Evaluation method, device and equipment for digital transformation of oil and gas business and storage medium
By training a backpropagation evaluation model and optimizing the model weights, the problem of inaccurate evaluation of digital transformation of oil and gas business was solved, and accurate evaluation of oil and gas business was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- RICHFIT INFORMATION TECH
- Filing Date
- 2024-12-11
- Publication Date
- 2026-06-12
Smart Images
Figure CN122198676A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of oil and gas exploration and development technology, and in particular to an evaluation method, apparatus, equipment and storage medium for the digital transformation of oil and gas business. Background Technology
[0002] As oil and gas exploration and development become increasingly challenging, digital transformation has become an important means for the oil and gas industry in its production and operations.
[0003] In related technologies, the digital transformation of oil and gas business lacks quantifiable evaluation methods, and the evaluation indicators are not subdivided according to the professional fields of oil and gas business. The evaluation of digital transformation in a specific professional field (such as oil and gas business) is not accurate enough. Summary of the Invention
[0004] This application provides an evaluation method, apparatus, equipment, and storage medium for the digital transformation of oil and gas operations, which can improve the accuracy of the evaluation of the digital transformation of oil and gas field operations. The technical solution provided by this application is as follows:
[0005] According to one aspect of the embodiments of this application, an evaluation method for the digital transformation of oil and gas business is provided, the method comprising:
[0006] Obtain historical sample data for the oil and gas business, which includes historical data corresponding to all indicators related to the digital transformation of the oil and gas business, and the all indicators include all indicators at the last level in the multi-level indicator system; determine at least one key indicator from the all indicators.
[0007] Based on the at least one key indicator and the full set of indicators, the historical sample data is used to train the backpropagation evaluation model to obtain the trained evaluation model.
[0008] The trained evaluation model is used to evaluate and predict the future digital transformation of the oil and gas business, and the evaluation results of the digital transformation of the oil and gas business are obtained.
[0009] In some embodiments, the evaluation model is trained using the historical sample data based on the at least one key indicator and the full set of indicators to obtain a trained evaluation model. This includes:
[0010] Identify key sample data in the historical sample data, wherein the key sample data refers to the historical data corresponding to each of the at least one key indicator;
[0011] Using the key sample data, the evaluation model is trained for the first time to obtain the evaluation model after the first training. The first training includes optimizing the weights of at least one key indicator in the evaluation model.
[0012] Based on the evaluation model completed in the first training, the evaluation model completed in the first training is trained a second time using the full sample data of the historical sample data to obtain the trained evaluation model. The full sample data refers to the historical data corresponding to each of the full indicators. The second training includes optimizing the weights of the full indicators in the evaluation model completed in the first training.
[0013] In some embodiments, the step of training the evaluation model a second time using the full set of historical sample data, based on the evaluation model trained in the first training iteration, to obtain the trained evaluation model, includes:
[0014] Set the initial conditions for the genetic algorithm;
[0015] Initialize the population, where each individual in the population represents a set of historical sample data from the full sample data;
[0016] Using a loss function, the fitness of each individual in the population is calculated, and the fitness is used to represent the prediction accuracy of the parameters in the evaluation model.
[0017] The population is sorted non-dominated based on the fitness of each individual in the population.
[0018] Based on the sorting results of the non-dominated sorting, selection, crossover, and mutation operations are performed on the population to solve for the parameters in the evaluation model, thereby obtaining the trained evaluation model.
[0019] In some embodiments, the step of using the key sample data to perform the first training of the evaluation model to obtain the evaluation model after the first training includes:
[0020] The key sample data is normalized to obtain normalized key sample data.
[0021] The evaluation model is trained for the first time using the normalized key sample data to obtain the evaluation model after the first training.
[0022] In some embodiments, initializing the population includes:
[0023] Based on the evaluation model completed in the first training, the initial values corresponding to the full set of indicators in the second training process are determined.
[0024] In some embodiments, determining the initial values corresponding to the full set of metrics during the second training process based on the evaluation model completed in the first training includes:
[0025] The weights of the key indicators in the evaluation model completed in the first training are used as the initial values of each key indicator in the second training.
[0026] In the second training, the weights of all indicators in the total indicators of the oil and gas business, except for the at least one key indicator, are set to 0. The initial values of the total indicators in the second training include the initial values of each of the key indicators and the initial values of the weights of the other indicators in the second training.
[0027] In some embodiments, the step of using the full sample data of the historical sample data to perform the second training on the evaluation model that has been trained in the first training to obtain the trained evaluation model includes:
[0028] The evaluation model, which has been trained for the first time, is used to generate a score for the digital transformation of the oil and gas business. The score is used to indicate the maturity of the digital transformation of the oil and gas business.
