Oil and gas field data collecting and processing method, device and system based on swan mongolian system

Abstract

The invention relates to the technical field, in particular to an oil and gas field data collecting and processing method, device and system based on a swan mongolian system, and the method comprises the steps: inputting a network delay record, a bandwidth occupation proportion and a data packet size sequence of a wellhead collecting end into a long and short-term memory network according to a time sequence in a data processing process, and carrying out the recursive calculation; according to the method, time sequence modeling and abnormity judgment of a delay trend are realized, time nodes with over-limit risks are identified, and priority selection of high-value nodes and binding management of link identifiers are effectively realized by utilizing a synchronous comparison method of delay values and bandwidth proportions and combining descending sorting of data packet sizes, so that the reliability of the link identifiers is improved. Node feature aggregation of a graph neural network is introduced when oil pipeline monitoring node delay and bandwidth values are combined, targeted optimization is carried out in combination with data features and service requirements in each link, and the acquisition and processing process has good performance improvement in the aspects of delay control, link resource optimization, task scheduling efficiency and the like.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field, in particular to an oil and gas field data acquisition and processing method, device and system based on a Hongmeng system. BACKGROUND

[0002] The data acquisition technical field includes a series of processes for acquiring, transmitting, receiving and storing data from a specified source by using a computer system and an operating platform, specifically, the data is automatically, batch-wise and periodically acquired through a software running environment and a data input interface, in the oil and gas field application scenario, the field includes acquisition of multiple real-time and timing data such as wellhead production parameters, pressure, temperature, flow and equipment running state, and reliable transmission of the data to a processing end through network communication to support subsequent storage and calculation processing, the field requires that the acquisition system has compatibility with a bottom layer device protocol, data format analysis capability and fault tolerance capability of abnormal acquisition.

[0003] The oil and gas field data acquisition and processing method based on the Hongmeng system refers to a data acquisition and processing program running in the Hongmeng operating system environment, which is used for acquiring multiple production and running data from oil and gas field production and monitoring equipment, and performing data organization, storage and calculation processes in a local system and a remote server, the purpose is to realize centralized management and high consistency recording of the oil and gas field production data, to ensure that the acquisition data is not lost and tampered in the transmission and processing process, and to provide complete and reliable data basis for oil and gas field production scheduling, equipment maintenance, safety monitoring and production optimization, so as to improve the informatization level of the oil and gas field production, shorten the fault response time and improve the accuracy of production operation decision basis.

[0004] Although the existing data acquisition and processing mode can complete the acquisition and transmission of wellhead production parameters, pressure, temperature, flow and equipment running state in the oil and gas field environment, it is driven by a fixed period and static logic in the data processing chain, lacks active prediction and targeted adjustment of the delay trend of the acquisition task, and causes that the delay-sensitive data cannot be transmitted in time when the network fluctuates and multiple tasks are highly concurrent, in node optimization, the existing scheme usually relies on a single parameter to screen the node state, cannot fully associate multiple business parameter information, and may cause that the key node cannot be preferentially guaranteed when the link resource is tight, in network path selection, due to the lack of multi-dimensional fusion of node and adjacent node state information, the high-load link is easily not identified and inefficiently shunted, in task scheduling, a static allocation strategy is usually used, and the dynamic balance of the current task amount of the core is not considered, which easily causes the core to be overloaded, causes the key monitoring data to be delayed, the link resource utilization rate to be reduced and the overall operation efficiency to be reduced. SUMMARY

[0005] The application aims to solve the problems in the prior art and provides an oil and gas field data acquisition and processing method, device and system based on a hyper system.

[0006] To achieve the above-mentioned purpose, the application adopts the following technical scheme: an oil and gas field data acquisition and processing method based on a hyper system, comprising the following steps: Step one: inputting each cycle parameter in time order into a long short-term memory network to generate a trend value and calculate a difference value, judging an overrun and marking by synchronously comparing the original sequence, generating a delay warning node set through a wellhead acquisition end network delay record, bandwidth occupation ratio and data packet size sequence; Step two: based on the delay warning node set, reading a delay value and bandwidth ratio and synchronously comparing and screening, extracting a node with large delay and high bandwidth and rearranging in descending order of data packet size, recording a communication link identifier and obtaining a priority link node table; Step three: based on the priority link node table, adding a weight value formed by a pipeline monitoring node link delay and bandwidth value in a node feature aggregation unit of a graph neural network, judging a high-load link and screening a shunt node, and generating a shunt path distribution table; Step four: based on the shunt path distribution table, reading a compressor station monitoring unit task type, preferentially inserting a safety monitoring task into an execution queue, comparing a running core queue length to select a valley value queue core to allocate a task, and obtaining a core execution task table; Step five: based on the core execution task table, reading each core task set to compare adjacent execution times, grouping tasks with a time difference less than a synchronization threshold value into a synchronization group and inserting a high-priority task, and integrating records to form a synchronization priority scheduling table.

