CN116048535A

CN116048535A - Data processing method and device, nonvolatile storage medium and electronic equipment

Info

Publication number: CN116048535A
Application number: CN202211565668.1A
Authority: CN
Inventors: 王新梦; 王彦伟; 贾继生; 武志学; 王宗文; 王文龙
Original assignee: Yantai Jereh Petroleum Equipment and Technologies Co Ltd
Current assignee: Yantai Jereh Petroleum Equipment and Technologies Co Ltd
Priority date: 2022-12-07
Filing date: 2022-12-07
Publication date: 2023-05-02

Abstract

The invention discloses a data processing method and device, a nonvolatile storage medium and electronic equipment. Wherein the method comprises the following steps: determining an operation condition of the target equipment and a key number corresponding to the operation condition based on the operation data of the target equipment, wherein the target equipment at least comprises: an oil or gas exploration device, an oil or gas collection device, an oil or gas energy storage device, and an oil or gas processing device; determining a target prediction model in a plurality of preset models based on the operation conditions; determining a key value corresponding to the key number; performing parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model; and processing the running data by adopting the configured target prediction model to obtain a data processing result. The invention solves the technical problems of low data processing efficiency and resource waste caused by the non-ideal degree of deployment configuration automation of equipment data processing in the related technology.

Description

Data processing method and device, nonvolatile storage medium and electronic equipment

Technical Field

The invention relates to the field of intellectualization of oil and gas service equipment, in particular to a data processing method and device, a nonvolatile storage medium and electronic equipment.

Background

The intelligent algorithm model is widely studied and primarily applied as the core and the brain of an intelligent system in the oil and gas industry, the operation of enterprises in industrial production enterprises and the completion of industrial production tasks are required to be completed, stable and efficient production equipment is required, and the intelligent system of the equipment is also required to have the capability of efficient deployment and continuous and stable operation. The oil gas intelligent model is simple and tedious in deployment mode in the initial application stage, requires a large amount of human resources and time resources, and often faces the conditions of system breakdown and service downtime in the processing process. The related technology fails to provide efficient algorithm deployment and a data processing implementation method with continuous and stable operation, so that resource waste is caused, the adaptability to various working conditions is poor, the processing efficiency is low, and the processing capacity on the site of equipment is not ideal.

In view of the above problems, no effective solution has been proposed at present.

Disclosure of Invention

The embodiment of the invention provides a data processing method and device, a nonvolatile storage medium and electronic equipment, which at least solve the technical problems of low data processing efficiency and resource waste caused by non-ideal deployment configuration automation degree of equipment data processing in the related technology.

According to an aspect of an embodiment of the present invention, there is provided a data processing method including: determining an operation working condition of target equipment and a key number corresponding to the operation working condition based on operation data of the target equipment, wherein the target equipment at least comprises: an oil or gas exploration device, an oil or gas collection device, an oil or gas energy storage device, and an oil or gas processing device; determining a target prediction model in a plurality of preset models based on the operation conditions, wherein model types of the plurality of preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment; determining a key value corresponding to the key number from a preset model list configuration file, wherein the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model; performing parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model; and processing the operation data by adopting the configured target prediction model to obtain a data processing result corresponding to the target equipment.

According to another aspect of an embodiment of the present invention, there is provided a data processing apparatus including: the first determining module is used for determining the operation working condition of the target equipment and the key number corresponding to the operation working condition based on the operation data of the target equipment; the second determining module is used for determining a target prediction model in a plurality of preset models based on the operation condition, wherein model types of the plurality of preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment; the first configuration module is used for determining a key value corresponding to the key number from a preset model list configuration file, wherein the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model; the second configuration module is used for carrying out parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model; and the processing module is used for processing the operation data by adopting the configured target prediction model to obtain a data processing result corresponding to the target equipment.

According to another aspect of embodiments of the present invention, there is provided a non-volatile storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform any one of the data processing methods.

According to another aspect of an embodiment of the present invention, there is provided an electronic apparatus including: one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the data processing method of any of the claims.

