​minus the second drilling pressure cross-domain query matching spatial structure value ,in, Indicates the second drilling pressure cross-domain query matching spatial structure value, The first one represents the drilling pressure cross-domain query matching feature vector eigenvalues, represents the number of eigenvalues ​​of all eigenvalues ​​of the drilling pressure cross-domain query matching eigenvector, Indicates the second drilling pressure cross-domain query matching long-range dependency value; The index value calculated by taking the first drilling pressure cross-domain query matching long-range dependency value as the exponent of the natural constant and the inverse of the second drilling pressure cross-domain query matching long-range dependency value are weighted summed to obtain the optimized eigenvalue corresponding to each eigenvalue ,in, Indicates the first drilling pressure cross-domain query matching long-range dependency value, Indicates that the second drilling pressure cross-domain query matches the long-range dependency value, represents a natural constant, and represents the weighted hyperparameter, Indicates the optimized eigenvalue corresponding to each eigenvalue; The optimized feature values ​​are combined into an optimized drilling pressure cross-domain query matching feature vector, and the optimized drilling pressure cross-domain query matching feature vector is input into an optimal drilling pressure recommender based on a decoder to obtain an optimal drilling pressure recommended decoded value.

[0051] That is, in view of the possible spatial structure loss of the feature set of the drilling pressure cross-domain query matching feature vector in the high-dimensional space, which causes the weight of the decoder to implicitly infer the spatial structure information based on the feature, resulting in inconsistent convergence. By establishing a long-distance feature dependency relationship based on the spatial structure representation of the drilling pressure cross-domain query matching feature vector relative to the overall feature scale of the drilling pressure cross-domain query matching feature vector, the feature local connectivity of the drilling pressure cross-domain query matching feature vector is established, and the spatial ambiguity information of the object feature value is captured by predicting the unstructured feature value points of the drilling pressure cross-domain query matching feature vector, thereby improving the spatial inductive deviation perception ability of the feature set of the drilling pressure cross-domain query matching feature vector, improving the convergence consistency of the decoder, and improving the accuracy of the optimal drilling pressure recommended decoding value obtained by decoding and regressing the optimal drilling pressure recommender based on the decoder of the drilling pressure cross-domain query matching feature vector input. In this way, the drilling pressure parameters of the drilling rig can be more accurately controlled based on the actual drilling state, thereby improving the intelligence level of remote drilling rig control, helping to quickly adapt to changes in geological conditions during the drilling process, and reducing safety accidents.

[0052] Based on the above embodiments, see Figure 4, which is a structural diagram of a digital-based drilling rig intelligent remote control system 800 in an embodiment of the present application. The digital-based intelligent remote control system 800 for drilling rigs includes: a data acquisition module 810, which is used to collect a time queue of drilling pressure values, a time queue of torque values, a time queue of pump pressure values, a time queue of mud flow, a time queue of mud density, a time queue of drilling depth, a time queue of temperature, and a time queue of humidity through a sensor network; a transmission module 820, which is used to transmit the time queue of drilling pressure values, the time queue of torque values, the time queue of pump pressure values, the time queue of mud flow, the time queue of mud density, the time queue of drilling depth, the time queue of temperature, and the time queue of humidity to a remote control server through a wireless communication module; a timing encoding module 830, which is used to perform timing encoding on the remote control server, respectively, on the remote control server, for the time queue of drilling pressure values, the time queue of torque values, the time queue of pump pressure values, the time queue of mud flow, the time queue of mud density, the time queue of drilling depth, the time queue of temperature, and the time queue of humidity to obtain a timing-related implicit feature vector of drilling pressure, a timing-related implicit feature vector of torque, a time queue of pump pressure values, a time queue of mud flow, a time queue of mud density, a time queue of drilling depth, a time queue of temperature, and a time queue of humidity. The invention relates to an implicit feature vector of drilling pressure time series association, an implicit feature vector of mud flow time series association, an implicit feature vector of mud density time series association, an implicit feature vector of drilling depth time series association, an implicit feature vector of temperature time series association and an implicit feature vector of temperature time series association; a cross-domain attention gate query encoding module 840, which is used to query the feature vector of drilling pressure time series association and the implicit feature vector of torque time series association, the implicit feature vector of pump pressure time series association, the implicit feature vector of mud flow time series association, the implicit feature vector of mud density time series association on the remote control server. The vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector as the initial key matrix, the query feature vector and the initial key matrix are input into the cross-domain attention query encoder to obtain the drilling pressure cross-domain query matching feature vector; the remote control instruction generation module 850 is used to generate an optimal drilling pressure recommended decoding value based on the drilling pressure cross-domain query matching feature vector on the remote control server, and generate a drilling pressure remote control instruction based on the optimal drilling pressure recommended decoding value and the drilling pressure value at the current time point.