[0007] As a further scheme of the application, the specific steps for generating the delay warning node set are as follows: Arranging three types of parameters in each acquisition cycle in time order in turn, subtracting the same parameter values in adjacent cycles in turn to obtain a difference value, and storing the difference value and time number in a paired manner in an ordered list to generate a cycle difference data set through a network delay record, bandwidth occupation ratio and data packet size sequence of a wellhead acquisition end; Based on the cycle difference data set, reading the difference value and corresponding three types of original parameter values in time number in turn, inserting the difference value into the end of a parameter group at the same time node to form a complete parameter unit, inputting the parameter units formed at each time node into an input layer of a long short-term memory network in order and recursively calculating an output in time order, fusing the output result and the original parameter unit at the same time node, and storing them in a new sequence in continuous order to generate a composite trend sequence; Based on the composite trend sequence, the delay values of each time node are sequentially read and compared with the set threshold value, the time nodes with delay values greater than the threshold value are marked and written into the marking list, the marked nodes are combined with the parameter unit to generate the target set, and the delay warning node set is generated.

[0008] As a further scheme of the application, the process of the long short-term memory network is as follows: after sequentially reading the difference values and corresponding three types of original parameter values according to time numbers, the difference values are inserted into the end of the parameter group of the same time node to form a complete parameter unit, the values of all parameter units are arranged according to time sequence and input into the input layer of the memory network, the input of each time node is sequentially processed in the implicit storage unit, and recursive operation is performed in combination with the state value of the previous time node, the current operation result is fused with the parameter unit of the corresponding time node item by item, and each time node is output in the time sequence of the fusion result and added to the newly created sequence to form a data set for generating a composite trend sequence.

[0009] As a further scheme of the application, the specific steps for generating the priority link node table are as follows: Based on the delay warning node set, the delay values and bandwidth occupation ratios of each node are read in time sequence, the nodes with high delay values and large bandwidth ratios are selected by comparison, the node numbers are matched with the corresponding data packet sizes and stored in the list, and the high delay high bandwidth node set is generated. Based on the high delay high bandwidth node set, the nodes are sorted by data packet size from large to small, the communication link identifiers of each node in the sorting result are extracted and sequentially inserted into the link list, the link list is stored corresponding to the node number, and the priority link node table is generated.

[0010] As a further scheme of the application, the specific steps for generating the shunt path allocation table are as follows: Based on the priority link node table, the delay values and bandwidth values of each link in the oil pipeline monitoring node are read, the two types of values of the same node and the corresponding values of adjacent nodes are added in the node feature aggregation unit of the graph neural network in time sequence to obtain merged values, and the merged values obtained by operation are written into the data column corresponding to the node and stored in the set in the node time sequence to generate the link weight value set. Based on the link weight value set, the weight values of each node are read and compared with the set load threshold value, the node numbers with weight values greater than the threshold value are written into the high load list, the candidate link data corresponding to the nodes in the list are obtained, and the table is formed according to the corresponding relationship between the nodes and the candidate link information to generate the high load candidate link table. Based on the high-load candidate link table, the delay values of each node candidate link are compared to select a link with a valley value delay, when the delays are equal, the bandwidth values are compared to select a link with a peak value bandwidth, the node number and the selected link are combined into a pairing record, and the record order is added to the result set to generate a shunt path allocation table.

[0011] As a further scheme of the application, the graph neural network execution process is based on the priority link node table, the link delay values and bandwidth values of the oil pipeline monitoring nodes and adjacent nodes are loaded as graph structure inputs according to node numbers, the delay values and bandwidth values of each node are combined to form a feature vector in time sequence, which is transmitted to the node feature aggregation unit of the network, and the feature vectors of directly connected nodes are paired and summed one by one in the unit, and the summed results are written into the feature storage area of the corresponding node, and the node features are rearranged according to time numbers and stored in a set as input data sources of the link weight value set.

[0012] As a further scheme of the application, the specific steps for generating the core execution task table are: Based on the shunt path allocation table, the compressor station monitoring unit task type is read according to the node number, the safety monitoring type node is extracted and inserted into the head end of the execution queue in order of task number, and the non-safety tasks are appended to the end of the queue in the original number order, and a corresponding record table of task number and queue position is generated to generate a priority task queue table; Based on the priority task queue table, the number of tasks in the current queue of each running core is counted, the core with the valley value of the task number is identified as the valley queue core, the tasks in the priority task queue table are distributed to the core according to the node number order, and the distributed task number and core number are recorded in the distribution list to generate the core execution task table.