[0029] Based on the score of the digital transformation of the oil and gas business and the historical sample data, the value of the loss function is calculated. The loss function is used to indicate the degree of fit of the output of the evaluation model to the historical sample data.
[0030] Based on the value of the loss function, the weights of each of the indicators are adjusted until the training stopping condition is met, training is stopped, and the evaluation model of the completed training is obtained.
[0031] In some embodiments, the activation function of the evaluation model is a non-linear activation function.
[0032] In some embodiments, determining at least one key indicator from the full set of indicators includes:
[0033] Based on the characteristics of the oil and gas business, at least one key indicator is determined;
[0034] The characteristics of the oil and gas business include at least one of the following: equity ratio, regional base, business type, cooperation agreement, business model, production stage, and construction scale.
[0035] According to one aspect of the embodiments of this application, an evaluation device for the digital transformation of oil and gas business is provided, the device comprising:
[0036] The data acquisition module is used to acquire historical sample data of the oil and gas business. The historical sample data includes historical data corresponding to all indicators related to the digital transformation of the oil and gas business. The all indicators include all indicators at the last level in the multi-level indicator system.
[0037] The indicator determination module is used to determine at least one key indicator from the full set of indicators.
[0038] The model training module is used to train the evaluation model based on the at least one key indicator and the full set of indicators, using the historical sample data, to obtain the trained evaluation model.
[0039] The evaluation module is used to evaluate and predict the future digital transformation of the oil and gas business using the trained evaluation model, and obtain the evaluation results of the digital transformation of the oil and gas business.
[0040] In some embodiments, the model training module includes:
[0041] The data determination submodule is used to determine the key sample data in the historical sample data, wherein the key sample data refers to the historical data corresponding to the at least one key indicator;
[0042] The model training submodule is used to perform the first training on the evaluation model using the key sample data to obtain the evaluation model after the first training. The first training includes optimizing the weights of at least one key indicator in the evaluation model.
[0043] The model training submodule is further configured to perform a second training on the evaluation model that was trained in the first training, using the full sample data of the historical sample data, to obtain the trained evaluation model. The full sample data refers to the historical data corresponding to each of the full indicators. The second training includes optimizing the weights of the full indicators in the evaluation model that was trained in the first training.
[0044] In some embodiments, the model training submodule includes:
[0045] The condition setting unit is used to set the initial conditions for the genetic algorithm;
[0046] An initialization unit is used to initialize a population, wherein each individual in the population represents a set of historical sample data in the full sample data;
[0047] The fitness calculation unit is used to calculate the fitness of each individual in the population using a loss function, wherein the fitness is used to represent the prediction accuracy of the parameters in the evaluation model.
[0048] The sorting unit is used to perform non-dominated sorting of the population based on the fitness of each individual in the population.
[0049] The model training unit is used to perform selection, crossover, and mutation operations on the population according to the sorting results of the non-dominated sorting, and to solve the parameters in the evaluation model to obtain the trained evaluation model.
[0050] In some embodiments, the model training submodule is used for:
[0051] The key sample data is normalized to obtain normalized key sample data.
[0052] The evaluation model is trained for the first time using the normalized key sample data to obtain the evaluation model after the first training.
[0053] In some embodiments, the initialization unit is configured to:
[0054] Based on the evaluation model completed in the first training, the initial values corresponding to the full set of indicators in the second training process are determined.
[0055] In some embodiments, the initialization unit is configured to:
[0056] The weights of the key indicators in the evaluation model completed in the first training are used as the initial values of each key indicator in the second training.
[0057] In the second training, the weights of all indicators in the total indicators of the oil and gas business, except for the at least one key indicator, are set to 0. The initial values of the total indicators in the second training include the initial values of each of the key indicators and the initial values of the weights of the other indicators in the second training.
[0058] In some embodiments, the model training unit is configured to:
[0059] The evaluation model, which has been trained for the first time, is used to generate a score for the digital transformation of the oil and gas business. The score is used to indicate the maturity of the digital transformation of the oil and gas business.
[0060] Based on the score of the digital transformation of the oil and gas business and the historical sample data, the value of the loss function is calculated. The loss function is used to indicate the degree of fit of the output of the evaluation model to the historical sample data.
[0061] Based on the value of the loss function, the weights of each of the indicators are adjusted until the training stopping condition is met, training is stopped, and the evaluation model of the completed training is obtained.
[0062] In some embodiments, the activation function of the evaluation model is a non-linear activation function.
[0063] In some embodiments, the indicator determination module is configured to:
[0064] Based on the characteristics of the oil and gas business, at least one key indicator is determined;
[0065] The characteristics of the oil and gas business include at least one of the following: equity ratio, regional base, business type, cooperation agreement, business model, production stage, and construction scale.
[0066] According to one aspect of the embodiments of this application, a computer device is provided, the computer device including a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to realize the above-described evaluation method for digital transformation of oil and gas business.
