Gathering and transportation intelligent operation and maintenance system and device based on missing data filling method

Abstract

The invention discloses a gathering and transportation intelligent operation and maintenance system and device based on a missing data filling method, and belongs to the technical field of intelligent operation and maintenance of an oil field gathering and transportation system.The gathering and transportation intelligent operation and maintenance system comprises a data acquisition module, a gathering and transportation grouping module, a real-time prediction module and a missing value supplementing module; comprising a pressure sensor, a temperature sensor, a flow sensor and a water content sensor and is used for collecting real-time pressure data, temperature data, flow data and crude oil water content data of a transportation pipeline in the oil field gathering and transportation process. According to the gathering and transportation intelligent operation and maintenance system and device based on the missing data filling method, through real-time data collection and intraday prediction model establishment, the system can intelligently manage pipeline data in the gathering and transportation process of an oil field, the operation and maintenance efficiency is improved, through the missing value supplementing module, the system can effectively process the data missing problem, and the operation and maintenance efficiency is improved. The data integrity and the continuous operation of the system are ensured, the real-time prediction module can dynamically update the prediction model, and the prediction accuracy and reliability are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent operation and maintenance of oilfield gathering and transportation systems, and specifically is a gathering and transportation intelligent operation and maintenance system and device based on a missing data filling method. Background Art

[0002] In the oilfield gathering and transportation system, monitoring and management rely on a large number of sensors and data acquisition devices, which monitor the temperature, pressure, and flow parameters of the production process in real time. However, the complex operating environment and technical limitations will lead to data loss during the data collection process. Data loss will lead to monitoring blind spots, affect equipment health management and process optimization, and increase production safety risks and operating costs. Traditional oilfield gathering and transportation data interpolation methods are mostly based on the historical operation data of oilfield pipelines to establish prediction equations to predict missing data. However, due to the different external factors in the daily pipeline operation process, it is easy to have large errors when filling the missing data of pipelines in the gathering and transportation process only through historical data, and it is impossible to group pipelines and make real-time predictions according to the actual pipeline specifications in the daily operation process, which has the problems of low practicality and functionality. This case proposes an intelligent operation and maintenance system and device for gathering and transportation based on the missing data filling method to solve the above technical problems. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an intelligent operation and maintenance system and device for gathering and transportation based on a missing data filling method, which solves the above technical problems by improving the detection method and processing method.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method includes data collection module, gathering and transportation grouping module, real-time prediction module and missing value supplement module; The data acquisition module includes a pressure sensor, a temperature sensor, a flow sensor, and a water content sensor, which are used to collect real-time pressure data, temperature data, flow data, and crude oil water content data of the transportation pipeline during the oil field gathering and transportation process, and collect oil field gathering and transportation benchmark data, including pipeline specifications and start and stop status, and transmit them to subsequent modules; The gathering and transportation grouping module groups the daily operating pipelines based on the collected current oilfield gathering and transportation pipeline specifications and start and stop status. Establish a daily operation pipeline database, group the pipelines activated daily based on pipeline specifications, divide the pipelines of the same model in operation on the same day into the same file, and record the real-time collection and transportation data collected by the data acquisition module; The real-time prediction module, for the pipelines in the same archive, establishes a prediction model for the daily pressure, temperature, flow rate, and water content of each pipeline based on the real-time gathering and transportation data of each pipeline in the archive, and synchronously verifies and updates the prediction model based on the real-time recorded pipeline data; The missing value supplement module is based on the pipeline data prediction model in the same file, and comprehensively fills the missing data of the missing pipeline by combining the output value of the prediction model of the missing data with the output value of the prediction model of other pipelines in the same file.