[0013] As a further scheme of the application, the specific steps for generating the synchronization priority scheduling table are: Based on the core execution task table, the task set is read according to the core number in turn, the task execution time in each set is arranged in order, the adjacent task time values are subtracted and compared with the synchronization threshold value, the task numbers with a time difference less than the threshold value are selected, the selected numbers are grouped according to the core number and added to the grouping list to generate a synchronization task grouping table; Based on the synchronization task grouping table, the core execution queue position corresponding to each group of task numbers is obtained, each group of tasks is moved to a high priority task while maintaining the original order of the tasks in the group, the adjusted queue and the synchronization task group number are correspondingly established to generate a data record table, and the data record table is combined to generate a synchronization priority scheduling table.

[0014] The oil and gas field data acquisition and processing device based on the Hongmeng system comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the oil and gas field data acquisition and processing method based on the Hongmeng system when executing the computer program.

[0015] The oil and gas field data acquisition and processing system based on the Hongmeng system is used for executing the oil and gas field data acquisition and processing method based on the Hongmeng system, and comprises: The trend analysis module: based on the wellhead acquisition end delay record, the bandwidth occupation ratio and the data packet size, the periodic data is sequentially input into the long short-term memory network to generate a trend value, and the difference value between the original data is compared to mark the over-limit time node, and a delay warning node set is generated; The preferred node module: based on the delay warning node set, the node delay value and the bandwidth ratio are read, the nodes with high delay and large bandwidth are screened, the data packet size is arranged in descending order, and the communication link identifier is recorded to form a priority link node table; The link allocation module: based on the priority link node table, the node delay and bandwidth are input into the graph neural network to generate a combined value, the high-load nodes are screened by comparing the threshold value, the qualified path is selected according to the delay and bandwidth, and a shunt path allocation table is obtained; The core scheduling module: based on the shunt path allocation table, the task type is read, the safety monitoring task is put into the front end of the execution queue, the number of core tasks is counted, the core with the valley value of the task number is selected for task allocation, and a core execution task table is obtained; The synchronization arrangement module: based on the core execution task table, the adjacent time difference of each core task is compared, the tasks with a time difference less than a threshold value are classified into a synchronization group, a high-priority task is inserted after the synchronization group, and a synchronization priority scheduling table is generated.

[0016] Compared with the prior art, the advantages and positive effects of the present application are: 1. In the present application, the network delay record, the bandwidth occupation ratio and the data packet size sequence of the wellhead acquisition end are input into the long short-term memory network in time sequence for recursive calculation in the data processing process, the time sequence modeling and abnormal judgment of the delay trend are realized, and the time nodes with over-limit risk are identified in advance; 2. In the present application, the key nodes with large delay and high bandwidth are screened by using the synchronous comparison method of delay value and bandwidth ratio, and the descending order sorting of data packet size is combined to effectively realize the priority selection of high-value nodes and the binding management of link identifier, so that the delay-sensitive tasks can be sequentially connected in the overall running link; 3. In the present application, by introducing the node feature aggregation of the graph neural network when combining the delay and bandwidth value of the oil pipeline monitoring node, and combining the data characteristics and business needs at each link for targeted optimization, the collection and processing process has good performance improvement in delay control, link resource optimization and task scheduling efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a workflow diagram of the present application; Figure 2 is a system flowchart of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical scheme and advantages of the present application more clear and explicit, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0019] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0020] Example 1 Please refer to Figure 1 The present application provides a technical scheme: an oil and gas field data collection and processing method based on a Hongmeng system, comprising the following steps: Step 1: Through the network delay record, bandwidth occupation ratio and data packet size sequence of the wellhead collection end, input the periodic parameters in time order into the long short-term memory network to recursively generate trend values and calculate adjacent difference values, and judge the out-of-limit and mark by synchronously comparing with the original sequence, to generate a delay warning node set; Step 2: Based on the delay warning node set, read the delay value and bandwidth ratio and synchronously compare and screen, extract the nodes with large delay and high bandwidth and reorder them in descending order of data packet size, record the communication link identifier and obtain a priority link node table; Step 3: Based on the priority link node table, add the link delay and bandwidth value of the oil pipeline monitoring node in the node feature aggregation unit of the graph neural network to form a weight value, judge the high-load link and screen the shunt nodes, and generate a shunt path distribution table; Step four: based on the shunt path distribution table, read the compressor station monitoring unit task type, insert the safety monitoring task into the execution queue, compare the running core queue length to select the valley value queue core task allocation, obtain the core execution task table; Step five: based on the core execution task table, read each core task set and compare the adjacent execution time, group the tasks with a time difference less than the synchronization threshold into a synchronization group and insert high-priority tasks, integrate the records to form a synchronization priority scheduling table.