In the embodiment of the invention, the operation working condition of the target equipment and the key number corresponding to the operation working condition are determined based on the operation data of the target equipment, wherein the target equipment at least comprises: an oil or gas exploration device, an oil or gas collection device, an oil or gas energy storage device, and an oil or gas processing device; determining a target prediction model in a plurality of preset models based on the operation conditions, wherein model types of the plurality of preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment; determining a key value corresponding to the key number from a preset model list configuration file, wherein the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model; performing parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model; and processing the operation data by adopting the configured target prediction model to obtain a data processing result corresponding to the target equipment. The method achieves the purposes of efficiently carrying out data processing in a flow manner and automatically selecting a processing algorithm model according to working conditions, achieves the technical effects of improving processing efficiency and reducing manpower and time resource investment, and further solves the technical problems of low data processing efficiency and resource waste caused by non-ideal degree of deployment and configuration of equipment data processing in the related technology.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:

FIG. 1 is a flow chart of an alternative data processing method provided in accordance with an embodiment of the present invention;

FIG. 2 is a schematic diagram of an alternative data processing method provided in accordance with an embodiment of the present invention;

FIG. 3 is a logic flow diagram of an alternative data processing method provided in accordance with an embodiment of the present invention;

fig. 4 is a schematic diagram of an alternative data processing apparatus provided in accordance with an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

The intelligent algorithm model is widely studied and primarily applied as the core and the brain of an intelligent system in the oil and gas industry, the operation of enterprises in industrial production enterprises and the completion of industrial production tasks are required to be completed, stable and efficient production equipment is required, and the intelligent system of the equipment is also required to have the capability of efficient deployment and continuous and stable operation. The oil gas intelligent model is simple and tedious in deployment mode in the initial application stage, requires a large amount of human resources and time resources, and often faces the conditions of system breakdown and service downtime in the processing process. The related technology fails to provide efficient algorithm deployment and a data processing implementation method with continuous and stable operation, so that resource waste is caused, the adaptability to various working conditions is poor, the processing efficiency is low, and the processing capacity on the site of equipment is not ideal. In the process of transformation, upgrading and intelligent development of the traditional industry, the traditional equipment needs to be effectively combined with an artificial intelligence technology method, so that practical application and scheme landing are realized. The intelligent service system needs to realize efficient deployment and continuous stable operation of industrial production, and a whole set of standard effective technical implementation method support is needed.

In accordance with an embodiment of the present invention, there is provided a method embodiment of data processing, it being noted that the steps shown in the flowchart of the figures may be performed in a computer system, such as a set of computer executable instructions, and, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order other than that shown or described herein.

FIG. 1 is a flow chart of a data processing method according to an embodiment of the present invention, as shown in FIG. 1, the method comprising the steps of:

step S102, determining an operation condition of the target device and a key number corresponding to the operation condition based on operation data of the target device, wherein the target device at least comprises: an oil or gas exploration device, an oil or gas collection device, an oil or gas energy storage device and an oil or gas processing device.

It is understood that the operation data is generated during the continuous operation of the target device, the operation condition of the target device is determined based on the operation data, and the key number corresponding to the operation condition is determined. The target equipment is oil-gas equipment and at least comprises: an oil or gas exploration device, an oil or gas collection device, an oil or gas energy storage device and an oil or gas processing device.

In an alternative embodiment, the method is applied to a web application framework established based on a Python language, and an execution program corresponding to the method is stored or executed in a web package or an execution file, wherein the web application framework at least comprises: a Django framework, a Tornado framework, or a flash framework.

It can be understood that the method provided by the invention is applied to a web application framework, wherein the web application framework is a development framework for supporting development of dynamic websites, network application programs and network services. The web application framework at least comprises: a Django framework, a Tornado framework, or a flash framework. The execution program corresponding to the method provided by the invention is stored or executed in the form of a web program package or an execution file. Through the arrangement, the operation environment and the implementation architecture of the method provided by the invention are shown.

The Django framework is a web application framework of open source code, which is obtained from Python language and adopts MTV (Model, template, view, i.e. model, template and view) design mode. The Tornado framework is a web application framework of open source codes, is obtained by the Python language, is a non-blocking server and provides real-time and rapid data processing services. The above-mentioned flash framework is called micro framework (micro) and "micro" does not mean that the whole web application is put into a Python file, and "micro" in the micro framework means that the flash aims to keep the code compact and easy to expand, and the main feature of the flash framework is that the core structure is simpler, but has very strong expansibility and compatibility.

Optionally, the web application framework further includes: the pyramid framework is a small, fast Python language-based framework.

Step S104, determining a target prediction model in a plurality of preset models based on the operation conditions, wherein the model types of the plurality of preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment.

It can be understood that the better processing effect can be achieved by adopting different models to process according to different working conditions, so that the target prediction model is determined in a plurality of preset models based on the operation working conditions. The multiple preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment. Through the processing, the processing model is determined according to the working conditions, so that the adaptability of the data processing result to the specific application environment is improved, and the data processing result with high accuracy is obtained.