[0053] Specifically, the specific functions and operations of each module in the above-mentioned digital-based drilling rig intelligent remote control system 800 have been referred to above. Figure 2 to Figure 3 The description of the digital-based drilling rig intelligent remote control method has been introduced in detail, and therefore, its repeated description will be omitted.

[0054] Based on the above embodiment, another exemplary embodiment of an electronic device is provided in the embodiment of the present application. In some possible implementations, the electronic device in the embodiment of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of the digital-based intelligent remote control method for drilling rigs in the above embodiment can be implemented when the processor executes the program.

[0055] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0056] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the digitized intelligent remote control method for drilling rigs according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc.

[0057] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the digitization-based intelligent remote control method for drilling rigs according to the embodiment of the present application.

[0058] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0059] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0060] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0061] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0062] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A digital-based intelligent remote control method for drilling rigs, characterized in that: include: Collecting a time queue of drilling pressure value, a time queue of torque value, a time queue of pump pressure value, a time queue of mud flow, a time queue of mud density, a time queue of drilling depth, a time queue of temperature and a time queue of humidity through a sensor network; Transmitting the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity to a remote control server through a wireless communication module; On the remote control server, respectively, for the remote control server, the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature, and the time queue of the humidity are time-series encoded to obtain a drilling pressure time-series associated implicit feature vector, a torque time-series associated implicit feature vector, a pump pressure time-series associated implicit feature vector, a mud flow time-series associated implicit feature vector, a mud density time-series associated implicit feature vector, a drilling depth time-series associated implicit feature vector, a temperature time-series associated implicit feature vector, and a temperature time-series associated implicit feature vector; In the remote control server, the drilling pressure time series associated implicit feature vector is used as a query feature vector and a feature matrix consisting of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector is used as an initial key matrix, and the query feature vector and the initial key matrix are input into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector; On the remote control server, an optimal drilling pressure recommended decoding value is generated based on the drilling pressure cross-domain query matching feature vector, and a drilling pressure remote control instruction is generated based on the optimal drilling pressure recommended decoding value and the drilling pressure value at the current time point.

2. The digitized intelligent remote control method for drilling rigs according to claim 1 is characterized in that: In the remote control server, respectively, the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity are time-series encoded to obtain a drilling pressure time-series associated implicit feature vector, a torque time-series associated implicit feature vector, a pump pressure time-series associated implicit feature vector, a mud flow time-series associated implicit feature vector, a mud density time-series associated implicit feature vector, a drilling depth time-series associated implicit feature vector, a temperature time-series associated implicit feature vector and a temperature time-series associated implicit feature vector, including: The time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity are respectively input into the sequence encoder based on the Bi-LSTM model to obtain the drilling pressure time series associated implicit feature vector, the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector.