[0021] The specific steps for generating the delay warning node set are: Through the network delay record, bandwidth occupation ratio, and data packet size sequence of the wellhead collection end, arrange the three types of parameters in each collection period in time sequence, subtract each adjacent period parameter value to obtain the difference value, store the difference value and time number pair according to the difference value and time number, generate a period difference value data set in an ordered list; Based on the period difference value data set, read the difference value and corresponding three types of original parameter values in time number, insert the difference value into the end of the parameter group at the same time node to form a complete parameter unit, input the parameter units formed at each time node into the input layer of the long short-term memory network in sequence and recursively calculate the output, fuse the output result with the original parameter unit at the same time node, and store it in a new sequence in continuous order to generate a composite trend sequence; Based on the composite trend sequence, read the delay value of each time node in sequence and compare it with the set threshold, mark the time nodes with delay values greater than the threshold and write them into a mark list, combine the marked nodes with the parameter units to generate a target set, and generate a delay warning node set; Through the network delay record, bandwidth occupation ratio, and data packet size sequence of the wellhead collection end, arrange the three types of parameters in each collection period in time sequence, subtract each adjacent period parameter value to obtain the difference value, store the difference value and time number pair according to the difference value and time number, generate a period difference value data set in an ordered list; Based on the period difference value data set, read the difference value and corresponding three types of original parameter values in time number, insert the difference value into the end of the parameter group at the same time node to form a complete parameter unit, input the parameter units formed at each time node into the input layer of the long short-term memory network in sequence and recursively calculate the output, fuse the output result with the original parameter unit at the same time node, and store it in a new sequence in continuous order to generate a composite trend sequence; Based on the composite trend sequence, the delay values ​​of each time node are read sequentially and compared with a set threshold of 200 milliseconds. The time node numbers with delay values ​​greater than the threshold are recorded as target numbers, and the complete data units corresponding to the numbers are read. Then, the time numbers and data units are combined into a two-element structure in the order of time number and data unit. The two-element structure is added to the set in order to generate a delay warning node set.

[0022] The execution process of the Long Short-Term Memory Network is as follows: after reading the difference and the corresponding three types of original parameter values ​​in sequence according to the time number, the difference is inserted into the end of the parameter group at the same time node to form a complete parameter unit. The values ​​of all parameter units are arranged in groups according to the time order and input into the input layer of the memory network. The input of each time node is processed sequentially in the hidden storage unit, and the recursive operation is performed in combination with the state value of the previous time node. The current operation result is fused with the parameter unit of the corresponding time node item by item. The time order of the fusion result is maintained, and each time node is output and added to the newly created sequence to form a data set for generating composite trend sequences. Long Short-Term Memory (LSTM) networks, according to the formula:

[0023] in: Indicates the current time step The hidden state output vector. Indicates the previous time step The hidden state output vector. This represents the weight matrix from the hidden state to the hidden state. This represents the weight matrix input to the hidden state. Indicates the current time step The input vector, The weight matrix represents the time interval parameter to the hidden state. Indicates the current time step Compared with the previous time step The time interval scalar, The weight matrix representing the link load ratio parameter to the implicit state. Indicates the current time step The link load ratio scalar. Represents the bias vector; Execution process: First, the delay difference, bandwidth usage difference, and data packet size difference for the current period are collected from the wellhead acquisition terminal to form the input vector. Obtain the hidden state of the previous time step. The data is stored in a vector, and the time step interval is calculated by subtracting the acquisition time of the previous cycle from the acquisition time of the current cycle. , and the instantaneous bandwidth usage ratio of the current link is measured synchronously in milliseconds, and the load ratio is obtained by dividing the current bandwidth usage by the maximum bandwidth capacity , and the is multiplied by the weight matrix to obtain the historical contribution of the hidden layer, and the is multiplied by the weight matrix to obtain the contribution of the real-time input of the current period, and the is multiplied by the weight matrix to obtain the weighted contribution of the time interval signal, and the is multiplied by the weight matrix to obtain the weighted contribution of the link load signal, and the above results are added element by element with the bias vector to form a comprehensive vector, and the new hidden state is calculated element by element using the hyperbolic tangent function .