In an alternative embodiment, the method further includes, before determining the target prediction model from the plurality of preset models based on the operating condition, the method further includes: acquiring a history working condition corresponding to a first device and various working condition data corresponding to the history working condition, wherein the first device is the target device and/or other devices with the same type as the target device, and the history working condition is a working condition of the first device in a preset running time range; and respectively training a plurality of preset initial models by adopting the plurality of working condition data to obtain a plurality of preset models.

It can be appreciated that the plurality of preset models may be considered as being obtained through pre-training, and first, a history condition corresponding to the first device is obtained, where the first device is the target device and/or other devices with the same device type as the target device, and the history condition is a condition that the first device appears in a preset operation time range. The data volume of the training set is expanded by acquiring various working condition data of the same type of equipment with working conditions, so that the training effect is improved. And respectively training a plurality of preset initial models by adopting a plurality of working condition data to obtain a plurality of preset models. Through the processing, the processing capacity of various preset models is improved, and the pre-training of the various preset models is completed.

In an alternative embodiment, the determining the target prediction model from a plurality of preset models based on the operation conditions includes: acquiring a history working condition corresponding to a first device and various working condition data corresponding to the history working condition, wherein the first device is the target device and/or other devices with the same type as the target device, and the history working condition is a working condition of the first device in a preset running time range; determining target working condition data corresponding to the operation working condition from the plurality of working condition data; determining a first model obtained by training the target working condition data in the multiple preset models; and taking the first model as the target prediction model.

It is understood that various operating condition data are obtained, and target operating condition data corresponding to the operating condition are determined from the various operating condition data. In order to obtain a first model with good training and strong processing capacity, a first model which is obtained by training by adopting target working condition data is determined in a plurality of preset models, and the first model is used as a target prediction model.

In an alternative embodiment, in a case where the history of operating conditions is plural, the determining the target operating condition data from the plural operating condition data based on the operating conditions includes: judging whether a first historical working condition which is the same as the operation working condition exists in the plurality of historical working conditions or not; if the first history working condition exists in the plurality of history working conditions, determining first working condition data corresponding to the first history working condition; and taking the first working condition data as the target working condition data.

It can be appreciated that in an actual application scenario, there are multiple history working conditions, where the multiple history working conditions may include or not include an operation working condition, different treatments are required, and under a condition that there is a first history working condition identical to the operation working condition in the multiple history working conditions, first working condition data corresponding to the first history working condition is determined, and the first working condition data is used as target working condition data.

In an alternative embodiment, the method further comprises: if the first historical working condition does not exist in the plurality of historical working conditions, determining the similarity between the plurality of historical working conditions and the operation working conditions respectively; determining a second history working condition with similarity larger than a preset similarity threshold value in the plurality of history working conditions; and taking the second working condition data corresponding to the second history working condition as the target working condition data.

It will be appreciated that in the event that there is no first history of the plurality of historic conditions that is the same as the operating condition, the operating condition is considered to be different from each of the plurality of historic conditions, and that in order to complete the process as much as possible in the existing plurality of preset models, a second, most similar, history is determined. Therefore, the similarity between the plurality of historical working conditions and the operation working conditions is determined, a second historical working condition, in which the similarity is larger than a preset similarity threshold, in the plurality of historical working conditions is determined, and second working condition data corresponding to the second historical working condition are used as the target working condition data.

Step S106, determining a key value corresponding to the key number from a preset model list configuration file, wherein the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model.

It can be understood that, since the configuration of the model parameters is a very complicated process, in order to improve the data processing efficiency, the key value corresponding to the key number is determined from the configuration file of the preset model list, and the initial parameters or the fixed parameters corresponding to the target prediction model are obtained by using the matching relationship between the key number and the key value.

And S108, carrying out parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model.

It can be understood that the obtained key value is adopted to carry out parameter configuration on the target prediction model, so as to obtain the configured target prediction model.

And step S110, processing the operation data by adopting the configured target prediction model to obtain a data processing result corresponding to the target equipment.

It can be understood that after two steps of model selection and parameter configuration, the configured target prediction model is regarded as a model which is the best match with the operation data in the multiple preset models, and a data processing result corresponding to the target device is obtained.

In an alternative embodiment, the method further comprises: based on the process of determining the target prediction model in the multiple preset models, obtaining first process information; performing parameter configuration on the target prediction model based on the key value to obtain second process information; a process of processing the operation data based on the configured target prediction model to obtain third process information; based on the data processing result, the first process information, the second process information and the third process information generate a log file, and the log file is stored in a preset storage position, wherein the log file is a local storage file or a preset database storage file.

