Drilling machine intelligent scheduling system

Abstract

The invention discloses a drilling machine intelligent scheduling system, which is provided with a drilling machine demand prediction module, can realize prediction of drilling machine demand quantity, solves the problem of low efficiency of manual prediction of drilling machine demand quantity in the prior art, and is also provided with a drilling machine supply and demand matching module and a drilling machine scheduling module. The problems of large matching error and inaccurate scheduling depending on experience in the prior art are solved, manual adjustment by related scheduling personnel is facilitated, and a powerful guarantee is provided for safe and efficient operation of a drilling machine scheduling system.

Description

Technical Field

[0001] The invention relates to the technical field of drilling rig scheduling, and in particular to an intelligent drilling rig scheduling system. Background Art

[0002] The operation and management of drilling projects involve complex links, making the intelligent scheduling system a comprehensive management system that integrates equipment demand forecasting, control, low-cost and efficient operation, etc.

[0003] Domestic oilfield companies are trying to establish a framework for information transmission between production applications from various perspectives in oilfield development, but a systematic intelligent rig scheduling management system has not yet been formed. In addition, the current rig scheduling model still mainly relies on manual experience to predict the demand and model of rigs, and adopts the traditional manual summary and analysis method, which is inefficient and has a relatively high error rate. There is also a certain lag in the response of various departments, which cannot guarantee the high-quality rig scheduling work, affecting the organization of rig resources and the coordinated scheduling between regions. Summary of the invention

[0004] The purpose of the present invention is to provide a drilling rig intelligent scheduling system to solve the problems existing in the prior art.

[0005] The present invention is achieved through the following technical solutions:

[0006] A drilling rig intelligent scheduling system, comprising a control module, a drilling rig demand prediction module, a drilling rig supply and demand matching module and a drilling rig scheduling module;

[0007] The control module is used to receive historical drilling data within a statistical period and a drilling rig demand forecasting instruction input by a staff member through human-computer interaction, transmit the historical drilling data and the drilling rig demand forecasting instruction to the drilling rig demand forecasting module, receive the drilling rig demand forecasting result fed back by the drilling rig demand forecasting module, and transmit the drilling rig demand forecasting result to a terminal corresponding to the staff member;

[0008] The drilling rig demand prediction module is used to obtain the drilling rig demand prediction result according to the historical drilling data and the drilling rig demand prediction instruction transmitted by the control module, and transmit the drilling rig demand prediction result to the control module;

[0009] The control module is also used to receive the drilling rig demand data transmitted by multiple drilling rig demand parties and the drilling rig supply data transmitted by multiple drilling rig suppliers, and transmit all the drilling rig demand data and drilling rig supply data to the drilling rig supply and demand matching module, receive the drilling rig supply and demand matching result fed back by the drilling rig supply and demand matching module, and transmit the drilling rig supply and demand matching result to the drilling rig demand party and the drilling rig supplier;

[0010] The drilling rig supply and demand matching module is used to obtain the drilling rig supply and demand matching result according to the drilling rig demand data and the drilling rig supply data transmitted by the control module, and transmit the drilling rig supply and demand matching result to the control module;

[0011] The drilling rig scheduling module is used to plan a route for each drilling rig provided by the drilling rig supplier according to the drilling rig supply and demand matching result, and feed back the planned drilling rig route to the drilling rig supplier through the control module.

[0012] In a possible implementation, the system further includes a data preprocessing module, which is used to clean up duplicate data, incomplete data, and abnormal data in historical drilling data.

[0013] In a possible implementation, it further includes a scheduling risk prediction module, wherein the scheduling risk prediction module is used to perform scheduling risk prediction on each drilling rig provided by the drilling rig supplier, obtain prediction results, and feed the prediction results back to the control module.

[0014] In a possible implementation, it also includes a real-time monitoring and early warning module, which is used to perform real-time monitoring and analysis on the drilling rig supply and demand matching results through a dynamic programming algorithm to obtain real-time monitoring and analysis results, and transmit the real-time monitoring and analysis results to the control module so that the drilling rig demander can accept and view the real-time monitoring and analysis results; the real-time monitoring and analysis results include that the current drilling rig does not correspond to the drilling rig supply and demand matching results corresponding to the drilling rig or the current drilling rig corresponds to the drilling rig supply and demand matching results corresponding to the drilling rig.