[0005] Furthermore, the real-time prediction module, for the pipelines in the same archive, establishes a prediction model for the daily pressure, temperature, flow rate, and water content of each pipeline based on the real-time gathering and transportation data of each pipeline in the archive, and synchronously verifies and updates the prediction model based on the real-time recorded pipeline data. The specific steps are as follows: For the pipeline data in the same group file in the daily operation pipeline database, based on the real-time data of each pipeline during the day's operation, pressure, temperature, flow rate, and water content data prediction models are established respectively; Based on time interval Collect real-time monitoring data of different pipelines in each archive, and conduct phased verification and update of the prediction model established on the day.

[0006] Furthermore, for the pipeline data in the same group file in the daily operation pipeline database, based on the real-time data of each pipeline in operation on that day, pressure, temperature, flow rate, and water content data prediction models are established respectively, and the specific steps are as follows: For the pipeline data in the same group file, the pressure, temperature, flow rate, and water content data in each pipeline are timestamped. Based on each pipeline in the same group file, the data before and after the pipeline operation on that day are collected. The pressure, temperature, flow rate and water content data at the moment will be Always Divide equally, based on The model establishes the pressure, temperature, flow rate and water content data prediction models of different pipelines in the same group file respectively. The specific steps are as follows: Collect the data before and after the pipeline runs on the day The data at the moment is recorded as ,in ,in Respectively represent the pressure, temperature, flow rate and water content data of the current pipeline; Respectively The data at each moment is divided into equal parts, each containing the time steps , mark each piece of data as ; Each piece of data As input, each time step , set the input dimension of the model to 4, divide the data set into training set, validation set and test set, and use the mean square error as the loss function. The algorithm formula is: ; in, They represent the true value and the predicted value respectively. The Adam optimizer is used to further update the model parameters to obtain the pressure, temperature, flow rate, and water content data prediction model for different pipelines in the same group.

[0007] Furthermore, the time interval based Collect the real-time monitoring data of different pipelines in each file, and conduct phase verification and update of the prediction model established on the day. The specific steps are as follows: Based on time interval Get the time points , at each time point The data in different pipeline files in the same group are extracted, including the actual values ​​of pipeline pressure, pipeline temperature, pipeline flow and water content data. ; The time point Input to the corresponding pipeline training of the day In the model, get the time point Current pipeline data prediction value , based on the mean absolute error method to calculate the actual value With the predicted value The error of is, and its algorithm formula is: ; in, Represents at a point in time The mean absolute error between the actual value and the predicted value at the time of right Make a judgment: when ≥ When , it means that the current pipeline data prediction model needs to be updated; when < When , it means that the current pipeline data prediction model does not need to be updated and continues to be used; against ≥ Time, collection time New data since then, based on Split into equal parts and retrain using the updated dataset model, and replace the old model with the new prediction model as the data prediction model for the current pipeline.

[0008] Furthermore, the real-time prediction module is also provided with a daily anomaly detection module, which detects the anomaly of the same day based on the same type of pipeline data in the same group. Perform real-time anomaly detection at all times. The specific steps are as follows: Collect the current day’s data from all pipeline files in the same group. The real-time pressure, temperature, flow rate, and water content data are collected every minute before the moment, and the original data sets are divided based on the data type. The abnormal parameters of each data set are analyzed using the isolation forest method. The specific steps are as follows: Randomly extract a sample subset from the original data set, the size is , for each tree starting from the root node, randomly select a feature and a random split value on the feature to divide the data set into left and right child nodes, and recursively repeat the above process for each child node until the preset tree depth is reached or there is only one data point in the child node; For each data point , calculate the path length in each tree , average the path lengths of all trees to get the average path length , calculate the anomaly score , and its algorithm formula is: ; ; ; in, Represents the total number of samples, that is, the number of all data points involved in anomaly detection, is the sample constant, Representative A harmonic number, when When the value is greater than the detection threshold, it means that the pipeline data points collected on that day are abnormal. The abnormal pipelines are marked in the same group file, and the abnormal data type of the abnormal pipelines is The model is isolated and the output value is not transmitted to the subsequent modules.