It can be understood that, in order to further improve the data processing capability, the processing procedure is recorded, and the parameter configuration procedure is performed on the target prediction model based on the key value, so as to obtain second procedure information; a process of processing the operation data based on the configured target prediction model, so as to obtain third process information; based on the data processing result, the first process information, the second process information and the third process information generate a log file, the log file is stored in a preset storage position, and the log file is a local storage file or a preset database storage file.

Through the steps, the data processing can be efficiently performed in a flow manner, the purpose of automatically selecting a processing algorithm model according to working conditions is achieved, the processing efficiency is improved, the technical effects of reducing manpower and time resource investment are achieved, and the technical problems of low data processing efficiency and resource waste caused by non-ideal deployment configuration automation degree of equipment data processing in the related technology are solved.

Based on the above embodiment and the optional embodiment, the invention provides an optional implementation mode, and mainly provides a method for realizing output processing and flow processing based on an oil gas intelligent model. The implementation of the method mainly comprises two large application function modules, namely an algorithm model service application module and a model result storage module. FIG. 2 is a schematic diagram of an alternative data processing method according to an embodiment of the present invention, where, as shown in FIG. 2, an algorithm model service application module includes functional items: the method comprises the steps of model calling service interfaces, intelligent model function main programs of oil and gas equipment, multi-model configuration calling of different measuring points of the equipment, model parameter configuration and automatic calling and a model running log recorder. The model result storage module comprises the following functions: and storing the model reasoning results in a multi-mode and storing the model log files in a dual-mode. The above modules are software function modules, and need to build an environment based on the Python language, and import the Python software package and the dependency package related to all the above functions locally.

Under the condition that the application scene is an oil-gas equipment fault prediction intelligent scene, the scene mainly comprises the steps of deploying a plurality of fault prediction algorithm models for different vulnerable parts of various equipment to realize the function of predicting faults of the vulnerable parts in advance, and fig. 3 is a logic flow chart of an optional data processing method provided according to an embodiment of the invention, as shown in fig. 3, the specific steps are realized as follows:

step one: the oil and gas equipment fault prediction algorithm model calls a service interface to be created and started. To obtain the target equipment operational data, the hydrocarbon equipment failure prediction model invokes a service interface including, but not limited to, providing equipment failure prediction service applications in two ways. One way is for a web application service interface to be through a web-based service framework that includes, but is not limited to: django, pyramid, tornado, or flash, etc. The method comprises the steps of packaging and packaging a main body program of a fault prediction model function of oil and gas equipment, a multi-model configuration call, model parameter configuration and automatic call and a model operation log recorder into a web service program package for program execution, and transmitting input parameters (namely operation data of target equipment) required by an algorithm model into the web service program package through an address and a port number appointed by a web service interface, so that a calculated value returned by algorithm model processing can be obtained from the web service interface.

The other is to package the multi-model configuration call, the model running log recorder, the oil and gas equipment fault prediction model function main program, the model parameter configuration and the automatic call into an executable file (suffix is. Exe), and the package modes of the executable file are various, including but not limited to using py2exe, pyinstall, cx_Freeze, nuitka method and the like. The input parameters (i.e., the operational data of the target device) may be directly imported into the executable file and the acquisition of the calculated values returned via the algorithmic model process performed.

The py2exe is a tool for converting the Python script into an independently executable program (extension. Exe) on windows system. The above-described pyinstler uses operating system support to load dynamic libraries, thereby ensuring compatibility with a variety of operating platforms. The cx_free is used to package the Python script into an executable program, like py2exe. In contrast, cx_Freeze is cross-platform, requiring Python2.3 or updated versions. The nuitka is a substitution compiler of Python, and can seamlessly substitute and expand interpretation and compiling work of Python.

In an alternative manner, the service interface is invoked by the device fault prediction algorithm model in a manual opening manner, and in a case where the scheme of providing data processing by the web service interface is executed in a command prompt, for convenience of understanding, for example: inputting a D/PythonD/main.py command, and starting a model service, wherein main.py is an oil and gas equipment fault prediction model web service interface starting program, and D/is an absolute path where Python.exe and main.py are respectively.

In another case, namely, in the case of a scheme for providing data processing in an executable file manner, for convenience of understanding, specific examples are: writing a command of D \\PythonD \main.py into a txt text format, and transferring the command into a file with an extension of ". Dat" (a file suffix) or ". Exe", and starting the file by double clicking the file of ". Dat" or ". Exe". Or the oil and gas equipment fault prediction model program package is packaged by an executable file (suffix is. Exe), and the service can be started by directly double-clicking the executable file.