[0015] In a possible implementation, the drilling rig demand prediction module includes a drilling number prediction unit, a drilling rig demand prediction unit, and a first data transmission unit;

[0016] The drilling number prediction unit is used to execute the drilling rig demand prediction instruction, predict the number of drilling rigs of different models in a year according to the historical drilling data transmitted by the control module, and obtain the drilling number prediction result of the drilling rig;

[0017] The drilling rig demand prediction unit is used to execute the drilling rig demand prediction instruction, predict the drilling rig demand according to the historical drilling data transmitted by the control module and the drilling rig drilling number prediction result, and obtain the drilling rig demand prediction result;

[0018] The first data transmission unit is used to transmit the drilling rig demand prediction result to the control module.

[0019] In a possible implementation, the drilling rig supply and demand matching module includes a supply and demand matching unit and a second data transmission unit;

[0020] The supply and demand matching unit is used to match t drilling rig supply and demand matching results of drilling rig suppliers for each drilling rig demander according to the drilling rig demand data and drilling rig supply data transmitted by the control module;

[0021] The second data transmission unit is used to transmit the drilling rig supply and demand matching result to the control module.

[0022] In a possible implementation, each drilling rig demander is matched with t drilling rig suppliers to obtain a drilling rig supply and demand matching result, including: using a KNN algorithm to match t drilling rig supply data for each drilling rig demand data, thereby matching t drilling rig suppliers for each drilling rig demander to obtain a drilling rig supply and demand matching result.

[0023] In a possible implementation, the drilling rig scheduling module includes a drilling rig route planning unit and a third data transmission unit;

[0024] The drilling rig route planning unit is used to determine the travel route of each drilling rig between the drilling rig demanders corresponding to the target drilling rig supply data for all target drilling rig supply data matched by each drilling rig provided by the drilling rig supplier by using a genetic algorithm to obtain a planned route for each drilling rig;

[0025] The third data transmission unit is used to feed back the planned route of each drilling rig to the drilling rig supplier corresponding to the drilling rig.

[0026] In a possible implementation, a genetic algorithm is used to determine the travel route of each drilling rig between the drilling rig demand parties corresponding to the target drilling rig supply data, and obtain the planned route of each drilling rig, including:

[0027] Assign different codes to n well sites corresponding to the drilling rigs, where the well sites are used to represent the objects to be constructed corresponding to the drilling rig demander;

[0028] According to the preset initial population size and the code corresponding to each well site, multiple individuals are randomly generated to obtain the population corresponding to the genetic algorithm;

[0029] For the population, the individuals in the population are updated by using the selection operation based on normal distribution, the sequential crossover operation and the exchange mutation operation to obtain the updated population;

[0030] When the maximum fitness value corresponding to an individual in the population is greater than the preset target or the number of updates reaches the maximum, the individual corresponding to the maximum fitness value is used as the travel route corresponding to the drilling rig to obtain the planned route of each drilling rig;

[0031] The fitness value corresponding to the individual is obtained through the fitness function, and the fitness function is the negative value of the objective function.

[0032] In a possible implementation, the scheduling risk prediction module includes a data preprocessing unit, an input data building unit, a model training unit, a risk value prediction unit, and a fourth data transmission unit;

[0033] The data preprocessing unit is used to normalize the drilling rig scheduling data in the historical drilling data to obtain the normalized drilling rig scheduling data, wherein the historical drilling data includes the drilling rig scheduling data and the risk value corresponding to the drilling rig scheduling data, and the drilling rig scheduling data includes the block to which the well site where the drilling rig works belongs, the drilling type, the well type, the drilling rig model, the drilling rig occupation period, the drilling rig transportation distance and the waiting period;

[0034] The input data construction unit is used to construct the normalized drilling rig scheduling data into input vector data of uniform data length;

[0035] The model training unit is used to use the input vector data as the input data corresponding to the risk prediction model constructed by the BRF neural network, use the corresponding risk value as the true value label corresponding to the risk prediction model, train the risk prediction model, and obtain the trained risk prediction model;

[0036] The risk value prediction unit is used to use the trained risk prediction model to predict the scheduling risk of each drilling rig provided by the drilling rig supplier to obtain a prediction result;

[0037] The fourth data transmission unit is used to feed back the prediction result to the control module.

[0038] The present invention provides an intelligent drilling rig scheduling system, which is provided with a drilling rig demand prediction module, which can realize the prediction of drilling rig demand and solve the problem of low efficiency of existing manual prediction of drilling rig demand. It is also provided with a drilling rig supply and demand matching module and a drilling rig scheduling module, which solves the problems of large matching errors and inaccurate scheduling in the existing technology relying on experience, facilitates manual adjustments by relevant scheduling personnel, and provides strong guarantee for the safe and efficient operation of the drilling rig scheduling system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of the structure of a drilling rig intelligent scheduling system provided by an embodiment of the present invention.

