CN115859693A

CN115859693A - Data processing method and device

Info

Publication number: CN115859693A
Application number: CN202310156976.7A
Authority: CN
Inventors: 赵亮; 李伟健; 张梦源; 黄国凌; 印卧涛
Original assignee: Alibaba Damo Institute Hangzhou Technology Co Ltd
Current assignee: Alibaba Damo Institute Hangzhou Technology Co Ltd
Priority date: 2023-02-17
Filing date: 2023-02-17
Publication date: 2023-03-28
Anticipated expiration: 2043-02-17
Also published as: CN115859693B

Abstract

The embodiment of the specification provides a data processing method and a data processing device, wherein the data processing method comprises the following steps: receiving a parameter adjusting instruction, wherein the parameter adjusting instruction carries parameter adjusting configuration information; acquiring at least one service model to be processed and at least one service data to be processed; generating a plurality of business model instances according to one business model to be processed and at least one business data to be processed based on a preset modeling language; and determining an initial solver, and adjusting solving parameters of the initial solver in a plurality of service model instances according to a preset parameter adjusting algorithm and parameter adjusting configuration information to obtain a target solver. The parameter adjusting and modeling language is combined, a plurality of business model examples are generated by utilizing the characteristic of model and data separation in the modeling language, the parameter adjusting is carried out on the solver on the plurality of model examples, the solving parameters do not need to be set by a user, the use threshold of the user is reduced, and the use experience of the user is improved.

Description

Data processing method and device

Technical Field

The embodiment of the specification relates to the technical field of computers, in particular to a data processing method.

Background

The modeling optimization problem exists in a plurality of business scenes, the optimization problem is widely applied to the fields of energy power, electronic commerce, supply chains, cloud computing, chemical engineering, finance, education and scientific research and the like, and the solving of the actual optimization problem usually comprises two steps of optimization modeling and solving, wherein an algebraic modeling language is a main mode for assisting a user in modeling. The optimization solver is the main numerical software for solving the optimization problem.

In the modeling language, a user inputs a complex optimization model through the semantics of an algebraic expression form, then introduces data, establishes the optimization model, and finally calls optimization solvers to solve. However, for a user of a modeling language, the user often cannot know the characteristics of each optimization problem and the solver, so that a proper solver and parameters are set to solve, and particularly when a plurality of optimization models or a problem (multi-instance problem) that the same model has a plurality of groups of different data exists, the problem is not well processed at present, so that how to help the user to refer to one group of optimization models in the modeling language and selecting proper solving parameters is an important problem related to the use of the modeling language.

Disclosure of Invention

In view of this, the embodiments of the present specification provide a data processing method. One or more embodiments of the present specification relate to a data processing apparatus, a computing device, a computer-readable storage medium, and a computer program, so as to solve the technical problems of the prior art.

According to a first aspect of the embodiments of the present specification, there is provided a data processing method applied to a cloud-side device, including:

receiving a parameter adjusting instruction, wherein the parameter adjusting instruction carries parameter adjusting configuration information;

acquiring at least one service model to be processed and at least one service data to be processed;

generating a plurality of business model instances according to the at least one business model to be processed and the at least one business data to be processed based on a preset modeling language;

determining an initial solver, and adjusting solving parameters of the initial solver in the plurality of service model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information to obtain a target solver.

According to a second aspect of embodiments of the present specification, there is provided a data processing method applied to a cloud-side device, including:

receiving a parameter adjusting instruction sent by a user at an end-side device for a power dispatching service, wherein the parameter adjusting instruction carries parameter adjusting configuration information;

acquiring a power cost model and at least one piece of power cost data corresponding to the power dispatching service;

generating a plurality of power cost model instances from the power cost model and the at least one power cost data based on a preset modeling language;

determining an initial solver, and adjusting solving parameters of the initial solver in the multiple power cost model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information to obtain a target solver.

According to a third aspect of embodiments herein, there is provided a data processing system comprising:

the cloud side equipment is used for sending a parameter adjusting instruction to the cloud side equipment, wherein the parameter adjusting instruction carries parameter adjusting configuration information;

the cloud-side device is configured to receive the parameter adjustment instruction, obtain at least one to-be-processed service model and at least one to-be-processed service data, generate a plurality of service model instances according to the at least one to-be-processed service model and the at least one to-be-processed service data based on a preset modeling language, determine an initial solver, adjust solution parameters of the initial solver in the plurality of service model instances according to a preset parameter adjustment algorithm and the parameter adjustment configuration information, obtain a target solver, and send the target solver to the end-side device.