In an alternative way, the device fault prediction algorithm model calls a service interface automatic opening mode, and the automatic opening mode is divided into two types, wherein one type is to configure a model service exe file into software capable of realizing automatic restarting crashing or suspending, including but not limited to a software program for monitoring whether a program runs normally or not, and the like, so as to realize automatic restarting or automatic starting with device starting after service crashing or suspending.

In an alternative manner, the deployment environment of the windows system (an operating system developed by microsoft) can be set as a system task planning program by setting an extension name ". Exe" file, and the system is self-started when being started.

Step two: according to the requirements of the fault prediction algorithm model, the data to be tested of the equipment are converted into model input data and are input into each algorithm model service of an algorithm model service application module together with the model dynamic parameters.

In the equipment failure prediction scene, the equipment to-be-detected data comprises, but is not limited to, the original data of vibration or other mechanical movement indexes or thermodynamic indexes are acquired and acquired by installing vibration or other types of sensors at the vulnerable part of the equipment. The model input data is a model characteristic index system data set constructed by means of characteristic index calculation, time-frequency domain data conversion and the like based on the equipment to-be-detected data. The model input data and the model dynamic parameters can be transmitted in the data formats of list, JSON character string, a dictionary of the subject and the like. The JSON (JavaScriptObjectNotation, JS object profile) is a lightweight data exchange format.

Step three: the fault prediction model service program firstly performs 'multi-model configuration calling', and performs proper model screening and calling according to the multi-model calling configuration rules according to the production working conditions or other production conditions of the input equipment.

After the input data and dynamic parameters of the oil and gas equipment fault prediction model are input into the fault prediction model service, the multi-model configuration call of the same component or equipment type is firstly carried out. The multi-model configuration call is mainly based on the equipment production conditions or other production conditions contained in the input data. For the purpose of facilitating understanding and carrying out specific examples, for equipment fault prediction scenes, the rotating speed or pressure in the running process of equipment is used as a main working condition influence factor, and the factors directly influence equipment fault prediction model training and influence the characteristic representation of data such as original vibration, power and the like of the equipment used in the testing process. Therefore, in this example, the collected data obtained by the same type of equipment (or the same type of collected part) needs to be classified according to two working condition influencing factors, such as the rotation speed or the pressure, for example, the collected data is divided into different working conditions according to multiple typical rotation speed or pressure working conditions, and the historical working conditions and multiple working condition data corresponding to the historical working conditions are obtained according to different working conditions. And training various initial models by adopting various working condition data. In other words, a plurality of preset models, i.e. failure prediction models, are obtained for the same vulnerable part of the same device.

Different models are adopted for processing corresponding to different working conditions, so that better processing effects can be achieved, and the target prediction model is determined in a plurality of preset models based on the operation working conditions. The multiple preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment. And the processing model is determined according to the working condition, so that the adaptability of the data processing result to the specific application environment is improved, and the data processing result with high accuracy is obtained. The multiple preset models can be regarded as being obtained through pre-training, and the historical working conditions corresponding to the first device are firstly obtained, wherein the first device is the target device and/or other devices with the same type as the target device, and the historical working conditions are working conditions of the first device in a preset running time range. The data volume of the training set is expanded by acquiring various working condition data of the same type of equipment with working conditions, so that the training effect is improved. And respectively training a plurality of preset initial models by adopting a plurality of working condition data to obtain a plurality of preset models. Through the processing, the processing capacity of various preset models is improved, and the pre-training of the various preset models is completed.

And the multi-model configuration call is used for selecting a model adapting to the working condition of the data to be tested, generating a plurality of different fault prediction models through training aiming at a plurality of different working conditions or other production conditions, and performing model call according to the following multi-model call rules. The model calling mode is specifically described as follows:

and acquiring various working condition data, and determining target working condition data corresponding to the operation working condition in the various working condition data. In order to obtain a first model with good training and strong processing capacity, a first model which is obtained by training by adopting target working condition data is determined in a plurality of preset models, and the first model is used as a target prediction model. In an actual application scene, a plurality of historical working conditions exist, the plurality of historical working conditions possibly comprise or do not comprise operation working conditions and need to be processed differently, first working condition data corresponding to the first historical working conditions are determined under the condition that the first historical working conditions identical to the operation working conditions exist in the plurality of historical working conditions, and the first working condition data are used as target working condition data. In the case where there is no first history condition that is the same as the operation condition among the plurality of history conditions, it is regarded that the operation condition is different from each of the plurality of history conditions, and in order to complete the processing as much as possible in the existing plurality of preset models, it is necessary to determine the most similar second history condition. Therefore, the similarity between the plurality of historical working conditions and the operation working conditions is determined, a second historical working condition, in which the similarity is larger than a preset similarity threshold, in the plurality of historical working conditions is determined, and second working condition data corresponding to the second historical working condition are used as the target working condition data.