[0041] Among them, 101-control module, 102-drilling rig demand prediction module, 103-drilling rig supply and demand matching module, 104-drilling rig scheduling module. DETAILED DESCRIPTION

[0042] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0043] like Figure 1 As shown, an embodiment of the present invention provides a drilling rig intelligent scheduling system, including a control module 101, a drilling rig demand prediction module 102, a drilling rig supply and demand matching module 103 and a drilling rig scheduling module 104.

[0044] The control module 101 is used to receive historical drilling data within a statistical period and drilling rig demand forecast instructions input by the staff through human-computer interaction, transmit the historical drilling data and drilling rig demand forecast instructions to the drilling rig demand forecast module 102, and receive the drilling rig demand forecast results fed back by the drilling rig demand forecast module 102, and transmit the drilling rig demand forecast results to the corresponding terminal of the staff.

[0045] The control module 101 can serve as an interaction module between the drilling rig demander and the drilling rig supplier and the drilling rig intelligent scheduling system provided by the embodiment of the present invention, so that the staff can perform demand forecasting, supply and demand matching, and scheduling route planning through the control module 101.

[0046] The historical drilling data may be data pre-stored in a database. When a drilling demand forecast instruction is received from a staff member through human-computer interaction, the historical drilling data within a preset statistical period or the historical drilling data within a statistical period determined by the staff member is automatically pulled.

[0047] The drilling rig demand prediction module 102 is used to obtain a drilling rig demand prediction result according to the historical drilling data and the drilling rig demand prediction instruction transmitted by the control module 101 , and transmit the drilling rig demand prediction result to the control module 101 .

[0048] In the prior art, the drilling rig demand forecast results are mainly obtained manually, which has the problems of low efficiency and easy manual errors. Therefore, the present invention automatically obtains the drilling rig demand forecast results according to the historical drilling data and drilling rig demand forecast instructions transmitted by the control module 101, which can effectively improve the forecast efficiency and avoid the problem of easy manual errors.

[0049] The control module 101 is also used to receive drilling rig demand data transmitted by multiple drilling rig demanders and drilling rig supply data transmitted by multiple drilling rig suppliers, and transmit all the drilling rig demand data and drilling rig supply data to the drilling rig supply and demand matching module 103, receive the drilling rig supply and demand matching results fed back by the drilling rig supply and demand matching module 103, and transmit the drilling rig supply and demand matching results to the drilling rig demanders and drilling rig suppliers.

[0050] Each rig demander can transmit its rig demand data, and each rig supplier can transmit its rig supply data, and intelligent matching can be performed to avoid inaccuracies caused by experience matching and the influence of subjective opinions. It is worth noting that since the rig models and quantities used in each well site are not single, a well site may be a rig demander or a rig supplier, that is, it may demand one or more types of rigs and supply other types of rigs.

[0051] The drilling rig supply and demand matching module 103 is used to obtain the drilling rig supply and demand matching result according to the drilling rig demand data and the drilling rig supply data transmitted by the control module 101 , and transmit the drilling rig supply and demand matching result to the control module 101 .

[0052] By intelligently matching drilling rig demand data with drilling rig supply data, problems caused by experience matching can be effectively avoided.

[0053] The drilling rig scheduling module 104 is used to plan a route for each drilling rig provided by the drilling rig supplier according to the drilling rig supply and demand matching result, and feed back the planned drilling rig route to the drilling rig supplier through the control module 101.

[0054] Since the time when each well site uses the drilling rig may be different, the problem of drilling rig scheduling route planning is involved. For example: Class A drilling rig is located in well site B, and well sites C and D both need Class A drilling rigs, and the earliest start time and the latest start time are the same. Between the earliest start time and the latest start time, Class A drilling rig can be used by well site C (or well site D) first, and then by well site D (or well site C), so Class A drilling rigs need to be routed.

[0055] In a possible implementation, the system further includes a data preprocessing module, which is used to clean up duplicate data, incomplete data, and abnormal data in historical drilling data.

[0056] The complexity of data preprocessing research is reflected in the fact that it contains a wealth of strategies and technologies. It is necessary to study and judge the authenticity, accuracy, completeness, timeliness, etc. of the data source. Only by ensuring the quality of the data source can the role of big data processing and analysis be brought into play. The data collected from the real world are generally incomplete and inconsistent data, and data processing cannot be performed directly, or the processing results are unsatisfactory. The present invention needs to clean and process the data, which will greatly improve the quality of the data model and reduce the time required for actual mining. The original data comes from various business systems, and the indicator caliber does not match. There will always be inconsistent, repeated, incomplete, erroneous or abnormal (deviating from the expected value) data. After data cleaning such as removing noise and irrelevant data, the data from multiple data sources are combined and stored in a consistent data storage, and the original data is converted into a form suitable for data mining, and finally the data concept is hierarchical and discretized.