[0067] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, the computer program being loaded and executed by a processor to realize the above-described evaluation method for digital transformation of oil and gas business.
[0068] According to one aspect of the embodiments of this application, a computer program product is provided, which is loaded and executed by a processor to realize the evaluation method for the digital transformation of oil and gas business described above.
[0069] The technical solutions provided in this application embodiment may have the following beneficial effects:
[0070] By training the evaluation model with historical sample data from oil and gas field operations, the evaluation model learns the weights of various indicators in the historical digital transformation of oil and gas field operations. This enables the evaluation model to more accurately predict the evaluation results of the digital transformation of oil and gas field operations, thereby improving the accuracy of the evaluation.
[0071] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description
[0072] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0073] Figure 1 This is a flowchart of an evaluation method for the digital transformation of oil and gas business provided in one embodiment of this application;
[0074] Figure 2 This is a schematic diagram of the structure of an evaluation model provided in one embodiment of this application;
[0075] Figure 3 This is a schematic diagram of the activation function curve provided in one embodiment of this application;
[0076] Figure 4 This is a schematic diagram of a genetic algorithm provided in one embodiment of this application;
[0077] Figure 5 This is a schematic diagram of a digital transformation maturity evaluation system provided in one embodiment of this application;
[0078] Figure 6 This is a block diagram of an evaluation device for the digital transformation of oil and gas business provided in one embodiment of this application;
[0079] Figure 7 This is a block diagram of a computer device provided in one embodiment of this application. Detailed Implementation
[0080] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of methods consistent with some aspects of this application as detailed in the appended claims.
[0081] The method provided in this application can be executed by a computer device, which refers to an electronic device with data computing, processing, and storage capabilities. This computer device can be a terminal such as a PC (Personal Computer), tablet computer, smartphone, wearable device, or intelligent robot; or it can be a server. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.
[0082] The technical solution of this application will be described and illustrated below through several embodiments.
[0083] Please refer to Figure 1 This document illustrates a flowchart of an evaluation method for the digital transformation of oil and gas operations according to an embodiment of this application. In this embodiment, the method is primarily illustrated by its application to the computer equipment described above. The method may include at least one of the following steps (110-140).
[0084] Step 110: Obtain historical sample data for the oil and gas business. The historical sample data includes historical data corresponding to all indicators related to the digital transformation of the oil and gas business. The total indicators include all indicators at the last level of the multi-level indicator system.
[0085] In some embodiments, as the difficulty of oil and gas exploration and development increases, digital transformation has become an important means for the oil and gas industry to cope with production and operation risks and challenges and achieve sustainable development. Quantitative guidance and evaluation of digital transformation work are issues that the group company and overseas oil and gas companies urgently need to address.
[0086] In some embodiments, when acquiring historical sample data, it is necessary to first determine the indicators to be evaluated, and then obtain the historical sample data corresponding to each indicator. In some embodiments, the indicator system used in the evaluation process can be a multi-level, multi-dimensional multi-level indicator system, thereby improving the comprehensiveness and accuracy of the evaluation results.
[0087] Step 120: Identify at least one key indicator from the total number of indicators.
[0088] In some embodiments, the multi-level indicator system includes three levels of indicators, mainly comprising 7 dimensions (first-level indicators), 16 indicators (second-level indicators), and 47 elements (third-level indicators). Among them, the 47 elements are the last level of indicators in the multi-level indicator system, and therefore these 47 elements constitute the total number of indicators.
[0089] In some embodiments, at least one key indicator is determined based on the characteristics of the oil and gas business; wherein the characteristics of the oil and gas business include at least one of the following: equity ratio, regional base, business type, cooperation agreement, business model, production stage, and construction scale.
[0090] In some embodiments, at least one key indicator (KPI) highly relevant to the oil and gas business is identified from the total number of indicators. In determining this KPI, the characteristics of different regions must be fully considered. Different regions have different project types, different equity ratios, and different geographical locations, leading to differences in the range of KPIs. For example, if the project type is exploration, indicators related to development, production, pipelines, and refining will not appear in the KPIs. If there are minority shareholders, the selection of indicators such as the top-level design and organizational management of digital transformation, cybersecurity, and infrastructure will differ. Furthermore, different regions have different regulations and policies, resulting in differences in the selection of indicators related to cybersecurity, data lake integration, and data sharing. For different projects / businesses, an appropriate range of KPIs can be selected based on their specific characteristics.
[0091] In some embodiments, steps 110 and 120 above constitute data preparation work before training the model, incorporating expert evaluation experience and the actual operation of enterprise digitalization into the model. For example, for each evaluation object (e.g., company, oil and gas field business, etc.), the actual values (historical data) of digital transformation or digitalization construction indicators over the past ten years are collected on an annual basis as the training basis data for the evaluation model (e.g., BP (Back Propagation) neural network model or other suitable neural network model) in this application.