[0009] Furthermore, the missing value filling module is based on the pipeline data prediction model in the same file, and comprehensively fills the missing data of the pipeline with missing data by combining the prediction model output value of the pipeline with missing data with the prediction model output values ​​of other pipelines in the same file. The specific steps are as follows: Collect real-time monitoring data of all pipelines in the same file and determine the time point when data is missing And the missing data pipeline, based on the missing data type through the data of the relevant type of data Model, by entering missing time points Get the predicted value of missing pipeline data, combined with other pipelines in the same file at the missing time point The output value of the prediction model is used to comprehensively generate the missing data value of the missing pipeline.

[0010] Furthermore, the real-time monitoring data of all pipelines in the same file are collected to determine the time point when the data is missing. And the missing data pipeline, based on the missing data type through the data of the relevant type of data Model, by entering missing time points Get the predicted value of missing pipeline data, combined with other pipelines in the same file at the missing time point The output value of the prediction model is used to comprehensively generate the missing data value of the pipeline. The specific steps are as follows: Missing data time point Substitute the relevant type of data in the missing data pipeline Prediction model to obtain the predicted value of missing pipeline data , and the missing time points Substitute the relevant type of data in other pipelines in the same group file Prediction model, obtain ,in The pipelines in the current group except the pipelines with missing data; Set weight value , ,in is the number of all pipelines in the current group. The missing data value of the pipeline is calculated based on the weight value. The specific algorithm formula is: ; in, This is the missing data value for the pipeline.

[0011] Furthermore, the gathering and transportation grouping module groups the daily operating pipelines based on the collected current oilfield gathering and transportation pipeline specifications and start and stop status. Establish a daily operation pipeline database, group the daily activated pipelines based on pipeline specifications, divide the same-model daily operation pipelines into the same file, and record the real-time collection and transportation data collected by the data acquisition module. The specific steps are as follows: Based on the collected current oilfield gathering and transportation pipeline specifications and the daily oilfield gathering and transportation pipeline task list, create archives for the activated oilfield gathering and transportation pipelines, and store the pipeline specifications in the archives, including pipeline diameter and pipeline material; For pipelines in operation on the day, pipelines with the same diameter and material are grouped together. Establish a daily operation pipeline database, and put the pipeline files grouped together into the same group in the daily operation pipeline database; Receive the real-time pipeline data collected by the sensor in the data acquisition module and record it in the archive of the corresponding pipeline.

[0012] The intelligent operation and maintenance device for gathering and transportation based on the missing data filling method is applied to the intelligent operation and maintenance system for gathering and transportation based on the missing data filling method, and includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the intelligent operation and maintenance system for gathering and transportation based on the missing data filling method described in the present invention are implemented.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. In the present invention, based on the gathering and transportation plan, the pipelines in operation on the day are grouped according to their specifications and models, and daily gathering and transportation data prediction equations are established for all pipelines in the same group, thereby avoiding the influence of external factors on pipeline data prediction and enhancing the accuracy of pipeline data prediction; 2. In the present invention, the constructed data prediction model is updated periodically based on a fixed time interval, so that the prediction model can adapt to changes in a timely manner, continuously and accurately reflect the latest characteristics and laws of the data, maintain good prediction performance, enhance the accuracy of subsequent missing data filling, and correct errors to make the model prediction results closer to the true value; 3. In the present invention, by combining the prediction results of missing data at missing time points of other prediction equations in the same group, combined with the missing data prediction value of the pipeline with missing data, the missing data value of the pipeline is obtained by comprehensive calculation, so as to fill the missing data of the pipeline with missing data. The combination of multiple prediction results can reduce the sensitivity of the model to extreme values ​​or abnormal inputs, and improve the robustness of the prediction results; 4. In the present invention, by grouping the pipelines activated on the day based on the specification type, and combining the results of other prediction equations in the same group and the prediction value of the pipeline with missing data for comprehensive calculation, multiple information sources can be fully utilized to avoid the limitations and deviations of a single prediction model, so that the filled value is closer to the true value, thereby improving the reliability of the data; 5. In the present invention, by combining the real-time data of each pipeline in the same group on the same day, a comprehensive judgment is made on the possible abnormal data values ​​based on the data of the same group on the same day, thereby avoiding the limitation of judging abnormal values ​​by historical data. Each pipeline in the same group on the same day is in a similar operating environment and conditions, and their data have strong correlation and comparability. Combining these real-time data to judge abnormal values ​​can make use of more information for judgment, thereby reducing the possibility of misjudgment; 6. In the present invention, abnormal values ​​are determined based on the real-time data of each pipeline in the same group on the same day. A quick comprehensive determination can be made based on the real-time data without waiting for the collection and analysis of a large amount of historical data. Decisions and responses can be made more quickly, which enhances adaptability and functionality. The entire intelligent operation and maintenance system and device for gathering and transportation based on the missing data filling method, through real-time data collection and the establishment of the daily prediction model, can intelligently manage the pipeline data in the oilfield gathering and transportation process and improve the operation and maintenance efficiency. Through the missing value filling module, the system can effectively handle the problem of missing data and ensure the integrity of the data and the continuous operation of the system. The real-time prediction module can dynamically update the prediction model, improve the accuracy and reliability of the prediction, and provide support for operation and maintenance decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a block diagram of the intelligent operation and maintenance system for gathering and transportation based on the missing data filling method of the present invention. DETAILED DESCRIPTION