[0057] In a possible implementation, a scheduling risk prediction module is further included, and the scheduling risk prediction module is used to perform scheduling risk prediction on each drilling rig provided by the drilling rig supplier, obtain prediction results, and feed the prediction results back to the control module 101.

[0058] In a possible implementation, it also includes a real-time monitoring and early warning module, which is used to perform real-time monitoring and analysis on the drilling rig supply and demand matching results through a dynamic programming algorithm to obtain real-time monitoring and analysis results, and transmit the real-time monitoring and analysis results to the control module 101, so that the drilling rig demander can accept and view the real-time monitoring and analysis results; the real-time monitoring and analysis results include that the current drilling rig does not correspond to the drilling rig supply and demand matching results corresponding to the drilling rig, or the current drilling rig corresponds to the drilling rig supply and demand matching results corresponding to the drilling rig.

[0059] Due to the delay of the construction period or the possible change of the drilling rig demand, the drilling rig demand plan and the drilling rig guarantee plan are judged and fed back to determine whether the prediction can achieve the preset goal. If the plan data value is less than or equal to the actual number of available drilling rigs, it is considered that the preset goal can be achieved. Otherwise, the iterative calculation is repeated.

[0060] In this embodiment, after the production and operation drilling rig scheduling related personnel log in to the drilling rig intelligent scheduling system, they can retrieve the attribute information of the drilling rig scheduling supply and demand sides from the database (which can be the oilfield drilling engineering operation system), and perform drilling rig intelligent scheduling analysis, and send the scheduling plan to the decision maker through the control module. Combined with the feedback information from the decision maker, the information is identified and analyzed to generate a drilling rig scheduling decision plan.

[0061] Compared with the prior art, the present invention establishes a connection with the oilfield production and operation related system, greatly improving the decision-making efficiency and improving the utilization rate of system data. In addition, the control module 101 can be directly connected with the decision-makers and field staff, and the feedback information of the decision-makers and field staff can be collected to optimize and adjust the drilling rig scheduling decision plan, provide intelligent analysis, intelligent decision-making drilling rig scheduling management system, remote analysis, scientific assistance oilfield decision-making, and assist computer information system to manage oilfields intelligently.

[0062] Optionally, all the results data are divided into two categories according to the technical fields they belong to: the demand side of well site equipment and the supply side of available drilling rigs. The demand data of well site drilling rigs include the number of exploration and deployment wells, the block to which the well site belongs, the drilling type (conventional well / shale gas), the designed well depth, the completed drilling layer, the models of drilling rigs available to each unit, the number of models of drilling rigs available, the blocks where the drilling rigs are located, and the approval cycle; the drilling rig supply capacity data samples collected by the model include seven categories: the block to which the well site belongs, the drilling type, the well type, the drilling rig model, the drilling rig occupation period, the transportation distance, and the waiting period.

[0063] In a possible implementation, the drilling rig demand prediction module 102 includes a drilling number prediction unit, a drilling rig demand prediction unit, and a first data transmission unit;

[0064] The drilling number prediction unit is used to execute the drilling rig demand prediction instruction, and predict the number of drilling rigs of different models in a year according to the historical drilling data transmitted by the control module 101, and obtain the drilling number prediction result of the drilling rig as follows:

[0065]

[0066] in, represents the number of wells that need to be drilled by the X-type drilling rig within the forecast period j, α represents the proportion of the number of wells drilled by the X-type drilling rig within the forecast period j in the historical drilling data, and the forecast period j is the same length as the statistical period i, represents the number of wells drilled by X-type drilling rig in statistical period i;

[0067] The drilling rig demand prediction unit is used to execute the drilling rig demand prediction instruction, and predict the drilling rig demand according to the historical drilling data and the drilling rig drilling number prediction result transmitted by the control module 101, and obtain the drilling rig demand prediction result as follows:

[0068]

[0069] in, represents the demand for x-type drilling rigs in forecast period j, β represents the balance coefficient, ∑(A+B+L) represents the number of drilling rigs occupied in the sample period of x-type drilling rigs in statistical period i, m j represents the number of months in the forecast period j.

[0070] The first data transmission unit is used to transmit the drilling rig demand prediction result to the control module 101.