[0092] In some embodiments, the multi-level indicator system includes multiple primary indicators in terms of dimensions, as shown in Table 1 below.
[0093] Table 1
[0094]
[0095] In some embodiments, the indicators at each level included in the multi-level indicator system can be referenced in Table 2 below.
[0096] Table 2
[0097]
[0098]
[0099]
[0100]
[0101]
[0102] Step 130: Use historical sample data to train the evaluation model and obtain the trained evaluation model.
[0103] In some embodiments, the evaluation model can be a neural network model. A neural network model is a multi-layer feedforward network trained using an error backpropagation algorithm. For example... Figure 2 As shown, neural network models can learn and store a large number of input-output pattern mappings without having to reveal the mathematical equations describing these mappings in advance.
[0104] In some embodiments, the activation function of the evaluation model is a non-linear activation function.
[0105] In some embodiments, considering the large number of input indicators for the evaluation model (e.g., up to 47), and the poor fit of linear models, a non-linear activation function, such as the sigmoid activation function, is chosen for the evaluation model. The evaluation results are then fitted. In some embodiments, reference is made to... Figure 2 Evaluation results Among them, activation function
[0106] The graph of the activation function is shown below. Figure 3 As shown, w1, b1, w2, and b2 are the parameters to be solved in the evaluation model.
[0107] Step 140: Using the trained evaluation model, evaluate and predict the future digital transformation of the oil and gas business, and obtain the evaluation results of the digital transformation of the oil and gas business.
[0108] In some embodiments, a trained evaluation model is used to evaluate and predict the future digital transformation of oil and gas operations, resulting in an evaluation result for the digital transformation of oil and gas operations. In some embodiments, the evaluation result is a digital maturity score.
[0109] In summary, the technical solution provided in this application uses historical sample data from oil and gas field operations to train the evaluation model, thereby enabling the evaluation model to learn the weights of various indicators in the historical digital transformation of oil and gas field operations. This allows the evaluation model to more accurately predict the evaluation results of the digital transformation of oil and gas field operations, thus improving the accuracy of the evaluation of the digital transformation of oil and gas field operations.
[0110] In some possible implementations, based on the at least one key indicator and the full set of indicators, historical sample data is used to train the evaluation model, resulting in a trained evaluation model. This includes the following steps:
[0111] 1. Identify key sample data in the historical sample data. Key sample data refers to the historical data corresponding to at least one key indicator.
[0112] 2. Using key sample data, the evaluation model is trained for the first time to obtain the evaluation model after the first training. The first training includes optimizing the weight of at least one key indicator in the evaluation model.
[0113] 3. Based on the evaluation model completed in the first training, the evaluation model completed in the first training is trained a second time using the full sample data of historical sample data to obtain the trained evaluation model. The full sample data refers to the historical data corresponding to each of the full indicators. The second training includes optimizing the weights of the full indicators in the evaluation model completed in the first training.
[0114] In some embodiments, the first training uses non-full sample data (i.e., at least one key indicator and its corresponding evaluation results defined in step 1), and calculates a portion of the model weights (i.e., the weights corresponding to at least one key indicator) using the method built into the evaluation model. The calculated weights are then used as some initial values for the second training, while other initial values are set to 0.
[0115] In some embodiments, the second training uses full sample data (i.e., historical data corresponding to all indicators) and uses a genetic algorithm to solve for the optimal parameters, finding the optimal solution near the key indicators. This makes the indicators of the final evaluation model more targeted to the relevant field (such as the oil and gas business field), thereby improving the accuracy and professionalism of the evaluation results.
[0116] In some embodiments, the evaluation model is trained for the first time using key sample data to obtain the evaluation model after the first training, including the following steps:
[0117] 1. Normalize the key sample data to obtain normalized key sample data;
[0118] 2. Using normalized key sample data, the evaluation model is trained for the first time to obtain the evaluation model after the first training.
[0119] In some embodiments, normalization involves transforming data to the range of 0 to 1. This normalization operation addresses the inconsistency between different indicator dimensions and eliminates the impact of data value space on the evaluation model. It also eliminates the influence of different units of measurement on the model. In some embodiments, for the data of one indicator, normalization can refer to the following formula:
[0120]
[0121] Where x is a data point to be normalized in this indicator, and x' is the normalized value of that data. min It is the minimum value in this indicator, x max It is the maximum value in this indicator.
[0122] In some embodiments, key metrics include: w 1_sel b 1_sel w 2_sel b 2_sel .