[0015] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Example 1: Figure 1 As shown, the intelligent operation and maintenance system for gathering and transportation based on the missing data filling method includes a data acquisition module, a gathering and transportation grouping module, a real-time prediction module, and a missing value supplement module; The data acquisition module includes a pressure sensor, a temperature sensor, a flow sensor, and a water content sensor, which are used to collect real-time pressure data, temperature data, flow data, and crude oil water content data of the transportation pipeline during the oil field gathering and transportation process. At the same time, it collects oil field gathering and transportation benchmark data, including pipeline specifications and start and stop status, and transmits them to subsequent modules; The gathering and transportation grouping module groups the daily operating pipelines based on the current oilfield gathering and transportation pipeline specifications and start and stop status. Establish a daily operation pipeline database, group the daily activated pipelines based on pipeline specifications, divide the same-model daily operation pipelines into the same file, and record the real-time collection and transportation data collected by the data acquisition module. The specific steps are as follows: Based on the collected current oilfield gathering and transportation pipeline specifications and the daily oilfield gathering and transportation pipeline task list, create archives for the activated oilfield gathering and transportation pipelines, and store the pipeline specifications in the archives, including pipeline diameter and pipeline material; For pipelines in operation on the day, pipelines with the same diameter and material are grouped together. Establish a daily operation pipeline database, and put the pipeline files grouped together into the same group in the daily operation pipeline database; Receive the real-time pipeline data collected by the sensor in the data acquisition module and record it in the archive of the corresponding pipeline; The real-time prediction module, for the pipelines in the same archive, establishes a prediction model for the daily pressure, temperature, flow rate, and water content of each pipeline based on the real-time gathering and transportation data of each pipeline in the archive. Based on the real-time recorded pipeline data, the prediction model is verified and updated synchronously. The specific steps are as follows: For the pipeline data in the same group file in the daily operation pipeline database, based on the real-time data of each pipeline in operation on that day, pressure, temperature, flow rate and water content data prediction models are established respectively. The specific steps are as follows: For the pipeline data in the same group file, the pressure, temperature, flow rate, and water content data in each pipeline are timestamped. Based on each pipeline in the same group file, the data before and after the pipeline operation on that day are collected. The pressure, temperature, flow rate and water content data at the moment will be Always Divide equally, based on The model establishes the pressure, temperature, flow rate and water content data prediction models of different pipelines in the same group file respectively. The specific steps are as follows: Collect the data before and after the pipeline runs on the day The data at the moment is recorded as ,in ,in Respectively represent the pressure, temperature, flow rate and water content data of the current pipeline; Respectively The data at each moment is divided into equal parts, each containing the time steps , mark each piece of data as ; Each piece of data As input, each time step , set the input dimension of the model to 4, divide the data set into training set, validation set and test set, and use the mean square error as the loss function. The algorithm formula is: ; in, They represent the true value and the predicted value respectively. The Adam optimizer is used to further update the model parameters to obtain the pressure, temperature, flow rate, and water content data prediction model for different pipelines in the same group.