Step four: according to the data to be tested and the dynamic parameters of the model, the functions of model parameter configuration and automatic call are adopted to perform model initial or fixed parameter configuration.

The device fault prediction model parameter configuration and automatic calling function realizes model initial or fixed parameter calling, mainly uses key numbers to make key values for initial or fixed parameters related to an algorithm model, and when the model is called, the initial or fixed parameters corresponding to a configuration file dictionary are matched by inputting the key numbers so as to realize model initial or fixed parameter calling input.

After the data to be tested and the dynamic parameters of the model are input, in the above example, the multi-model configuration calling function performs matching model screening according to the target rotation speed or pressure working condition, and the rotation speed or pressure working condition is input into the preset model list configuration file as a key number, and the corresponding key value is matched through the key value relationship, namely, the initial parameters or the fixed parameters of the fault prediction model under the working condition, including but not limited to: the method comprises the steps of data acquisition frequency, sample data length, database account information of a database table, equipment structure parameters and the like. The matched key values are simultaneously input into a fault prediction model corresponding to the target working condition to be detected.

Step five: the equipment fault prediction model function main body program calculates model prediction or identification results or label data according to model input data, model dynamic parameters and model initial or fixed parameters.

The main program of the equipment fault prediction model function is mainly an algorithm model main program constructed by Python language programming and comprises the whole structure and the calculation logic process of an algorithm model, wherein the model function can comprise but is not limited to: a mechanism model function body of the rule threshold class, for example: 1. and setting an overrun threshold rule according to evaluation indexes such as mechanical motion indexes, thermodynamic performance indexes and the like of the equipment, and constructing a rule model. 2. Neural network model, the model main body carries out network structure construction based on Python neural network frame, and the related frames include but are not limited to: tensorflow, pytorch, keras, etc.

The tensorf low is a symbol mathematical system based on data stream programming, is widely applied to programming realization of various machine learning algorithms, and the precursor of the tensorf low is a google neural network algorithm library. The system has a multi-level structure, can be deployed on various servers, PC terminals and webpages, and supports high-performance numerical computation. The pyrach is an open source Python machine learning library, and is used for application programs such as natural language processing. Support dynamic graphs and provide a Python interface. The above-mentioned keras is an open source artificial neural network library written by Python, and is used for designing, debugging, evaluating, applying and visualizing deep learning model.

The main program for realizing the model function mainly comprises the following steps: a neural network structure framework program, an index system construction program, a to-be-measured data prediction error calculation program, a to-be-measured data classification label output program and the like. After the model input data, the model dynamic parameters and the model initial or fixed parameters are all input into the target prediction model, the input data can be processed through model calculation to obtain a prediction discrimination error or a prediction classification label and the like to obtain a data processing result corresponding to the target equipment.

Step six: after the model result storage module receives the data processing result output by the algorithm model service application module, the model prediction result storage under the target mode is carried out according to the multi-mode storage mode.

The multi-mode storage of the reasoning result of the fault prediction model mainly comprises three storage modes of the reasoning result of the model, namely: reasoning result local file storage (including but not limited to text files, spreadsheets, etc.), database library table storage (including but not limited to mysql, oracle, etc.), or generating output result JSON strings. The mysql is used as a relational database management system and has the characteristics of small volume, high speed, low total possession cost, open source code and the like. The oracle is an information management software, which also serves the database.

Step seven: the model running log recorder synchronously records the running whole process log and error reporting record in the models from the step one to the step six.

The equipment fault prediction model operates the log recorder, and the log recorder, the log processor and the log formatter are built, the log formatter is added into the log processor, the log processor is added into the log recorder, and the model function main body program, the multi-model configuration calling program, the model parameter configuration program and the automatic calling program are used for carrying out operation conditions and fault reporting record. When the log files are stored locally, setting a single log file size limit and a total log file number limit, and performing fixed number and fixed size cyclic coverage on the log files.

The model log file dual mode storage mainly comprises a model log file local storage (including but not limited to text files, spreadsheets and the like), a database table storage (including but not limited to mysql, oracle and the like).

At least any one of the following effects is achieved by the above-described alternative embodiments: the intelligent algorithm model of the two kinds of oil gas equipment of the web service interface and the executable file is realized to process data, and a plurality of web service frames are supported to build the service interface. And selecting and directionally calling the multi-type target algorithm model configuration according to the screening conditions such as working conditions. And realizing the multi-mode storage of the operation result of the algorithm model and the dual-mode storage of the log file. The automatic configuration and automatic calling of the initial or fixed parameters of the model are realized, the data processing efficiency is improved, and an automatic and procedural intelligent data processing method for the oil and gas equipment is provided.