[0071] As shown in Table 1, this embodiment provides an example of a drilling rig demand prediction result:

[0072] Table 1

[0073]

[0074] In a possible implementation, the drilling rig supply and demand matching module 103 includes a supply and demand matching unit and a second data transmission unit;

[0075] The supply and demand matching unit is used to match t drilling rig supply and demand matching results of drilling rig suppliers for each drilling rig demander according to the drilling rig demand data and drilling rig supply data transmitted by the control module 101;

[0076] The second data transmission unit is used to transmit the drilling rig supply and demand matching result to the control module 101.

[0077] For specific required data, the text content of the data is expressed in a mathematically processable form, that is, as a point in a high-dimensional space. This point is a test sample, and the K-nearest neighbor classification algorithm is used to determine which category the sample belongs to.

[0078] Matching each rig demander with t rig suppliers to obtain a rig supply and demand matching result includes: using a KNN algorithm to match each rig demand data with t rig supply data, thereby matching each rig demander with t rig suppliers to obtain a rig supply and demand matching result. Thus, the staff can determine the final rig demander based on the rig supply and demand matching result.

[0079] Only by examining the distribution of points in high-dimensional space can we know the distribution characteristics of these points in space and then perform data matching. By mining data, we can collect, clean and integrate the available drilling rigs and regular demand data of each unit, analyze and construct the allocation route of available drilling rigs, establish the relationship network model between available drilling rigs and well site needs, and realize the intelligent docking of drilling rig supply and demand under batch data.

[0080] The matching degree can also be obtained to assist the staff to further determine the final matching result. According to the matching result of rig supply and demand, the matching degree between the rig demand data corresponding to the rig demander and the rig supply data corresponding to the rig supplier is determined as follows:

[0081]

[0082] Among them, m rRepresents the matching degree, r i' represents the number of correct texts that the allocator judges as the i'th category, a i' Indicates the actual number of texts in the i'th category.

[0083] This embodiment provides a matching degree example for the Guanqian 1 well. The drilling rig requirement time is January 20, 2021, and a ZJ50 drilling rig is required. The specific matching degree calculation is shown in Table 2.

[0084] Table 2

[0085]

[0086] Therefore, Leshan 1 Well is the most suitable matching result and can be used as the final matching result. Alternatively, all three matching results can be fed back to the drilling rig demander, and the final matching result will be determined by the staff of the drilling rig demander.

[0087] In a possible implementation, the drilling rig scheduling module 104 includes a drilling rig route planning unit and a third data transmission unit.

[0088] The drilling rig route planning unit is used to use a genetic algorithm to determine the travel route of each drilling rig between the drilling rig demanders corresponding to the target drilling rig supply data for all target drilling rig supply data matched by each drilling rig provided by the drilling rig supplier, so as to obtain the planned route of each drilling rig.

[0089] The third data transmission unit is used to feed back the planned route of each drilling rig to the drilling rig supplier corresponding to the drilling rig.

[0090] In a possible implementation, a genetic algorithm is used to determine the travel route of each drilling rig between the drilling rig demand parties corresponding to the target drilling rig supply data, and obtain the planned route of each drilling rig, including:

[0091] Different codes are assigned to n well sites corresponding to the drilling rigs, and the well sites are used to characterize the objects to be constructed corresponding to the drilling rig demander.

[0092] According to the preset initial population size and the code corresponding to each well site, multiple individuals are randomly generated to obtain the population corresponding to the genetic algorithm.

[0093] For the population, the selection operation based on normal distribution, the sequential crossover operation and the exchange mutation operation are used to update the individuals in the population to obtain the updated population.

[0094] When the maximum fitness value corresponding to an individual in the population is greater than the preset target or the number of updates reaches the maximum, the individual corresponding to the maximum fitness value is used as the travel route corresponding to the drilling rig to obtain the planned route of each drilling rig.

[0095] The fitness value corresponding to the individual is obtained through the fitness function, and the fitness function is the negative value of the objective function.

[0096] In this embodiment, the rig site orientation layout problem is transformed into a TSP with hard time constraints. The following objective function can be established (assuming starting from block 1):

[0097]

[0098] Where: l is the objective function value; d i,j is the distance between two well sites i and j, x ij is a 0-1 decision variable, k1j and k2j are the earliest and latest construction times allowed for well j, respectively, and t j is the actual start-up time of well j, and Z is the time penalty coefficient. In order to obtain the minimum value of l, a genetic algorithm is used for analysis.