[0017] It should be noted that the training set is used for model parameter learning, the validation set is used to adjust hyperparameters, and the test set is used to evaluate model performance. The ratio is 70% training set, 15% validation set, and 15% test set. The time value is usually 120 minutes. The equal parts are usually set to 120, that is, the real-time pipeline pressure data, temperature data, flow data, and crude oil water content data obtained by the sensor are taken as a data point every minute. Time and The equal parts can also be adjusted according to the actual situation. The model can output the predicted values ​​of pressure, temperature, flow rate and water content in the future. The model establishes prediction models for pressure, temperature, flow, and moisture content data for different pipelines in the same group file, realizing real-time prediction and dynamic update of data. At the same time, the daily model is established based on the pipeline operation status of each day, which can unify the external influences on the pipeline caused by other external factors, such as weather and humidity, to ensure the accuracy of current data prediction.

[0018] Based on time interval Collect the real-time monitoring data of different pipelines in each archive, and conduct phase verification and update of the prediction model established on the day. The specific steps are as follows: Based on time interval Get the time points , at each time point The data in different pipeline files in the same group are extracted, including the actual values ​​of pipeline pressure, pipeline temperature, pipeline flow and water content data. ; The time point Input to the corresponding pipeline training of the day In the model, get the time point Current pipeline data predicted value , based on the mean absolute error method to calculate the actual value With the predicted value The error of is, and its algorithm formula is: ; in, Represents at a point in time The mean absolute error between the actual value and the predicted value at the time of right Make a judgment: when ≥ When , it means that the current pipeline data prediction model needs to be updated; when < When , it means that the current pipeline data prediction model does not need to be updated and continues to be used; against ≥ Time, collection time New data since then, based on Split into equal parts and retrain using the updated dataset model, and replace the old model with the new prediction model as the data prediction model for the current pipeline.

[0019] It should be noted that the mean absolute error threshold Different mean absolute error thresholds need to be set based on different data types. , respectively calculate the mean value, standard deviation and other statistics of the daily data of pressure, flow, temperature and water content, and use the empirical method to determine the mean absolute error threshold When the mean absolute error of a model on a certain variable exceeds a specific threshold, the stage update of the model related to the variable is triggered, so as to keep the prediction accuracy of the model at all times. It needs to be set according to actual usage. It is usually set to 20 minutes, and can be increased or decreased as needed.

[0020] The real-time prediction module also has a daily anomaly detection module, which detects the same type of pipeline data in the same group. Perform real-time anomaly detection at all times. The specific steps are as follows: Collect the current day’s data from all pipeline files in the same group. The real-time pressure, temperature, flow rate, and water content data are collected every minute before the moment, and the original data sets are divided based on the data type. The abnormal parameters of each data set are analyzed using the isolation forest method. The specific steps are as follows: Randomly extract a sample subset from the original data set, the size is , for each tree starting from the root node, randomly select a feature and a random split value on the feature to divide the data set into left and right child nodes, and recursively repeat the above process for each child node until the preset tree depth is reached or there is only one data point in the child node; For each data point , calculate the path length in each tree , average the path lengths of all trees to get the average path length , calculate the anomaly score , and its algorithm formula is: ; ; ; in, Represents the total number of samples, that is, the number of all data points involved in anomaly detection, is the sample constant, Representative A harmonic number, when When the value is greater than the detection threshold, it means that the pipeline data points collected on that day are abnormal. The abnormal pipelines are marked in the same group file, and the abnormal data type of the abnormal pipelines is The model is isolated and the output value is not transmitted to the subsequent modules.