[0123] In some embodiments, the step of training the evaluation model a second time using the full set of historical sample data, based on the evaluation model trained in the first training, to obtain the trained evaluation model, includes the following steps:
[0124] 1. Set the initial conditions for the genetic algorithm;
[0125] 2. Initialize the population, where each individual in the population represents a set of historical sample data from the full sample data;
[0126] 3. Using a loss function, calculate the fitness of each individual in the population, whereby the fitness is used to represent the prediction accuracy of the parameters in the evaluation model;
[0127] 4. Based on the fitness of each individual in the population, perform a non-dominated sorting of the population;
[0128] 5. Based on the sorting results of the non-dominated sorting, perform selection, crossover, and mutation operations on the population, solve for the parameters in the evaluation model, and obtain the trained evaluation model.
[0129] The basic idea of a genetic algorithm is to start from an initial population, select individuals using the natural law of survival of the fittest, and generate a new generation through crossover and mutation. This process continues generation by generation until the objective is met. Genetic algorithms simulate the selection, crossover, and mutation operations in biological evolution, gradually evolving from an initial population to find the optimal solution. In this embodiment, the population includes multiple individuals, each representing a set of parameters for a possible evaluation model.
[0130] In this embodiment, a fast non-dominated sorting genetic algorithm is employed. Fitness ensures a relatively even distribution of individuals, maintaining population diversity and preventing the over-proliferation of super-individuals, thus avoiding premature convergence. Furthermore, compared to ordinary non-dominated sorting, fast non-dominated sorting reduces the computational complexity of the algorithm, thereby improving the optimization efficiency of the objective function. Using the fast non-dominated sorting genetic algorithm for a second training of the evaluation model can improve the training efficiency and effectiveness, thereby enhancing the accuracy of oil and gas business assessments.
[0131] In some embodiments, initializing the population includes: determining the initial values corresponding to the full set of indicators in the second training process based on the evaluation model completed in the first training.
[0132] In some embodiments, by training twice, the nonlinearity of the evaluation model is preserved, and the weights of the trained model can be optimized towards key indicators, thereby optimizing the evaluation results of oil and gas field operations as much as possible.
[0133] In some embodiments, based on the evaluation model completed in the first training, the initial values corresponding to the full set of indicators in the second training process are determined, including the following steps:
[0134] 1. Use the weights of the key indicators in the evaluation model completed in the first training as the initial values of each key indicator in the second training.
[0135] 2. In the second training, set the weights of all indicators in the oil and gas business, except for the key indicators, to 0. The initial values of all indicators in the second training include the initial values of each key indicator and the initial values of the weights of other indicators.
[0136] In some embodiments, during the second training process, the key metric w 1_sel b 1_sel w 2_sel b 2_sel The initial values are the weights of the key metrics obtained from the first training; other metrics w1 others b1 others w2 others b2 others The initial value is set to 0.
[0137] In some embodiments, the evaluation model that has been trained once is trained a second time using the full sample data of historical sample data to obtain a fully trained evaluation model, including the following steps:
[0138] 1. Use the evaluation model that has been trained for the first time to generate a score for the digital transformation of oil and gas business. The score is used to indicate the maturity of the digital transformation of oil and gas business.
[0139] 2. Based on the scores and historical sample data of the digital transformation of oil and gas business, calculate the value of the loss function. The loss function is used to indicate the degree of fit of the evaluation model's output to the historical sample data.
[0140] 3. Based on the value of the loss function, adjust the weights of each indicator until the training stopping condition is met, then stop training and obtain the evaluation model after training is complete.
[0141] In some embodiments, the loss function evaluates the goodness of fit of the evaluation model to the samples. The closer the predicted result is to the actual value, the stronger the model's fit and the higher the goodness of fit, and the smaller the corresponding loss function value. Conversely, the greater the difference between the predicted result and the actual value, the weaker the model's fit and the lower the goodness of fit, and the larger the corresponding loss function value. In some embodiments, for ease of calculation, the loss function value can be calculated based on Euclidean distance, using the minimum squared error (MSE) between the actual value and the estimated value as the loss function. The calculation can be referenced from the following formula:
[0142] minC(Y,G(X))=‖G(X)-Y‖ 2 =∑ i (G(x i )-y i ) 2 ,
[0143] Where minC(Y,G(X)) is the value of the loss function, G(X) is the actual value (i.e., historical sample data), and G(x) is the loss function value. i Let y be the actual value of the i-th indicator, and Y be the prediction result of the evaluation model. i This represents the prediction result for the i-th indicator.
[0144] In some embodiments, the model training accuracy is set, such as setting the number of generations for genetic algorithm training or the allowable error iters=1000.
[0145] During model training, optimization constraints are set. For each generation of the population, two modeling errors are calculated: error_key for the model with only key indicators and error_all for the model with all indicators. During each selection operation of the genetic algorithm, the top 20 individuals are first sorted by error_key from lowest to highest, and then sorted by error_all to select the top 10 individuals.