[0099] The use of genetic algorithms to arrange the on-site dispatch routes of drilling rigs can optimize efficient and dynamic on-site drilling rig management, avoid construction delays caused by subjective judgments of managers during scheduling, and maximize consideration of various factors, such as the time constraints of construction overlaps. The optimization solution obtained is the optimal solution after the integration of multiple objective factors. Genetic algorithms can quickly solve practical problems, and the obtained construction sequence has a smaller objective function value than the actual construction sequence. Genetic algorithms are computational models that simulate the biological evolution process of Darwin genetic selection and natural elimination. For the drilling rig dispatch route, the designed algorithm is as follows:

[0100] (1) Parameter coding uses sequential number coding. When solving, the segment length is represented by the sequential construction of n well sites. This sequential number coding based on path representation requires that the genes encoded by the chromosome of an individual cannot be repeated, that is, a well must be constructed only once.

[0101] (2) The population is initialized using a random function, and the first generation of the population is generated based on the set initial population size and the number of wells input.

[0102] (3) Design of fitness function The purpose of using genetic algorithm is to find the chromosome with the highest fitness. The designed objective function is to obtain the minimum value. The model avoids complex scaling and only constructs the fitness function by reversing the sign of the objective function. That is, the maximum fitness function value corresponds to the minimum objective function value.

[0103] (4) Design of genetic operations Since we chose a path-based sequential number encoding and tried to make the solution to the problem reach global optimization, we adopted normal distribution-based selection (norm-GeomSelect), order-based crossover, and three swap mutation.

[0104] (5) Setting of termination conditions: The algorithm is terminated when the genetic generation number of the algorithm exceeds the set maximum genetic generation number; the algorithm is terminated when the improvement of the optimal solution is less than a given value.

[0105] In a possible implementation, the scheduling risk prediction module includes a data preprocessing unit, an input data building unit, a model training unit, a risk value prediction unit, and a fourth data transmission unit.

[0106] The data preprocessing unit is used to normalize the drilling rig scheduling data in the historical drilling data to obtain the normalized drilling rig scheduling data, the historical drilling data includes the drilling rig scheduling data and the risk value corresponding to the drilling rig scheduling data, the drilling rig scheduling data includes the block where the well site where the drilling rig works belongs, the drilling type, the well type, the drilling rig model, the drilling rig occupancy period, the drilling rig transportation distance and the waiting period.

[0107] The input data construction unit is used to construct the normalized drilling rig scheduling data into input vector data with uniform data length.

[0108] The model training unit is used to use input vector data as input data corresponding to a risk prediction model constructed by a BRF (Radial Basis Function) neural network, and use the corresponding risk value as a true value label corresponding to the risk prediction model to train the risk prediction model and obtain a trained risk prediction model.

[0109] The risk value prediction unit is used to use the trained risk prediction model to perform scheduling risk prediction on each drilling rig provided by the drilling rig supplier to obtain a prediction result.

[0110] The fourth data transmission unit is used to feed back the prediction result to the control module 101.

[0111] There are many indicators for early warning evaluation of drilling equipment scheduling, and there is some information duplication between the indicators. In order to more accurately evaluate the early warning of drilling equipment scheduling, a dynamic early warning and evaluation method for drilling equipment based on factor analysis and neural network is proposed. The results of the above-mentioned drilling equipment demand prediction model are used to predict the risk of oilfield drilling equipment support capabilities, and early warning of drilling equipment scheduling is performed. Factor analysis, RBF neural network, and ant colony algorithm are combined to comprehensively evaluate the support capabilities of oilfield drilling equipment and provide early warnings.

[0112] When modeling and evaluating the risk change law of drilling rig equipment scheduling, it is necessary to first design a scientific and objective drilling rig equipment scheduling risk evaluation index system. The embodiment of the present invention designs an index system as shown in Table 3.

[0113] Table 3

[0114]

[0115] As can be seen from Table 3, there are many indicators for rig equipment scheduling risk assessment, and there are direct or indirect correlations between these indicators. Assuming there are k indicators in total, the indicators have different dimensions, which makes their values ​​differ greatly, affecting the convergence efficiency of RBF neural network learning. For this reason, they are preprocessed:

[0116]

[0117] In the formula, ww i represents the original value of the rig equipment scheduling risk assessment index, w i Represents the processed drilling rig equipment scheduling risk assessment index value.

[0118] Use m (m < k) to represent the original k drilling equipment scheduling risk assessment indicators, and the establishment form is as follows:

[0119]

[0120]

[0121] In the formula, e i represents a special factor, f i represents the common factor, a ij Represents the loading of the common factor.