[0024] The specific steps for generating the priority link node table are as follows: Based on the delay warning node set, read the delay value and bandwidth occupation ratio of each node in time sequence, select the node with high delay value and large bandwidth ratio by comparing the two, match the node number with the corresponding data packet size and store it in the list to generate a high delay and high bandwidth node set; Based on the high delay and high bandwidth node set, sort the nodes by packet size from large to small, extract the communication link identifier of each node in the sorting result and insert it into the link list in turn, store the link list corresponding to the node number, and generate the priority link node table; Based on the delay warning node set, read the delay value and bandwidth occupation ratio of each node in time sequence, set the delay threshold to one hundred and fifty milliseconds and the bandwidth occupation ratio threshold to 75%, judge the values of the delay value of each node and the delay threshold, and the bandwidth ratio and the bandwidth threshold, determine whether it meets the screening standard through logical and operation of the two conditions, extract the node number that meets the condition, record the corresponding data packet size as an integer value, and combine it with the node number into a paired item, write it into the list container in turn according to the input time sequence, and generate a high delay and high bandwidth node set; Based on the high delay and high bandwidth node set, sort the nodes by packet size from large to small, use the packet size as the primary sorting key and the node number as the auxiliary key, compare the sizes of adjacent elements in the initial node array and exchange them until the whole array is sorted in descending order, maintain the original node number order synchronously during the sorting process, read the communication link identifier of the sorted node and add it to the link list in turn according to the sorting order, establish the corresponding relationship between the link list and the node number in the mapping table, record the link identifier corresponding to each node number through the mapping table, and generate the priority link node table.

[0025] The specific steps for generating the shunt path allocation table are as follows: Based on the priority link node table, the delay value and the bandwidth value of each link in the oil pipeline monitoring node are read, and the two types of values of the same node and the corresponding values of the adjacent nodes are sequentially added in the node feature aggregation unit of the graph neural network to obtain a merged value, and the merged value is written into the node corresponding data column and stored in a set in the node time sequence to generate a link weight value set; Based on the link weight value set, the weight value of each node is read and compared with the set load threshold value, the node number with a weight value greater than the threshold value is written into a high load list, the candidate link data corresponding to the node in the list is obtained, and a table is formed according to the node and the candidate link information according to the corresponding relationship to generate a high load candidate link table; Based on the high load candidate link table, the delay values of the candidate links of each node are compared to select the link with the minimum delay value, and when the delay values are equal, the bandwidth values are compared to select the link with the maximum bandwidth value, the node number and the selected link are combined into a pairing record, and the record is sequentially added to a result set to generate a shunt path allocation table; Based on the priority link node table, the delay value and the bandwidth value of each link in the oil pipeline monitoring node are read, and the two types of values of the same node and the corresponding values of the adjacent nodes are sequentially added in the node feature aggregation unit of the graph neural network to obtain a merged value, and the merged value is written into the node corresponding data column and stored in a set in the node time sequence to generate a link weight value set; Based on the link weight value set, the weight value of each node is read and compared with the set load threshold value, the node number with a weight value greater than the threshold value is written into a high load list, the candidate link data corresponding to the node in the list is obtained, and a table is formed according to the node and the candidate link information according to the corresponding relationship to generate a high load candidate link table; Based on the high load candidate link table, the delay values in the candidate link set are first compared, the delay comparison direction is set to be arranged from small to large, the link with the minimum delay value is selected as the target link record of the current node, when there are multiple links with equal delay values, the second condition comparison bandwidth value is entered, the bandwidth comparison direction is set to be arranged from large to small, and the link with the maximum bandwidth value is selected as the target link of the node, the node number and the selected link data are combined into a pairing entry, and sequentially added to a new result set in the node number ascending order to generate a shunt path allocation table.

[0026] The execution process of the graph neural network is as follows: Based on the priority link node table, the link delay values ​​and bandwidth values ​​of the monitoring nodes of the oil pipeline and their adjacent nodes are loaded as graph structure input according to the node number. The delay values ​​and bandwidth values ​​of each node are combined in time order to form a feature vector, which is passed to the node feature aggregation unit of the network. The feature vectors of the directly connected nodes are paired one by one in the unit and summed. The summing result is written into the feature storage area of ​​the corresponding node. The node features are rearranged according to the time number and stored in the set as the input data source of the link weight value set. Graph neural networks, according to the formula:

[0027] in: Indicates the first The updated feature vector of each node in the current aggregation round Indicates the node number currently being processed. Indicates the first The feature vector of each node in the previous aggregation round This indicates the time step corresponding to the previous node. This represents the weight matrix from the node's feature mapping to the current round, with subscripts indicating the weights. This represents the feature mapping of the current node itself. Indicates the first The set of adjacent nodes of a node. This represents the weight matrix that maps the features of adjacent nodes to the current node, with subscripts... This represents the characteristic part of adjacent nodes. Indicates adjacent nodes The previous round of feature vectors, Indicates that the neighboring nodes are in the set The index position in Represents the input feature mapping matrix, with subscripts... Represents external input features. Indicates the first The input feature vector of each node in the current round, Represents the time interval mapping matrix, with subscripts Represents time, element Indicates the first The time weights corresponding to each feature dimension Indicates the first The data collection time interval between each node and the previous round of calculation Represents the bias vector, elements Indicates adding to the first A constant term for each feature dimension. Represents a nonlinear activation function; Execution process: first, generate network topology according to the priority link node table currently constructed, associate each node with a set of adjacent nodes Establish connection relationship and generate input feature vector , containing the delay value, bandwidth occupation ratio and packet size collected by the node in real time, read the feature vector of the node itself in the last round , and multiply it by matrix to get the mapping value of the current round self-feature, traverse each neighbor node in the set , take the feature vector of the last round , multiply it by matrix to get the neighbor contribution, then multiply the input feature of the node by matrix to get the contribution of external input feature, and sum up the two according to the neighbor nodes in turn, read the time interval of the current node , multiply it by matrix to get the time feature contribution, and finally add the self-feature contribution, neighbor feature contribution, input feature contribution, time feature contribution and bias vector in the feature dimension to form a comprehensive vector.

[0028] The specific steps for generating the core execution task table are as follows: Based on the shunt path allocation table, read the compressor station monitoring unit task type according to the node number, extract the safety monitoring type node and insert it into the execution queue head end in order of task number, append the non-safety task in the original number order at the end of the queue, and generate a corresponding record table of task number and queue position, generate the priority task queue table; Based on the priority task queue table, count the number of tasks in the current queue of each running core, identify the core with the valley value as the valley queue core when the task number is the valley value, distribute the tasks in the priority task queue table to the core according to the node number order, and record the distributed task number and core number in the distribution list, generate the core execution task table; Based on the shunt path allocation table, read the compressor station monitoring unit task type according to the node number, set the classification condition as the task type field value equal to safety monitoring, read the task number in order from small to large according to the task number, and insert it into the execution queue head end, and the tasks that do not meet the condition are added to the end of the queue in the original number order, generate a corresponding list of node number and queue position, the position in the list is recorded in the order of zero-based integer sequence number, generate the priority task queue table; ​Based on the priority task queue list, the number of tasks in the current queue of each running core is counted, the core number and the current number of tasks are paired to form a key-value record, and the core number is sorted from small to large according to the number of tasks. The core number ranked first is selected as the valley value queue core. The tasks in the priority task queue list are sequentially assigned to the core according to the node number from small to large. When assigning, a mapping table between the task number and the core number is established synchronously. The information content of the task and the corresponding core in the mapping table is written into the assignment list, and the core execution task table is generated.

[0029] The specific steps of generating the synchronous priority scheduling table are: Based on the core execution task table, the task set is read according to the core number, the task planning execution time in each set is arranged in order, the difference between adjacent task time values is calculated and compared with the synchronization threshold, and the task number with a time difference less than the threshold is selected. The selected number is grouped according to the core number and added to the grouping list to generate the synchronous task grouping table. Based on the synchronous task grouping table, the core execution queue position corresponding to each group task number is obtained, the whole group task is moved to the high priority task, the original order of the tasks in the group is maintained, the adjusted queue is established with the data record table corresponding to the synchronous task group number, and the data record table is combined to generate the synchronous priority scheduling table. Based on the core execution task table, the task set is read according to the core number, the task planning time in each task set is arranged in order from early to late, and the arrangement is based on the numerical value of the time value from the minimum value to the maximum value in order. Subsequently, the time difference between the time values of the adjacent two tasks is directly subtracted, the synchronization threshold is set to 200 milliseconds, the size of the time difference and the synchronization threshold is judged, the task number with a time difference less than the threshold is selected, the selected number is grouped according to the core number, and the grouping result is recorded in the grouping list in the form of core number as key and task number array as value, to generate the synchronous task grouping table. Based on the synchronous task grouping table, the core execution queue position corresponding to each group task number is obtained, the relative position of the tasks in the same group in the queue is moved to the high priority task, and the original order of the tasks in the group is maintained. The insertion point is determined according to the next index value of the end position of the high priority task during the movement. The corresponding data record is established according to the task number and the synchronization group number of the execution queue after the insertion. The synchronization group number is taken as the index, and the adjusted queue sequence is taken as the value in the record table. The record table is finally combined into a single set in order of core number, and the synchronous priority scheduling table is generated.