[0146] In some embodiments, such as Figure 4 The genetic algorithm shown solves for the optimal parameters. It employs selection, crossover, and mutation operations to solve the problem, and after training, obtains a set of optimal weights w1, b1, w2, and b2 for the evaluation model. After training, the evaluation model is saved.
[0147] In some embodiments, such as Figure 5 The digital transformation maturity assessment system shown evaluates the digital transformation of oil and gas field operations.
[0148] In the aforementioned implementation methods, the digital transformation indicator evaluation methods and systems can help projects / oil and gas field operations more scientifically and accurately assess their digital maturity, identify shortcomings and weaknesses, analyze the reasons for these shortcomings, and provide reference and direction for the next step of digital transformation. Furthermore, optimization is performed on a smaller number of key indicators, followed by optimization training on all indicators. This results in an evaluation model that considers all indicators while also being specific to key indicators, thereby improving the accuracy of the evaluation model in assessing the digital transformation of corresponding fields (such as oil and gas operations).
[0149] In one example, the technical solution provided in the embodiments of this application has been initially applied to a gas field that mainly produces natural gas. We are the operator with 100% equity and a product sharing contract, which includes the following steps.
[0150] 1. Collect historical data samples of the gas field over the past 10 years.
[0151] Expert scores for the Amu Darya gas field over the past 10 years were collected, with a maximum score of 100. A total of 44 pairs of individual and total scores for each indicator were collected. The digital maturity score for this gas field over the past 10 years can be found in Table 3 below.
[0152] Table 3
[0153]
[0154]
[0155] 2. Training and evaluating the model.
[0156] Ten sample data pairs (x1, x2, x3, x44, y_score) were normalized, and the model was trained using the evaluation model. The optimal parameters were solved using a genetic algorithm, and the evaluation model was finally saved.
[0157] 3. Launch the evaluation model to predict the digital maturity score of the gas field over the next 3 years.
[0158] The trained evaluation model is launched, predicting a total digital maturity score of 96.52 in 2024, 96.88 in 2025, and 97.12 in 2026.
[0159] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.
[0160] Please refer to Figure 6This diagram illustrates a block diagram of an evaluation device for the digital transformation of oil and gas operations, provided in one embodiment of this application. The device has the functionality to implement the aforementioned evaluation method example for the digital transformation of oil and gas operations. This functionality can be implemented in hardware or by hardware executing corresponding software. The device can be the computer equipment described above, or it can be installed on a computer device. The device 600 may include: a data acquisition module 610, an indicator determination module 620, a model training module 630, and an evaluation module 640.
[0161] The data acquisition module 610 is used to acquire historical sample data of the oil and gas business. The historical sample data includes historical data corresponding to all indicators related to the digital transformation of the oil and gas business. The all indicators include all indicators at the last level in the multi-level indicator system.
[0162] The indicator determination module 620 is used to determine at least one key indicator from the full set of indicators.
[0163] The model training module 630 is used to train the evaluation model based on the at least one key indicator and the full set of indicators, using the historical sample data, to obtain the trained evaluation model.
[0164] The evaluation module 640 is used to evaluate and predict the future digital transformation of the oil and gas business using the trained evaluation model, and obtain the evaluation result of the digital transformation of the oil and gas business.
[0165] In some embodiments, the model training module 630 includes a data determination submodule and a model training submodule.
[0166] The data determination submodule is used to determine the key sample data in the historical sample data, wherein the key sample data refers to the historical data corresponding to the at least one key indicator.
[0167] The model training submodule is used to perform the first training of the evaluation model using the key sample data to obtain the evaluation model after the first training. The first training includes optimizing the weights of at least one key indicator in the evaluation model.
[0168] The model training submodule is further configured to perform a second training on the evaluation model that was trained in the first training, using the full sample data of the historical sample data, to obtain the trained evaluation model. The full sample data refers to the historical data corresponding to each of the full indicators. The second training includes optimizing the weights of the full indicators in the evaluation model that was trained in the first training.
[0169] In some embodiments, the model training submodule includes: a condition setting unit, an initialization unit, a fitness calculation unit, a sorting unit, and a model training unit.
[0170] The condition setting unit is used to set the initial conditions of the genetic algorithm.
[0171] The initialization unit is used to initialize the population, where each individual in the population represents a set of historical sample data in the full sample data.
[0172] The fitness calculation unit is used to calculate the fitness of each individual in the population using a loss function. The fitness is used to represent the prediction accuracy of the parameters in the evaluation model.
[0173] The sorting unit is used to perform non-dominated sorting of the population based on the fitness of each individual in the population.
[0174] The model training unit is used to perform selection, crossover, and mutation operations on the population according to the sorting results of the non-dominated sorting, and to solve the parameters in the evaluation model to obtain the trained evaluation model.
[0175] In some embodiments, the model training submodule is used for:
[0176] The key sample data is normalized to obtain normalized key sample data.