It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.

In this embodiment, a data processing device is further provided, and the device is used to implement the foregoing embodiments and preferred embodiments, and will not be described in detail. As used below, the terms "module," "apparatus" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

According to an embodiment of the present invention, there is further provided an embodiment of an apparatus for implementing a data processing method, and fig. 4 is a schematic diagram of a data processing apparatus according to an embodiment of the present invention, as shown in fig. 4, where the data processing apparatus includes: the first determining module 402, the second determining module 404, the first configuring module 406, the second configuring module 408, and the processing module 410 are described below.

A first determining module 402, configured to determine an operation condition of a target device and a key number corresponding to the operation condition based on operation data of the target device;

the second determining module 404 is connected to the first determining module 402, and is configured to determine a target prediction model from a plurality of preset models based on the operation conditions, where model types of the plurality of preset models at least include a neural network model, and a mechanism model obtained based on an operation rule of the target device;

a first configuration module 406, connected to the second determination module 404, configured to determine, from a preset model list configuration file, a key value corresponding to the key number, where the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model;

the second configuration module 408 is connected to the first configuration module 406, and is configured to perform parameter configuration on the target prediction model based on the key value, so as to obtain a configured target prediction model;

and the processing module 410 is connected to the second configuration module 408, and is configured to process the operation data by using the configured target prediction model, so as to obtain a data processing result corresponding to the target device.

In the data processing apparatus provided by the embodiment of the present invention, the first determining module 402 is configured to determine, based on operation data of a target device, an operation condition of the target device and a key number corresponding to the operation condition; the second determining module 404 is connected to the first determining module 402, and is configured to determine a target prediction model from a plurality of preset models based on the operation conditions, where model types of the plurality of preset models at least include a neural network model, and a mechanism model obtained based on an operation rule of the target device; a first configuration module 406, connected to the second determination module 404, configured to determine, from a preset model list configuration file, a key value corresponding to the key number, where the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model; the second configuration module 408 is connected to the first configuration module 406, and is configured to perform parameter configuration on the target prediction model based on the key value, so as to obtain a configured target prediction model; and the processing module 410 is connected to the second configuration module 408, and is configured to process the operation data by using the configured target prediction model, so as to obtain a data processing result corresponding to the target device. The method achieves the purposes of efficiently carrying out data processing in a flow manner and automatically selecting a processing algorithm model according to working conditions, achieves the technical effects of improving processing efficiency and reducing manpower and time resource investment, and further solves the technical problems of low data processing efficiency and resource waste caused by non-ideal degree of deployment and configuration of equipment data processing in the related technology.

It should be noted that each of the above modules may be implemented by software or hardware, for example, in the latter case, it may be implemented by: the above modules may be located in the same processor; alternatively, the various modules described above may be located in different processors in any combination.

Here, the first determining module 402, the second determining module 404, the first configuring module 406, and the second configuring module 408 correspond to steps S102 to S110 in the embodiment, and the above modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the disclosure of the above embodiments. It should be noted that the above modules may be run in a computer terminal as part of the apparatus.

It should be noted that, the optional or preferred implementation manner of this embodiment may be referred to the related description in the embodiment, and will not be repeated herein.

The data processing apparatus may further include a processor and a memory, where the first determining module 402, the second determining module 404, the first configuring module 406, the second configuring module 408, the processing module 410, and the like are stored as program units, and the processor executes the program units stored in the memory to implement corresponding functions.

The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel may be provided with one or more. The memory may include volatile memory, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flashRAM), among other forms in computer readable media, the memory including at least one memory chip.

The embodiment of the invention provides a nonvolatile storage medium, on which a program is stored, which when executed by a processor, implements a data processing method.

The embodiment of the invention provides an electronic device, which comprises a processor, a memory and a program stored on the memory and capable of running on the processor, wherein the following steps are realized when the processor executes the program: determining an operation condition of the target device and a key number corresponding to the operation condition based on operation data of the target device, wherein the target device at least comprises: an oil or gas exploration device, an oil or gas collection device, an oil or gas energy storage device, and an oil or gas processing device; determining a target prediction model in a plurality of preset models based on the operation conditions, wherein model types of the plurality of preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment; determining a key value corresponding to the key number from a preset model list configuration file, wherein the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model; performing parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model; and processing the operation data by adopting the configured target prediction model to obtain a data processing result corresponding to the target equipment. The device herein may be a server, a PC, etc.