According to a fourth aspect of the embodiments of the present specification, there is provided a data processing apparatus applied to a cloud-side device, including:

the device comprises a receiving module, a parameter adjusting module and a parameter adjusting module, wherein the receiving module is configured to receive a parameter adjusting instruction, and the parameter adjusting instruction carries parameter adjusting configuration information;

the acquisition module is configured to acquire at least one to-be-processed service model and at least one to-be-processed service data;

the generating module is configured to generate a plurality of business model instances according to the at least one business model to be processed and the at least one business data to be processed based on a preset modeling language;

and the adjusting module is configured to determine an initial solver, adjust solving parameters of the initial solver in the plurality of service model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information, and obtain a target solver.

According to a fifth aspect of embodiments herein, there is provided a computing device comprising:

a memory and a processor;

the memory is used for storing computer-executable instructions, and the processor is used for executing the computer-executable instructions, and the computer-executable instructions realize the steps of the data processing method when being executed by the processor.

According to a sixth aspect of embodiments herein, there is provided a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the data processing method described above.

According to a seventh aspect of embodiments herein, there is provided a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned data processing method.

One embodiment of the present description implements receiving a parameter adjustment instruction, where the parameter adjustment instruction carries parameter adjustment configuration information; acquiring at least one service model to be processed and at least one service data to be processed; generating a plurality of business model instances according to the at least one business model to be processed and the at least one business data to be processed based on a preset modeling language; determining an initial solver, and adjusting solving parameters of the initial solver in the plurality of service model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information to obtain a target solver.

According to the method provided by the embodiment of the specification, the parameter adjusting method is combined with the modeling language, the characteristic that the model and the data in the modeling language are separated is fully utilized, the at least one service model to be processed and the at least one service data to be processed are generated into a plurality of service model examples, the preset parameter adjusting algorithm is called to adjust the parameters of the solver on the plurality of model examples, the final target solver is obtained, the overall time of parameter adjustment is shortened, the solving parameters are not required to be set by a user, the use threshold of the user is lowered, and the use experience of the user is improved.

Drawings

FIG. 1 is a diagram illustrating a data processing method according to an embodiment of the present disclosure;

FIG. 2 is a flow chart of a data processing method provided by an embodiment of the present specification;

FIG. 3 is a flowchart illustrating a processing procedure of a data processing method applied to a scenario of unit combination prediction according to an embodiment of the present disclosure;

FIG. 4 is a flowchart illustrating a processing procedure of a data processing method applied to a multi-service scenario according to an embodiment of the present disclosure;

FIG. 5 is a flow chart of another data processing method provided by an embodiment of the present description;

FIG. 6 is a data processing system provided by one embodiment of the present description;

fig. 7 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present specification;

fig. 8 is a block diagram of a computing device according to an embodiment of the present disclosure.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present description. This description may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, as those skilled in the art will be able to make and use the present disclosure without departing from the spirit and scope of the present disclosure.

The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It will be understood that, although the terms first, second, etc. may be used herein in one or more embodiments to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first can also be referred to as a second and, similarly, a second can also be referred to as a first without departing from the scope of one or more embodiments of the present description. The word "if" as used herein may be interpreted as "at" \8230; "or" when 8230; \8230; "or" in response to a determination ", depending on the context.

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) referred to in this specification are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the relevant data need to comply with relevant laws and regulations and standards of relevant countries and regions, and are provided with corresponding operation entrances for the user to choose to authorize or reject.

First, the noun terms to which one or more embodiments of the present specification relate are explained.

Algebraic modeling language (algebra modelling language, AML): the complex optimization model is modeled through the semantics of an algebraic expression form, so that the use of an API (application programming interface) is avoided, and the aim is that a user can concentrate on the algebraic model. The modeling language does not solve the actual problem, but depends on connecting each optimization solver, so that a user only needs to input an algebraic model and introduce data on the development optimization model, and then the solver can be called to solve and further analyze, so that the development threshold is reduced, the development process is accelerated, and the modeling language is a programming language for the data planning problem.