[0177] The evaluation model is trained for the first time using the normalized key sample data to obtain the evaluation model after the first training.
[0178] In some embodiments, the initialization unit is configured to:
[0179] Based on the evaluation model completed in the first training, the initial values corresponding to the full set of indicators in the second training process are determined.
[0180] In some embodiments, the initialization unit is configured to:
[0181] The weights of the key indicators in the evaluation model completed in the first training are used as the initial values of each key indicator in the second training.
[0182] In the second training iteration, the weights of all indicators in the total indicators of the oil and gas business, excluding the at least one key indicator, are set to 0. The initial values of the total indicators in the second training iteration include the initial values of each of the key indicators and the initial values of the weights of the other indicators. In some embodiments, the model training unit is configured to:
[0183] The evaluation model, which has been trained for the first time, is used to generate a score for the digital transformation of the oil and gas business. The score is used to indicate the maturity of the digital transformation of the oil and gas business.
[0184] Based on the score of the digital transformation of the oil and gas business and the historical sample data, the value of the loss function is calculated. The loss function is used to indicate the degree of fit of the output of the evaluation model to the historical sample data.
[0185] Based on the value of the loss function, the weights of each of the indicators are adjusted until the training stopping condition is met, training is stopped, and the evaluation model of the completed training is obtained.
[0186] In some embodiments, the activation function of the evaluation model is a non-linear activation function.
[0187] In some embodiments, the indicator determination module 620 is configured to:
[0188] Based on the characteristics of the oil and gas business, at least one key indicator is determined;
[0189] The characteristics of the oil and gas business include at least one of the following: equity ratio, regional base, business type, cooperation agreement, business model, production stage, and construction scale.
[0190] In summary, the technical solution provided in this application uses historical sample data from oil and gas field operations to train the evaluation model, thereby enabling the evaluation model to learn the weights of various indicators in the historical digital transformation of oil and gas field operations. This allows the evaluation model to more accurately predict the evaluation results of the digital transformation of oil and gas field operations, thus improving the accuracy of the evaluation of the digital transformation of oil and gas field operations.
[0191] It should be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional modules when implementing its functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.
[0192] Please refer to Figure 7 This diagram illustrates the structural block diagram of a computer device according to an embodiment of this application. The computer device is used to implement the evaluation method for the digital transformation of oil and gas operations provided in the above embodiments. Specifically:
[0193] The computer device 700 includes a CPU (Central Processing Unit) 701, a system memory 704 including RAM (Random Access Memory) 702 and ROM (Read-Only Memory) 703, and a system bus 705 connecting the system memory 704 and the central processing unit 701. The computer device 700 also includes a basic I / O (Input / Output) system 706 that facilitates information transfer between various components within the computer, and a mass storage device 707 for storing the operating system 713, application programs 714, and other program modules 715.
[0194] The basic input / output system 706 includes a display 708 for displaying information and an input device 709 for user input, such as a mouse or keyboard. Both the display 708 and the input device 709 are connected to the central processing unit 701 via an input / output controller 710 connected to the system bus 705. The basic input / output system 706 may also include the input / output controller 710 for receiving and processing input from multiple other devices such as a keyboard, mouse, or electronic stylus. Similarly, the input / output controller 710 also provides output to a display screen, printer, or other types of output devices.
[0195] The mass storage device 707 is connected to the central processing unit 701 via a mass storage controller (not shown) connected to the system bus 705. The mass storage device 707 and its associated computer-readable media provide non-volatile storage for the computer device 700. That is, the mass storage device 707 may include computer-readable media (not shown) such as a hard disk or a CD-ROM (Compact Disc Read-Only Memory) drive.
[0196] Without loss of generality, the computer-readable medium may include computer storage media and communication media. Computer storage media include volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include RAM, ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash memory or other solid-state storage, CD-ROM, DVD (Digital Video Disc) or other optical storage, magnetic tape cassettes, magnetic tape, disk storage, or other magnetic storage devices. Of course, those skilled in the art will recognize that the computer storage media are not limited to the above-mentioned types. The system memory 704 and mass storage device 707 described above can be collectively referred to as memory.
[0197] According to various embodiments of this application, the computer device 700 can also be connected to a remote computer on a network, such as the Internet. That is, the computer device 700 can be connected to a network 712 via a network interface unit 711 connected to the system bus 705, or the network interface unit 711 can be used to connect to other types of networks or remote computer systems (not shown).
[0198] In an exemplary embodiment, a computer-readable storage medium is also provided, wherein a computer program is stored therein, which, when executed by a processor, implements the aforementioned evaluation method for the digital transformation of oil and gas business.
[0199] In an exemplary embodiment, a computer program product is also provided, which is loaded and executed by a processor to realize the evaluation method for the digital transformation of oil and gas business described above.