3. The digitized intelligent remote control method for drilling rigs according to claim 2 is characterized in that: In the remote control server, the drilling pressure time series associated implicit feature vector is used as a query feature vector and a feature matrix consisting of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector is used as an initial key matrix, and the query feature vector and the initial key matrix are input into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector, including: After linearly transforming the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector respectively, the linearly transformed time series associated implicit feature vectors are matrixed to obtain the initial key matrix; Calculating the characteristic distribution difference energy coefficients between the WOB time series associated implicit characteristic vector and each row vector in the initial key matrix to obtain a set of drilling parameter time series semantic characteristic distribution difference energy coefficients; Comparing each drilling parameter time series semantic feature distribution difference energy coefficient in the set of drilling parameter time series semantic feature distribution difference energy coefficients with a predetermined threshold, and in response to the drilling parameter time series semantic feature distribution difference energy coefficient being less than the predetermined threshold, taking the mean vector between the row vector in the initial key matrix corresponding to the drilling parameter time series semantic feature distribution difference energy coefficient and the implicit feature vector associated with the drilling pressure time series as the updated row vector to obtain an updated key matrix; Using a value embedding matrix and a value embedding vector to process the drilling pressure time series associated implicit feature vector to obtain a drilling pressure time series semantic value vector; Taking the WOB time series associated implicit feature vector as a query vector, each updated row vector in the update key matrix as a key vector and the WOB time series semantic value vector as a value vector, the query vector, the key vector and the value vector are input into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of drilling parameter time series semantic cross-domain query attention vectors; The position-based mean vector of the sequence of the drilling parameter time series semantic cross-domain query focus vector is calculated to obtain the drilling pressure cross-domain query matching feature vector.

4. The digitized intelligent remote control method for drilling rigs according to claim 3 is characterized in that: After linear transformation is performed on the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector respectively, the linearly transformed time series associated implicit feature vectors are matrixed to obtain the initial key matrix, including: The torque time-series associated implicit feature vector, the pump pressure time-series associated implicit feature vector, the mud flow time-series associated implicit feature vector, the mud density time-series associated implicit feature vector, the drilling depth time-series associated implicit feature vector, the temperature time-series associated implicit feature vector and each of the time-series associated implicit feature vectors are multiplied by the corresponding linear weight matrix respectively and then added with the linear bias vector by position to obtain a plurality of linearly transformed time-series associated implicit feature vectors; The multiple linearly transformed temporally associated implicit feature vectors are arranged in a matrix to obtain the initial key matrix.

5. The digitized intelligent remote control method for drilling rigs according to claim 4 is characterized in that: Calculating the characteristic distribution difference energy coefficient between the implicit characteristic vector associated with the drilling pressure time series and each row vector in the initial key matrix to obtain a set of drilling parameter time series semantic characteristic distribution difference energy coefficients, including: Respectively calculating the positional difference, positional dot product and positional addition between the WOB timing-associated implicit feature vector and the row vector in the initial key matrix to obtain a WOB timing semantics-drilling parameter timing semantics differential feature vector, a WOB timing semantics-drilling parameter timing semantics dot product feature vector and a WOB timing semantics-drilling parameter timing semantics sum feature vector; The drilling pressure time series semantics-drilling parameter time series semantics differential feature vector, the drilling pressure time series semantics-drilling parameter time series semantics point product feature vector and the drilling pressure time series semantics-drilling parameter time series semantics sum feature vector are cascaded and processed through a one-dimensional convolution layer, and the obtained feature vector is subjected to maximum pooling processing to obtain the drilling pressure time series semantics-drilling parameter time series semantics feature association representation vector; Extracting the maximum value of the feature in the drilling pressure time series semantics-drilling parameter time series semantics feature association representation vector to obtain the maximum representation value of the drilling pressure time series semantics-drilling parameter time series semantics association; Calculating the mean and variance of the row vectors in the initial key matrix to obtain the mean and variance of the drilling parameter time series semantic features; Adding the drilling parameter time series semantic feature variance to the hyperparameter to obtain a first feature distribution difference energy factor; Calculate the square of the difference between the maximum representation value of the drilling pressure time series semantics-drilling parameter time series semantics association and the mean value of the drilling parameter time series semantic features to obtain a semantic feature difference value; The drilling parameter time series semantic feature variance is multiplied by a constant of two and then added to the semantic feature difference value and the hyperparameter to obtain a second feature distribution difference energy factor; The first feature distribution difference energy factor is divided by the second feature distribution difference energy factor to obtain the drilling parameter time series semantic feature distribution difference energy coefficient.