Optimization solver (optimizationsolver): the main numerical software for solving the optimization problem includes many open-source and commercial solvers.

Modeling platform (modelingplatform): an integrated solution that integrates algebraic modeling language, optimization solver, and Integrated Drive Electronics (IDE) or a platform on the cloud. A complete set of modeling platforms typically includes an underlying algebraic modeling language Software Development Kit (SDK), an optimization solver SDK, and an overlying user interface IDE. With the increasing trend of cloud on infrastructure and data and the increasing complexity of optimization problems in actual business, the demand of using a modeling tool in a cloud end is increased, and the cloud modeling mode has obvious advantages in the aspects of multi-person cooperation, data import, application production and the like.

Parameter adjuster (tuningtolrtuner): each solver can exert better performance by setting solving parameters, for example, for a mixed integer linear programming problem, its cuts selection strategy, probingstrategy, nodeheuristic, etc. can greatly affect the performance of solving the problem, and the process of selecting suitable parameters is also called parameter adjustment. Many solvers provide some parameter tuning capability.

The optimization problem is widely applied to many fields such as energy and power, electronic commerce, supply chains, cloud computing, chemical engineering, finance, education and scientific research, and the like, and the method for solving the actual optimization problem usually comprises two steps of optimization modeling and solution. In the modeling language, a user inputs a complex optimization model into the model through the semantics of an algebraic expression form, then introduces data, establishes an optimization model, and finally calls an optimization solver to solve. The advantage of a generic modeling language is that it can support separate input forms for models and data, and the number of solvers supported is very high. Each solver can exert better performance by setting solving parameters, for example, for a mixed integer linear programming problem, its cuts selection strategy, probingstrategy, nodeheuristic, etc. can greatly affect the performance of solving the problem, and the process of selecting suitable parameters is also called parameter adjustment.

However, users of modeling languages often cannot understand the characteristics of each optimization problem and solver, and therefore set appropriate solvers and parameters to solve. Especially, when a user needs to solve a set of optimization models or a problem that the same model transforms different data when using a modeling language (referred to as a multi-instance problem in the embodiments provided in this specification), and desires to obtain suitable parameters on the set of problem, it cannot be supported on the basis of existing products. Therefore, how to automatically help a user to refer to a group of optimization models in the modeling language and selecting proper solving parameters is an important subject related to the use of the modeling language.

In the present specification, a data processing method is provided, and the present specification relates to a data processing apparatus, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

Referring to fig. 1, fig. 1 is a schematic diagram illustrating a data processing method according to an embodiment of the present disclosure, and as shown in fig. 1, the data processing method provided in the embodiment of the present disclosure is applied to a terminal 100, where the terminal 100 may be a terminal device such as a notebook computer, an intelligent terminal, a server, and a cloud server.

In the terminal 100, a user opens a parameter adjusting mode in a modeling language, and inputs configuration information of a parameter adjusting device, such as maximum parameter adjusting time, parameter adjusting target and other information, where the maximum parameter adjusting time specifically refers to how long it takes to obtain a parameter adjusting result, and the parameter adjusting target specifically refers to desired information of the user, such as a threshold value reached by a user desired precision, or short parameter adjusting time.

Meanwhile, the terminal 100 may also receive at least one to-be-processed service data and at least one to-be-processed service model uploaded by a user, where it should be noted that the to-be-processed service data and the to-be-processed service model may be different to-be-processed service data of the same to-be-processed service model, or may be multiple groups of to-be-processed models and to-be-processed service data corresponding to each to-be-processed model, which is not limited in the embodiments provided in this specification.

After the service data to be processed and the service model to be processed are obtained, the separated service data to be processed and the separated service model to be processed are converted through a preset modeling language to generate a plurality of model examples corresponding to each other, then a proper solver is selected, and solving parameters of the solver are adjusted on the multi-model examples according to configuration information of a parameter adjustor and a preset parameter adjusting algorithm, wherein the configuration information is input by a user in advance, so that the solver can select solving parameters which are proper relative to the plurality of model examples.