[0122] Expressing the formula in matrix form gives:

[0123] W=AF+ε

[0124] in:

[0125]

[0126] Solving the above equation can be transformed into:

[0127]

[0128] Remember AA T The eigenvector corresponding to the largest eigenvalue greater than λ1 is b1, that is, Similarly, we can get In this way, multiple eigenvectors are obtained. The factor value is solved to obtain the number of risk assessment indicators:

[0129]

[0130] Let W = [w1,w2,...w n ] T As input vector, the radial vector is U = [u1,u2,...,u j ] T , then:

[0131]

[0132] in, is the center vector of node j, and:

[0133] x ji (k) = x ji (k-1)+ηΔx ji +α(x ji (k-1)-...-x ji (kp))

[0134] Among them, η is the learning efficiency, α is the momentum factor, and:

[0135]

[0136] Among them, b j is the base width parameter. The network weight vector is V = [v1, v2, ...] T , and there are:

[0137] V j (k) = v j (k-1)+η(yout(k)-ym(k))h j +α(v j (k-1)-…-v j (kp))

[0138] The output values ​​are:

[0139] y m (k)=v1h1+v2h2+...+v p h p

[0140] The risk prediction process in the present invention is as follows:

[0141] 1) Analyze the workflow and characteristics of traditional oilfield drilling rig scheduling and establish a drilling rig equipment scheduling risk indicator system.

[0142] 2) Collect the corresponding rig equipment scheduling risk indicator data and remove some erroneous data points to ensure the performance of rig equipment scheduling risk assessment.

[0143] 3) Use factor analysis to process the risk assessment indicators of drilling rig equipment scheduling, select the optimal indicators, and eliminate useless information.

[0144] 4) Score the samples according to each set of drilling rig equipment scheduling risk assessment index values ​​to obtain the drilling rig equipment scheduling risk value. It is worth noting that the risk value corresponding to the training sample can be input by human-computer interaction or can be data pre-stored in the database.

[0145] 5) The risk value and the corresponding rig equipment scheduling risk assessment index data are combined into a test sample set, and the oilfield rig support capability parameters and index data are used as training samples.

[0146] 6) The risk prediction model is trained using the training samples of drilling rig equipment scheduling risk assessment, and the algorithm-related parameters are optimized through the ant colony algorithm to obtain the trained risk prediction model. The trained risk prediction model is then used to predict the scheduling risk of each drilling rig provided by the drilling rig supplier.

[0147] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A drilling rig intelligent scheduling system, characterized in that: It includes control module, drilling rig demand prediction module, drilling rig supply and demand matching module and drilling rig scheduling module; The control module is used to receive historical drilling data within a statistical period and a drilling rig demand forecasting instruction input by a staff member through human-computer interaction, transmit the historical drilling data and the drilling rig demand forecasting instruction to the drilling rig demand forecasting module, receive the drilling rig demand forecasting result fed back by the drilling rig demand forecasting module, and transmit the drilling rig demand forecasting result to a terminal corresponding to the staff member; The drilling rig demand prediction module is used to obtain the drilling rig demand prediction result according to the historical drilling data and the drilling rig demand prediction instruction transmitted by the control module, and transmit the drilling rig demand prediction result to the control module; The control module is also used to receive the drilling rig demand data transmitted by multiple drilling rig demand parties and the drilling rig supply data transmitted by multiple drilling rig suppliers, and transmit all the drilling rig demand data and drilling rig supply data to the drilling rig supply and demand matching module, receive the drilling rig supply and demand matching result fed back by the drilling rig supply and demand matching module, and transmit the drilling rig supply and demand matching result to the drilling rig demand party and the drilling rig supplier; The drilling rig supply and demand matching module is used to obtain the drilling rig supply and demand matching result according to the drilling rig demand data and the drilling rig supply data transmitted by the control module, and transmit the drilling rig supply and demand matching result to the control module; The drilling rig scheduling module is used to plan a route for each drilling rig provided by the drilling rig supplier according to the drilling rig supply and demand matching result, and feed back the planned drilling rig route to the drilling rig supplier through the control module.

2. The drilling rig intelligent scheduling system according to claim 1, characterized in that: It also includes a data preprocessing module, which is used to clean up duplicate data, incomplete data and abnormal data in historical drilling data.

3. The drilling rig intelligent scheduling system according to claim 1, characterized in that: It also includes a scheduling risk prediction module, which is used to predict the scheduling risk of each drilling rig provided by the drilling rig supplier, obtain the prediction result, and feed the prediction result back to the control module.

4. The drilling rig intelligent scheduling system according to claim 1, characterized in that: It also includes a real-time monitoring and early warning module, which is used to perform real-time monitoring and analysis on the drilling rig supply and demand matching results through a dynamic programming algorithm to obtain real-time monitoring and analysis results, and transmit the real-time monitoring and analysis results to the control module so that the drilling rig demander can accept and view the real-time monitoring and analysis results; The real-time monitoring and analysis result includes that the current drilling rig does not correspond to the drilling rig supply and demand matching result corresponding to the drilling rig or the current drilling rig corresponds to the drilling rig supply and demand matching result corresponding to the drilling rig.