[0200] It should be understood that "multiple" as used in this article refers to two or more. The character " / " generally indicates that the objects before and after it are in an "or" relationship.
[0201] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
Claims
1. An evaluation method for the digital transformation of oil and gas operations, characterized in that, The method includes: Obtain historical sample data for the oil and gas business. The historical sample data includes historical data corresponding to all indicators related to the digital transformation of the oil and gas business. The all indicators include all indicators at the last level in the multi-level indicator system. From the total number of indicators, at least one key indicator is identified; Based on the at least one key indicator and the full set of indicators, the historical sample data is used to train the evaluation model, and the trained evaluation model is obtained. The trained evaluation model is used to evaluate and predict the future digital transformation of the oil and gas business, and the evaluation results of the digital transformation of the oil and gas business are obtained.
2. The method according to claim 1, characterized in that, The step of training the evaluation model based on the at least one key indicator and the full set of indicators, using the historical sample data, to obtain the trained evaluation model includes: Identify key sample data in the historical sample data, wherein the key sample data refers to the historical data corresponding to each of the at least one key indicator; Using the key sample data, the evaluation model is trained for the first time to obtain the evaluation model after the first training. The first training includes optimizing the weights of at least one key indicator in the evaluation model. Based on the evaluation model completed in the first training, the evaluation model completed in the first training is trained a second time using the full sample data of the historical sample data to obtain the trained evaluation model. The full sample data refers to the historical data corresponding to each of the full indicators. The second training includes optimizing the weights of the full indicators in the evaluation model completed in the first training.
3. The method according to claim 2, characterized in that, The evaluation model, based on the first training iteration, is further trained a second time using the full set of historical sample data to obtain the fully trained evaluation model, including: Set the initial conditions for the genetic algorithm; Initialize the population, where each individual in the population represents a set of historical sample data from the full sample data; Using a loss function, the fitness of each individual in the population is calculated, and the fitness is used to represent the prediction accuracy of the parameters in the evaluation model. The population is sorted non-dominated based on the fitness of each individual in the population. Based on the sorting results of the non-dominated sorting, selection, crossover, and mutation operations are performed on the population to solve for the parameters in the evaluation model, thereby obtaining the trained evaluation model.
4. The method according to claim 3, characterized in that, The initialization of the population includes: Based on the evaluation model completed in the first training, the initial values corresponding to the full set of indicators in the second training process are determined.
5. The method according to claim 4, characterized in that, The determination of initial values for all metrics during the second training process, based on the evaluation model completed in the first training iteration, includes: The weights of the key indicators in the evaluation model completed in the first training are used as the initial values of each key indicator in the second training. In the second training, the weights of all indicators in the total indicators of the oil and gas business, except for the at least one key indicator, are set to 0. The initial values of the total indicators in the second training include the initial values of each of the key indicators and the initial values of the weights of the other indicators in the second training.
6. The method according to claim 4, characterized in that, The process of using the full set of historical sample data to train the evaluation model after the first training to obtain the trained evaluation model includes: The evaluation model, which has been trained for the first time, is used to generate a score for the digital transformation of the oil and gas business. The score is used to indicate the maturity of the digital transformation of the oil and gas business. Based on the score of the digital transformation of the oil and gas business and the historical sample data, the value of the loss function is calculated. The loss function is used to indicate the degree of fit of the output of the evaluation model to the historical sample data. Based on the value of the loss function, the weights of each of the indicators are adjusted until the training stopping condition is met, training is stopped, and the evaluation model of the completed training is obtained.
7. The method according to any one of claims 1 to 6, characterized in that, The step of identifying at least one key indicator from the full set of indicators includes: Based on the characteristics of the oil and gas business, at least one key indicator is determined; The characteristics of the oil and gas business include at least one of the following: equity ratio, regional base, business type, cooperation agreement, business model, production stage, and construction scale.
8. An evaluation device for the digital transformation of oil and gas operations, characterized in that, The device includes: The data acquisition module is used to acquire historical sample data of the oil and gas business. The historical sample data includes historical data corresponding to all indicators related to the digital transformation of the oil and gas business. The all indicators include all indicators at the last level in the multi-level indicator system. The indicator determination module is used to determine at least one key indicator from the full set of indicators. The model training module is used to train the evaluation model based on the at least one key indicator and the full set of indicators, using the historical sample data, to obtain the trained evaluation model. The evaluation module is used to evaluate and predict the future digital transformation of the oil and gas business using the trained evaluation model, and obtain the evaluation results of the digital transformation of the oil and gas business.
9. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing a computer program, which is loaded and executed by the processor to implement the evaluation method for digital transformation of oil and gas business as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is loaded and executed by a processor to implement the evaluation method for business digital transformation as described in any one of claims 1 to 7.