In the data processing method provided by an embodiment of the specification, a multi-model parameter adjusting method based on a modeling language is designed, a multi-model and single-model multi-data-source multi-instance problem in the modeling language can be adjusted, a parameter adjusting system and the modeling language are combined for use, and the characteristic of model and data separation of the modeling language is fully utilized, so that an abstract model and different optimized model instances generated by multiple groups of data can be adjusted in the modeling language, and meanwhile, multiple groups of model parameter adjusting are supported. In order to reduce the overall time of parameter adjustment and consumption of other computing resources, the parameter adjustment process is divided into two stages, a multi-model example is generated by utilizing a modeling language in the first stage, a parameter adjustment algorithm is applied to adjust solving parameters of a solver on the multi-model example in the second stage, parameter adjustment of multiple models is realized, and meanwhile, a parameter adjustment system and the modeling language are effectively combined, so that parameter adjustment can be directly used in the modeling language, a user does not need to set the solving parameters, the use threshold of the user is reduced, and the use experience of the user is improved.

Referring to fig. 2, fig. 2 is a flowchart illustrating a data processing method according to an embodiment of the present disclosure, which specifically includes the following steps.

Step 202: receiving a parameter adjusting instruction, wherein the parameter adjusting instruction carries parameter adjusting configuration information.

The parameter adjusting instruction is specifically an instruction for adjusting solving parameters of a solver corresponding to at least one model based on a parameter adjuster. The parameter adjustment instruction may carry parameter adjustment configuration information set by a user, and specifically, the parameter adjustment configuration information may include maximum parameter adjustment time, a parameter adjustment target, and the like, where the maximum parameter adjustment time may be understood as time required by the whole parameter adjustment process, and the parameter adjustment target may be understood as a result to be achieved after the user desires parameter adjustment, for example, the parameter adjustment time is short, the parameter adjustment precision is high, and the like.

In practical application, a user can open a parameter-adjusting mode in a Modeling Language, the Modeling Language is software for assisting the user in generating a corresponding mathematical model based on a service problem to be solved, common Modeling languages include AMPL (artificial Programming Language), GAMS (the general algebrac Modeling System) and the like, taking AMPL as an example, AMPL is a comprehensive data model Language for solving linear, nonlinear and integer mathematical Programming problems in an optimization process, which cannot directly solve the problems, and has a function similar to a compiler to convert a model (.mod) into a special file (.nl), and after reading in a model file and a data file, a solver capable of solving various mathematical Programming problems is called to solve the problems.

After selecting the corresponding modeling language, the user sets the relevant configuration information of the parameter adjuster in the corresponding parameter adjuster mode, generates a parameter adjusting instruction based on the parameter configuration information, and sends the parameter adjusting instruction to the terminal. The terminal can receive the parameter adjusting instruction carrying the parameter adjusting configuration information.

Further, a user may send a parameter adjustment instruction to the cloud-side device through the terminal in a form of inputting an interactive command, for example, the user opens a parameter adjuster mode (tuner) in the modeling language, sets a configuration of the parameter adjuster (tuner _ options), and inputs' options; optionals _ options "tuner _ time = 10, run _obj = time"'. Wherein tuner _ time represents the maximum parameter adjusting time, and run _ obj represents the parameter adjusting target.

In addition, the user can also input the parameter adjusting configuration information of the user in a visual interface of the modeling language, specifically, the parameter adjusting configuration information can be selected or input in a checking and inputting mode, so that the cloud side equipment receives the parameter adjusting configuration information.

Step 204: and acquiring at least one service model to be processed and at least one service data to be processed.

In the modeling language, a user can input a complex service model into the model through semantics in an algebraic expression form and then introduce service data, for example, a certain target service is abstracted into a service model to be processed "minax + b", and the service data to be processed corresponding to the service model to be processed may be "a =1, b =1".

In practical application, the to-be-processed service model is a mathematical programming problem formed by abstracting a problem to be solved by the to-be-processed service, the to-be-processed service data specifically refers to service data corresponding to the to-be-processed service model, and in an embodiment provided in this specification, the service specifically refers to a specific service application scenario in practical application. For example, the unit combination problem, the transformer combination problem in an energy power scenario; dock allocation problems with vessel berthing; a routing line distribution problem of a network distribution scenario; channel allocation problems for satellite communication scenarios; a warehouse scenario cargo distribution problem, etc.