[0021] It should be noted that in the isolation forest, the anomaly score The closer it is to 1, the more likely the data point is an outlier. The closer it is to 0, the more likely it is normal data. The detection threshold is generally set to 0.8, and can also be adjusted between 0.75 and 0.93 according to actual conditions. In the moment Value Needs and Construction Model The value is consistent with the abnormal data type of the abnormal pipeline. The model is isolated and the output value is not transmitted to the subsequent modules, which can reduce the error when supplementing the missing values ​​later.

[0022] Embodiment 2: A missing value supplement module is based on the pipeline data prediction model in the same file. The missing data of the pipeline is filled in by combining the output value of the prediction model of the missing data with the output value of the prediction model of other pipelines in the same file. The specific steps are as follows: Collect real-time monitoring data of all pipelines in the same file and determine the time point when data is missing And the missing data pipeline, based on the missing data type through the data of the relevant type of data Model, by entering missing time points Get the predicted value of missing pipeline data, combined with other pipelines in the same file at the missing time point The output value of the prediction model is used to comprehensively generate the missing data value of the missing pipeline. The specific steps are as follows: Missing data time point Substitute the relevant type of data in the missing data pipeline Prediction model to obtain the predicted value of missing pipeline data , and the missing time points Substitute the relevant type of data in other pipelines in the same group file Prediction model, obtain ,in The pipelines in the current group except the pipelines with missing data; Set weight value , ,in is the number of all pipelines in the current group. The missing data value of the pipeline is calculated based on the weight value. The specific algorithm formula is: ; in, This is the missing data value for the pipeline.

[0023] It should be noted that It is generally set to 0.5, and the remaining weight values ​​are 0.5, which is based on the average value of the number of pipelines in the current group except for the pipelines with missing data. By combining the prediction results of the missing data at the missing time point by other prediction equations in the same group, combined with the predicted value of the missing data of the pipeline with missing data, the missing data value of the pipeline is obtained by comprehensive calculation, so as to fill the missing data of the pipeline with missing data and ensure the integrity and accuracy of the data.

[0024] Embodiment 3: A gathering and transportation intelligent operation and maintenance device based on a missing data filling method is applied to a gathering and transportation intelligent operation and maintenance system based on a missing data filling method, comprising a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the gathering and transportation intelligent operation and maintenance system based on a missing data filling method of the present invention are implemented.

[0025] The intelligent operation and maintenance system and device for gathering and transportation based on the missing data filling method of the present invention, when in use, uses pressure sensors, temperature sensors, flow sensors, and water content sensors to collect real-time pressure data, temperature data, flow data, and crude oil water content data of the transportation pipeline during the gathering and transportation process of the oil field, and groups the pipelines based on the daily oil field gathering and transportation activation and the oil field specifications, and establishes a prediction model for the pipelines in the group in combination with the real-time data; The prediction model is updated periodically based on the actual measured values. At the same time, all pipeline prediction models in the same group are integrated to perform comprehensive calculations on the missing data to obtain the missing data of the pipelines with missing data.

[0026] In the embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation; the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the method of this embodiment.