[0030] Please refer to Figure 2 , the oil and gas field data acquisition and processing system based on the Hongmeng system is used to execute the above-mentioned oil and gas field data acquisition and processing method based on the Hongmeng system. The system comprises: Trend analysis module: based on the wellhead acquisition end delay record, bandwidth occupation ratio, data packet size, the cycle data is input into the long short-term memory network calculation in order to generate trend value, and the difference value comparison mark is compared with the original data. The time node is generated, and the delay warning node set is generated; Preferred node module: based on the delay warning node set, read the node delay value and bandwidth ratio, filter the nodes with high delay and large bandwidth, arrange them in descending order according to the data packet size, and record the communication link identification to form the priority link node table; Link allocation module: based on the priority link node table, input the node delay and bandwidth into the graph neural network to generate the combined value, compare with the threshold to filter the high load nodes, select the qualified path according to the delay and bandwidth, and obtain the shunt path allocation table; Core scheduling module: based on the shunt path allocation table, read the task type, put the safety monitoring task into the front end of the execution queue, count the number of each core task, select the core with the valley value of the task number for task allocation, and obtain the core execution task table; Synchronous arrangement module: based on the core execution task table, compare the adjacent time difference of each core task, group the tasks with time difference less than the threshold into a synchronous group, insert high priority tasks, and generate a synchronous priority scheduling table.

[0031] The above is only the preferred embodiment of the present application, not other forms of the present application, any skilled in the art may use the above disclosed technical content to change or modify the equivalent embodiment applied to other fields, but any simple modification, equivalent change and modification of the above embodiment without departing from the technical solution content of the present application, according to the technical essence of the present application, still belongs to the protection scope of the technical solution of the present application.

Claims

1. A method for oil and gas field data acquisition and processing based on a hyper system, characterized in that, The method comprises the following steps: Step 1: input the network delay record, bandwidth occupation ratio and data packet size sequence of the wellhead collection end into a long short-term memory network to generate trend values and calculate adjacent difference values, and judge the out-of-limit and mark the delay warning node set by synchronously comparing the original sequence; Step 2: based on the delay warning node set, read the delay value and bandwidth ratio and synchronously compare and screen, extract the nodes with large delay and high bandwidth, and reorder them in descending order of data packet size, record the communication link identifier and obtain the priority link node table; Step 3: based on the priority link node table, add the pipeline monitoring node link delay and bandwidth value in the node feature aggregation unit of the graph neural network to form a weight value, judge the high-load link and screen the shunt node, and generate a shunt path distribution table; Step 4: based on the shunt path distribution table, read the compressor station monitoring unit task type, insert the safety monitoring task into the execution queue preferentially, compare the running core queue length to select the valley value queue core to allocate tasks, and obtain a core execution task table; Step 5: based on the core execution task table, read each core task set and compare the adjacent execution time, group the tasks with a time difference less than a synchronization threshold value into a synchronization group and insert a high-priority task, and integrate and record to form a synchronization priority scheduling table.

2. The method for oil and gas field data acquisition and processing based on a Hmong system according to claim 1, characterized in that, The specific steps for generating the delay warning node set are: arrange the three types of parameters of each collection period in time sequence, subtract the same parameter values of adjacent periods item by item to obtain difference values, and store the difference values and time numbers in an ordered list according to the difference values, to generate a period difference data set; based on the period difference data set, read the difference values and corresponding three types of original parameter values in time number sequence, insert the difference values into the end of the parameter group at the same time node to form a complete parameter unit, input the parameter units formed at each time node into the input layer of the long short-term memory network in sequence and recursively calculate the output, fuse the output results and the original parameter units at the same time node, and store them in a new sequence in continuous order to generate a composite trend sequence; based on the composite trend sequence, read the delay values of each time node in sequence and compare them with a set threshold value, mark the time nodes with delay values greater than the threshold value and write them into a mark list, combine the marked nodes and parameter units to generate a target set, and generate a delay warning node set.

3. The method according to claim 1, wherein, The execution process of the long short-term memory network is as follows: after reading the difference values and corresponding three types of original parameter values in time number sequence, insert the difference values into the end of the parameter group at the same time node to form a complete parameter unit, arrange the numerical values of all parameter units in time sequence and input them into the input layer of the memory network, process the inputs of each time node in the hidden storage unit in sequence, combine the state values of the previous time nodes for recursive operation, fuse the current step operation results and the parameter units of the corresponding time nodes item by item, maintain the time sequence of the fusion results, output each time node and add it to a new sequence to form a data set for generating a composite trend sequence.

4. The method according to claim 1, wherein, The specific steps for generating the priority link node table are: Based on the delay early warning node set, read the node delay value and bandwidth occupation ratio in time sequence, compare the two to select the node with high delay value and large bandwidth ratio, match the node number with the corresponding data packet size and store it in the list to generate a high delay high bandwidth node set; Based on the high delay high bandwidth node set, sort the nodes by packet size from large to small, extract the communication link identifier of each node in the sorting result and insert it into the link list in turn, store the link list corresponding to the node number, and generate a priority link node table.