6. The digitized intelligent remote control method for drilling rigs according to claim 5 is characterized in that: The drilling pressure timing-associated implicit feature vector is processed using a value embedding matrix and a value embedding vector to obtain a drilling pressure timing-associated semantic value vector, including: calculating matrix multiplication between the drilling pressure timing-associated implicit feature vector and the value embedding matrix, and then adding the vector to the value embedding vector by position to obtain the drilling pressure timing-associated semantic value vector.

7. The digitized intelligent remote control method for drilling rigs according to claim 6 is characterized in that: The WOB time series associated implicit feature vector is used as a query vector, each updated row vector in the update key matrix is ​​used as a key vector, and the WOB time series semantic value vector is used as a value vector. The query vector, the key vector, and the value vector are input into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of drilling parameter time series semantic cross-domain query attention vectors, including: After calculating the product between the query vector and the transposed vector of the key vector, dividing the obtained query-key feature association value by the square root of the length of the key vector to obtain a query-key feature association coefficient; Inputting the query-key feature association coefficient into a softmax function to obtain a query-key feature association weight value; The query-key feature association weight value is used as a weight, and the transposed vector of the value vector is weighted to obtain a drilling parameter temporal semantic cross-domain query attention vector.

8. The digitized intelligent remote control method for drilling rigs according to claim 7 is characterized in that: The remote control server generates an optimal drilling pressure recommended decoded value based on the drilling pressure cross-domain query matching feature vector, and generates a drilling pressure remote control instruction based on the optimal drilling pressure recommended decoded value and the drilling pressure value at the current time point, including: At the remote control server, the drilling pressure cross-domain query matching feature vector is input into an optimal drilling pressure recommender based on a decoder to obtain an optimal drilling pressure recommended decoded value; The remote control server generates a drilling pressure remote control instruction based on the comparison between the optimal drilling pressure recommended decoded value and the drilling pressure value at the current time point.

9. The digitized intelligent remote control method for drilling rigs according to claim 8, characterized in that: In response to the optimal recommended decoding value of the bit weight being greater than the bit weight value at the current time point, the bit weight remote control instruction is to reduce the bit weight value.

10. The digital-based intelligent remote control system for drilling rigs is characterized by: include: A data acquisition module, used for collecting a time queue of drilling pressure value, a time queue of torque value, a time queue of pump pressure value, a time queue of mud flow, a time queue of mud density, a time queue of drilling depth, a time queue of temperature and a time queue of humidity through a sensor network; a transmission module, used for transmitting the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity to a remote control server through a wireless communication module; A timing encoding module is used for performing timing encoding on the remote control server respectively, on the remote control server, of the time queue of the drilling pressure value, the time queue of the torque value, the time queue of the pump pressure value, the time queue of the mud flow rate, the time queue of the mud density, the time queue of the drilling depth, the time queue of the temperature and the time queue of the humidity to obtain a timing-related implicit feature vector of the drilling pressure, a timing-related implicit feature vector of the torque, a timing-related implicit feature vector of the pump pressure, a timing-related implicit feature vector of the mud flow rate, a timing-related implicit feature vector of the mud density, a timing-related implicit feature vector of the drilling depth, a timing-related implicit feature vector of the temperature and a timing-related implicit feature vector of the temperature; A cross-domain attention gate query encoding module is used for, on the remote control server, using the drilling pressure time series associated implicit feature vector as a query feature vector and a feature matrix consisting of the torque time series associated implicit feature vector, the pump pressure time series associated implicit feature vector, the mud flow time series associated implicit feature vector, the mud density time series associated implicit feature vector, the drilling depth time series associated implicit feature vector, the temperature time series associated implicit feature vector and the temperature time series associated implicit feature vector as an initial key matrix, and inputting the query feature vector and the initial key matrix into a cross-domain attention query encoder to obtain a drilling pressure cross-domain query matching feature vector; The remote control instruction generation module is used to generate an optimal drilling pressure recommended decoding value based on the drilling pressure cross-domain query matching feature vector on the remote control server, and generate a drilling pressure remote control instruction based on the optimal drilling pressure recommended decoding value and the drilling pressure value at the current time point.

Images