[0027] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method is characterized by: It includes data collection module, gathering and transportation grouping module, real-time prediction module and missing value supplement module; The data acquisition module includes a pressure sensor, a temperature sensor, a flow sensor, and a water content sensor, which are used to collect real-time pressure data, temperature data, flow data, and crude oil water content data of the transportation pipeline during the oil field gathering and transportation process, and collect oil field gathering and transportation benchmark data, including pipeline specifications and start and stop status, and transmit them to subsequent modules; The gathering and transportation grouping module groups the daily operating pipelines based on the collected current oilfield gathering and transportation pipeline specifications and start and stop status. Establish a daily operation pipeline database, group the pipelines activated daily based on pipeline specifications, divide the pipelines of the same model in operation on the same day into the same file, and record the real-time collection and transportation data collected by the data acquisition module; The real-time prediction module, for the pipelines in the same archive, establishes a prediction model for the daily pressure, temperature, flow rate, and water content of each pipeline based on the real-time gathering and transportation data of each pipeline in the archive, and synchronously verifies and updates the prediction model based on the real-time recorded pipeline data; The missing value supplement module is based on the pipeline data prediction model in the same file, and comprehensively fills the missing data of the missing pipeline by combining the output value of the prediction model of the missing data with the output value of the prediction model of other pipelines in the same file.

2. The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method according to claim 1 is characterized by: The real-time prediction module, for the pipelines in the same archive, establishes a prediction model for the daily pressure, temperature, flow rate, and water content of each pipeline based on the real-time gathering and transportation data of each pipeline in the archive, and synchronously verifies and updates the prediction model based on the real-time recorded pipeline data. The specific steps are as follows: For the pipeline data in the same group file in the daily operation pipeline database, based on the real-time data of each pipeline during the day's operation, pressure, temperature, flow rate, and water content data prediction models are established respectively; Based on time interval Collect real-time monitoring data of different pipelines in each archive, and conduct phased verification and update of the prediction model established on the day.

3. The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method according to claim 2 is characterized in that: The pipeline data in the same group file in the daily operation pipeline database are used to establish pressure, temperature, flow rate and water content data prediction models based on the real-time data of each pipeline in operation on that day. The specific steps are as follows: For the pipeline data in the same group file, the pressure, temperature, flow rate, and water content data in each pipeline are timestamped. Based on each pipeline in the same group file, the data before and after the pipeline operation on that day are collected. The pressure, temperature, flow rate and water content data at the moment will be Always Divide equally, based on The model establishes the pressure, temperature, flow rate and water content data prediction models of different pipelines in the same group file respectively. The specific steps are as follows: Collect the data before and after the pipeline runs on the day The data at the moment is recorded as ,in ,in Respectively represent the pressure, temperature, flow rate and water content data of the current pipeline; Respectively The data at each moment is divided into equal parts, each containing the time steps , mark each piece of data as ; Each piece of data As input, each time step , set the input dimension of the model to 4, divide the data set into training set, validation set and test set, and use the mean square error as the loss function. The algorithm formula is: ； in, They represent the true value and the predicted value respectively. The Adam optimizer is used to further update the model parameters to obtain the pressure, temperature, flow rate, and water content data prediction model for different pipelines in the same group.

4. The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method according to claim 3 is characterized by: The time interval based Collect the real-time monitoring data of different pipelines in each file, and conduct phase verification and update of the prediction model established on the day. The specific steps are as follows: Based on time interval Get the time points , at each time point The data in different pipeline files in the same group are extracted, including the actual values ​​of pipeline pressure, pipeline temperature, pipeline flow and water content data. ; The time point Input to the corresponding pipeline training of the day In the model, get the time point Current pipeline data predicted value , based on the mean absolute error method to calculate the actual value With the predicted value The error of is, and its algorithm formula is: ； in, Represents at a point in time The mean absolute error between the actual value and the predicted value at the time of right Make a judgment: when ≥ When , it means that the current pipeline data prediction model needs to be updated; when < When , it means that the current pipeline data prediction model does not need to be updated and continues to be used; against ≥ Time, collection time New data since then, based on Split into equal parts and retrain using the updated dataset model, and replace the old model with the new prediction model as the data prediction model for the current pipeline.