The invention also provides a computer program product adapted to perform, when executed on a data processing device, a program initialized with the method steps of: determining an operation condition of the target device and a key number corresponding to the operation condition based on operation data of the target device, wherein the target device at least comprises: an oil or gas exploration device, an oil or gas collection device, an oil or gas energy storage device, and an oil or gas processing device; determining a target prediction model in a plurality of preset models based on the operation conditions, wherein model types of the plurality of preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment; determining a key value corresponding to the key number from a preset model list configuration file, wherein the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model; performing parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model; and processing the operation data by adopting the configured target prediction model to obtain a data processing result corresponding to the target equipment.

It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.

The foregoing is merely exemplary of the present invention and is not intended to limit the present invention. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are to be included in the scope of the claims of the present invention.

Claims

1. A method of data processing, comprising:

determining an operation working condition of target equipment and a key number corresponding to the operation working condition based on operation data of the target equipment, wherein the target equipment at least comprises: an oil or gas exploration device, an oil or gas collection device, an oil or gas energy storage device, and an oil or gas processing device;

Determining a target prediction model in a plurality of preset models based on the operation conditions, wherein model types of the plurality of preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment;

determining a key value corresponding to the key number from a preset model list configuration file, wherein the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model;

performing parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model;

and processing the operation data by adopting the configured target prediction model to obtain a data processing result corresponding to the target equipment.

2. The method according to claim 1, wherein the method further comprises:

obtaining first process information based on a process of determining the target prediction model in the plurality of preset models;

performing parameter configuration on the target prediction model based on the key value to obtain second process information;

a process of processing the operation data based on the configured target prediction model, so as to obtain third process information;

Based on the data processing result, the first process information, the second process information and the third process information generate a log file, and the log file is stored in a preset storage position, wherein the log file is a local storage file or a preset database storage file.

3. The method of claim 1, wherein prior to said determining a target predictive model among said plurality of preset models based on said operating conditions, said method further comprises:

acquiring a history working condition corresponding to first equipment and various working condition data corresponding to the history working condition, wherein the first equipment is the target equipment and/or other equipment with the same equipment type as the target equipment, and the history working condition is a working condition of the first equipment in a preset running time range;

and respectively training a plurality of preset initial models by adopting the plurality of working condition data to obtain the plurality of preset models.

4. The method of claim 1, wherein determining a target prediction model among a plurality of preset models preset based on the operating conditions comprises:

Determining target working condition data corresponding to the operation working condition from the plurality of working condition data;

determining a first model obtained by training the target working condition data in the multiple preset models;

and taking the first model as the target prediction model.

5. The method of claim 4, wherein, in the event that the historical operating condition is a plurality of, the determining target operating condition data from the plurality of operating condition data based on the operating condition comprises:

judging whether a first historical working condition which is the same as the running working condition exists in the plurality of historical working conditions or not;

if the first history working condition exists in the plurality of history working conditions, determining first working condition data corresponding to the first history working condition;

and taking the first working condition data as the target working condition data.

6. The method of claim 5, wherein the method further comprises:

if the first historical working condition does not exist in the plurality of historical working conditions, determining the similarity between the plurality of historical working conditions and the operation working conditions respectively;

determining a second historical working condition with similarity larger than a preset similarity threshold value in the plurality of historical working conditions;

And taking the second working condition data corresponding to the second history working condition as the target working condition data.

7. The method according to any one of claims 1 to 6, wherein the method is applied to a web application framework established based on Python language, and an execution program corresponding to the method is stored or executed in a form of a web package or an execution file, and wherein the web application framework at least comprises: a Django framework, a Tornado framework, or a flash framework.

8. A data processing apparatus, comprising:

the first determining module is used for determining the operation working condition of the target equipment and the key number corresponding to the operation working condition based on the operation data of the target equipment;

the second determining module is used for determining a target prediction model in a plurality of preset models based on the operation condition, wherein model types of the plurality of preset models at least comprise a neural network model and a mechanism model obtained based on the operation rule of the target equipment;

the first configuration module is used for determining a key value corresponding to the key number from a preset model list configuration file, wherein the key value corresponds to an initial parameter or a fixed parameter corresponding to the target prediction model;

The second configuration module is used for carrying out parameter configuration on the target prediction model based on the key value to obtain a configured target prediction model;

and the processing module is used for processing the operation data by adopting the configured target prediction model to obtain a data processing result corresponding to the target equipment.

9. A non-volatile storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the data processing method of any one of claims 1 to 7.

10. An electronic device, comprising: one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the data processing method of any of claims 1 to 7.