5. The drilling rig intelligent scheduling system according to claim 1, characterized in that: The drilling rig demand prediction module includes a drilling number prediction unit, a drilling rig demand prediction unit and a first data transmission unit; The drilling number prediction unit is used to execute the drilling rig demand prediction instruction, predict the number of drilling rigs of different models in a year according to the historical drilling data transmitted by the control module, and obtain the drilling number prediction result of the drilling rig; The drilling rig demand prediction unit is used to execute the drilling rig demand prediction instruction, predict the drilling rig demand according to the historical drilling data transmitted by the control module and the drilling rig drilling number prediction result, and obtain the drilling rig demand prediction result; The first data transmission unit is used to transmit the drilling rig demand prediction result to the control module.

6. The drilling rig intelligent scheduling system according to claim 1, characterized in that: The drilling rig supply and demand matching module includes a supply and demand matching unit and a second data transmission unit; The supply and demand matching unit is used to match t drilling rig supply and demand matching results of drilling rig suppliers for each drilling rig demander according to the drilling rig demand data and drilling rig supply data transmitted by the control module; The second data transmission unit is used to transmit the drilling rig supply and demand matching result to the control module.

7. The drilling rig intelligent scheduling system according to claim 6, characterized in that: Matching t drilling rig suppliers for each drilling rig demander to obtain a drilling rig supply and demand matching result, including: using a KNN algorithm to match t drilling rig supply data for each drilling rig demand data, thereby matching t drilling rig suppliers for each drilling rig demander to obtain a drilling rig supply and demand matching result.

8. The drilling rig intelligent scheduling system according to claim 1, characterized in that: The drilling rig scheduling module includes a drilling rig route planning unit and a third data transmission unit; The drilling rig route planning unit is used to determine the travel route of each drilling rig between the drilling rig demanders corresponding to the target drilling rig supply data for all target drilling rig supply data matched by each drilling rig provided by the drilling rig supplier by using a genetic algorithm to obtain a planned route for each drilling rig; The third data transmission unit is used to feed back the planned route of each drilling rig to the drilling rig supplier corresponding to the drilling rig.

9. The drilling rig intelligent scheduling system according to claim 8, characterized in that: The genetic algorithm is used to determine the route of each drilling rig between the drilling rig demand parties corresponding to the target drilling rig supply data, and the planned route of each drilling rig is obtained, including: Assign different codes to n well sites corresponding to the drilling rigs, where the well sites are used to represent the objects to be constructed corresponding to the drilling rig demander; According to the preset initial population size and the code corresponding to each well site, multiple individuals are randomly generated to obtain the population corresponding to the genetic algorithm; For the population, the individuals in the population are updated by using the selection operation based on normal distribution, the sequential crossover operation and the exchange mutation operation to obtain the updated population; When the maximum fitness value corresponding to an individual in the population is greater than the preset target or the number of updates reaches the maximum, the individual corresponding to the maximum fitness value is used as the travel route corresponding to the drilling rig to obtain the planned route of each drilling rig; The fitness value corresponding to the individual is obtained through the fitness function, and the fitness function is the negative value of the objective function.

10. The drilling rig intelligent scheduling system according to claim 3, characterized in that: The scheduling risk prediction module includes a data preprocessing unit, an input data building unit, a model training unit, a risk value prediction unit and a fourth data transmission unit; The data preprocessing unit is used to normalize the drilling rig scheduling data in the historical drilling data to obtain the normalized drilling rig scheduling data, wherein the historical drilling data includes the drilling rig scheduling data and the risk value corresponding to the drilling rig scheduling data, and the drilling rig scheduling data includes the block to which the well site where the drilling rig works belongs, the drilling type, the well type, the drilling rig model, the drilling rig occupation period, the drilling rig transportation distance and the waiting period; The input data construction unit is used to construct the normalized drilling rig scheduling data into input vector data of uniform data length; The model training unit is used to use the input vector data as the input data corresponding to the risk prediction model constructed by the BRF neural network, use the corresponding risk value as the true value label corresponding to the risk prediction model, train the risk prediction model, and obtain the trained risk prediction model; The risk value prediction unit is used to use the trained risk prediction model to predict the scheduling risk of each drilling rig provided by the drilling rig supplier to obtain a prediction result; The fourth data transmission unit is used to feed back the prediction result to the control module.

Images