5. The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method according to claim 4 is characterized in that: The real-time prediction module is also provided with a daily anomaly detection module, which detects the anomaly of the day based on the same type of pipeline data in the same group. Perform real-time anomaly detection at all times. The specific steps are as follows: Collect the current day’s data from all pipeline files in the same group. The real-time pressure, temperature, flow rate, and water content data are collected every minute before the moment, and the original data sets are divided based on the data type. The abnormal parameters of each data set are analyzed using the isolation forest method. The specific steps are as follows: Randomly extract a sample subset from the original data set, the size is , for each tree starting from the root node, randomly select a feature and a random split value on the feature to divide the data set into left and right child nodes, and recursively repeat the above process for each child node until the preset tree depth is reached or there is only one data point in the child node; For each data point , calculate the path length in each tree , average the path lengths of all trees to get the average path length , calculate the anomaly score , and its algorithm formula is: ； ； ； in, Represents the total number of samples, that is, the number of all data points involved in anomaly detection, is the sample constant, Representative A harmonic number, when When the value is greater than the detection threshold, it means that the pipeline data points collected on that day are abnormal. The abnormal pipelines are marked in the same group file, and the abnormal data type of the abnormal pipelines is The model is isolated and the output value is not transmitted to the subsequent modules.

6. The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method according to claim 2 is characterized in that: The missing value supplement module is based on the pipeline data prediction model in the same file, and comprehensively fills the missing data of the pipeline with missing data by combining the prediction model output value of the pipeline with missing data with the prediction model output values ​​of other pipelines in the same file. The specific steps are as follows: Collect real-time monitoring data of all pipelines in the same file and determine the time point when data is missing And the missing data pipeline, based on the missing data type through the data of the relevant type of data Model, by entering missing time points Get the predicted value of missing pipeline data, combined with other pipelines in the same file at the missing time point The output value of the prediction model is used to comprehensively generate the missing data value of the missing pipeline.

7. The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method according to claim 6 is characterized in that: The real-time monitoring data of all pipelines in the same file are collected to determine the time point when the data is missing And the missing data pipeline, based on the missing data type through the data of the relevant type of data Model, by entering missing time points Get the predicted value of missing pipeline data, combined with other pipelines in the same file at the missing time point The output value of the prediction model is used to comprehensively generate the missing data value of the pipeline. The specific steps are as follows: Missing data time point Substitute the relevant type of data in the missing data pipeline Prediction model to obtain the predicted value of missing pipeline data , and the missing time points Substitute the relevant type of data in other pipelines in the same group file Prediction model, obtain ,in The pipelines in the current group except the pipelines with missing data; Set weight value , ,in is the number of all pipelines in the current group. The missing data value of the pipeline is calculated based on the weight value. The specific algorithm formula is: ； in, This is the missing data value for the pipeline.

8. The intelligent operation and maintenance system for gathering and transportation based on the missing data filling method according to claim 1 is characterized by: The gathering and transportation grouping module groups the daily operating pipelines based on the collected current oilfield gathering and transportation pipeline specifications and start and stop status. Establish a daily operation pipeline database, group the daily activated pipelines based on pipeline specifications, divide the same-model daily operation pipelines into the same file, and record the real-time collection and transportation data collected by the data acquisition module. The specific steps are as follows: Based on the collected current oilfield gathering and transportation pipeline specifications and the daily oilfield gathering and transportation pipeline task list, create archives for the activated oilfield gathering and transportation pipelines, and store the pipeline specifications in the archives, including pipeline diameter and pipeline material; For pipelines in operation on the day, pipelines with the same diameter and material are grouped together. Establish a daily operation pipeline database, and put the pipeline files grouped together into the same group in the daily operation pipeline database; Receive the real-time pipeline data collected by the sensor in the data acquisition module and record it in the archive of the corresponding pipeline.

9. A gathering and transportation intelligent operation and maintenance device based on a missing data filling method, characterized in that: The device is applied to a gathering and transportation intelligent operation and maintenance system based on a missing data filling method, and includes a memory and a processor: a memory for non-transitory storage of computer readable instructions; a processor for executing the computer-readable instructions; Wherein, when the computer-readable instructions are executed by the processor, the system described in any one of claims 1-8 is executed.

Images