5. The method of claim 1, wherein the method is a method of oil and gas field data acquisition and processing based on a Hmong system. The specific steps for generating the shunt path allocation table are: Based on the priority link node table, read the delay value and bandwidth value of each link in the oil pipeline monitoring node, perform addition operation in time sequence in the node feature aggregation unit of the graph neural network to obtain the merged value, and write the operation result to the node corresponding data column, and store it in the set in time sequence to generate a link weight value set; Based on the link weight value set, read the weight value of each node and compare it with the set load threshold value, write the node number with weight value greater than the threshold value into the high load list, obtain the candidate link data corresponding to the node in the list, and form a table according to the corresponding relationship between the node and the candidate link information to generate a high load candidate link table; Based on the high load candidate link table, compare the delay values of the candidate links of each node to select the link with the valley value delay, and compare the bandwidth values when the delay is equal to select the link with the peak value, combine the node number and the selected link to form a pairing record, and add it to the result set in record order to generate a shunt path allocation table.

6. The method for oil and gas field data acquisition and processing based on a Hmong system according to claim 1, characterized in that, The graph neural network execution process is based on the priority link node table, the link delay value and bandwidth value of the oil pipeline monitoring node and the adjacent node are loaded as graph structure input according to the node number, the delay value and the bandwidth value of each node are combined to form a feature vector in time sequence, which is transmitted to the node feature aggregation unit of the network, and the feature vectors of the directly connected nodes are paired one by one in the unit and are added, and the addition result is written to the feature storage area of the corresponding node, and the node features are rearranged according to the time number and stored in the set as the input data source of the link weight value set.

7. The method of claim 1, wherein the method is a method of oil and gas field data acquisition and processing based on a Hmong system. The specific steps for generating the core execution task table are: Based on the shunt path allocation table, read the compressor station monitoring unit task type according to the node number, extract the safety monitoring type node and insert it into the head end of the execution queue in task number order, and append the non-safety task in the original number order at the end of the queue, while generating a corresponding record table of task number and queue position to generate a priority task queue table; Based on the priority task queue table, count the number of tasks in the current queue of each running core, identify the core with the valley value queue as the valley value queue core, distribute the tasks in the priority task queue table to the core according to the node number order, and bind the distributed task number and core number in the distribution list to generate a core execution task table.

8. The method of claim 1, wherein the method is a method of oil and gas field data acquisition and processing based on a Hmong system. The specific steps for generating the synchronous priority scheduling table are: Based on the core execution task table, the task sets are read in sequence according to the core numbers, the task execution times in each set are arranged in order, the adjacent task time values are subtracted and compared with the synchronization threshold value, the task numbers with a time difference less than the threshold value are screened out, the screened numbers are grouped according to the core numbers and added to the grouping list, and a synchronization task grouping table is generated; Based on the synchronization task grouping table, the core execution queue positions corresponding to the task numbers of each group are obtained, the tasks of each group are moved as a whole to high-priority tasks, the original order of the tasks in the group is maintained, a data record table is established corresponding to the adjusted queue and the synchronization task group numbers, and a synchronization priority scheduling table is generated by merging the data record table.

9. A device for oil and gas field data acquisition and processing based on a HOMOGENEOUS SYSTEM, comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor executes the computer program to realize the steps of the oil and gas field data acquisition and processing method based on the Hongmeng system in any one of claims 1-7.

10. A data acquisition and processing system for oil and gas fields based on a HOMOGENEOUS SYSTEM, characterized in that, The oil and gas field data acquisition and processing system based on the Hongmeng system is used to execute the oil and gas field data acquisition and processing method based on the Hongmeng system in any one of claims 1-7, and the system comprises: The trend analysis module: based on the wellhead acquisition end delay record, bandwidth occupation ratio, and data packet size, the periodic data is input into a long short-term memory network in order to calculate the trend value, and the difference value between the trend value and the original data is compared to mark the over-limit time node, and a delay warning node set is generated; The preferred node module: based on the delay warning node set, the node delay value and bandwidth ratio are read, the nodes with high delay and large bandwidth are screened out, the data packet size is arranged in descending order, and the communication link identifier is recorded to form a priority link node table; The link allocation module: based on the priority link node table, the node delay and bandwidth are input into a graph neural network to aggregate and generate a combined value, the high-load nodes are screened out by comparing the threshold value, the qualified path is selected according to the delay and bandwidth, and a shunt path allocation table is obtained; The core scheduling module: based on the shunt path allocation table, the task type is read, the safety monitoring task is placed at the front end of the execution queue, the number of tasks of each core is counted, the core with the valley value of the task number is selected for task allocation, and a core execution task table is obtained; The synchronization arrangement module: based on the core execution task table, the adjacent time difference of each core task is compared, the tasks with a time difference less than the threshold value are grouped into a synchronization group, and a high-priority task is inserted after the synchronization priority scheduling table is generated.

Images