Taking an energy power problem as an example, the energy power problem can be abstracted to an economic scheduling problem, the economic scheduling problem is an optimization problem of minimizing supply cost by meeting energy requirements and deciding output (power output) of different generators on a time section under the condition of meeting the operation constraint of a power system, and in a basic model of the economic scheduling problem, the output cost is a linear function related to the output on the assumption that only a thermal generator set exists, so that the economic scheduling problem can be abstracted to a linear programming problem.

In the problem, assuming that the combination of generators is G, the generator G belongs to G, wherein the service data to be processed comprises: fixed cost of generator g operation is f _g With varying cost ofu _g With predicted power demand d and decision variable being the output of each generator gp _g . This problem can be abstracted as making the total cost of the generator combinationcg（p _g ) Minimum, wherein the force cost function can be reduced to a linear function related to the force, i.e. forcecg（p _g ）=f _g +u _g *p _g . In addition to this, there are constraints, such as constraining the minimum maximum output limit for each generator:

. The sum of the power outputs of each generator is to meet the predicted power demand d.

In practical applications, the to-be-processed service model and the to-be-processed service data may be uploaded by a user, and based on this, in a specific embodiment provided by the present application, the obtaining at least one to-be-processed service model and at least one to-be-processed service data includes:

responding to a data uploading instruction of a user, and displaying a data uploading page;

and receiving at least one to-be-processed business model and at least one to-be-processed business data uploaded by the user based on the data uploading page.

The data uploading page is a visual page displayed to a user by the terminal based on a data uploading instruction sent by the user, the data uploading page is displayed to the user by the user, a visual processing page is provided for the user, and the user can conveniently upload the service model to be processed and the service data to be processed in the data uploading page.

Furthermore, when the user performs corresponding processing on the operation terminal, the user can send a data uploading instruction to the terminal, and the terminal responds to the instruction to display a data uploading page for the user, so that the to-be-processed service model and the to-be-processed service data can be uploaded based on the data uploading page, and the function of acquiring the to-be-processed service model and the to-be-processed service data is realized.

In practical application, the at least one to-be-processed service model and the at least one to-be-processed service data may be one to-be-processed service model and a plurality of to-be-processed service data, or may be a plurality of to-be-processed service models and to-be-processed service data corresponding to each to-be-processed service model. The two cases will be explained in detail below.

In a specific embodiment provided in this specification, the acquiring at least one to-be-processed service model and at least one to-be-processed service data includes:

and acquiring a service model to be processed and a plurality of service data to be processed.

In the method provided by the present specification, solution parameters of a solver of a multi-model are adjusted correspondingly, and accordingly, a plurality of model instances need to be generated subsequently, and in the process of generating a plurality of model instances, one to-be-processed service model may be generated by combining a plurality of to-be-processed service data. For example, in the case of dock allocation in which the service to be processed is a ship dock, the model of the service to be processed corresponding to the dock allocation service is abstracted to "3ax +2by =0", where a corresponds to different types of dates, such as a =1 for a working day, a =2 for a weekend, a =3 for a holiday, etc., and b corresponds to different time periods of a day, e.g., b =1 may represent 8.

Taking a service model to be processed as a unit combination allocation scenario as an example, in the service scenario, a unit combination for power generation on the next day needs to be predicted every day, and for the unit combination problem, the corresponding service models to be processed are the same, but the power consumption requirements in different time periods every day are different, so that the service data to be processed corresponding to different time periods every day are different.

In another specific embodiment provided in this specification, acquiring at least one to-be-processed service model and at least one to-be-processed service data includes:

and acquiring a plurality of to-be-processed service models and to-be-processed service data corresponding to each to-be-processed service model.

In addition to the above-mentioned one to-be-processed model and a plurality of to-be-processed service data, the method provided in this specification may also select appropriate solution parameters for a plurality of different models, and in practical applications, for a modeling platform, various service models may be received, and if each service model is separately solved, computational resources may be wasted.

Based on this, an embodiment of the present specification may further receive a plurality of to-be-processed service models, where each to-be-processed model has to-be-processed service data corresponding to the to-be-processed model, for example, the received to-be-processed service model 1 is "3ax +2by =0", and the corresponding to-be-processed service data1 is "a =1, b =1"; the to-be-processed business model 2 is' cx ² +3y =0", and its corresponding to-be-processed service data2 is" c =3"; the pending Business model 3 is "(ay) ² -bz)/cz =0", and its corresponding pending traffic data 3 is" a =3, b =2, c =5 "\8230;.

In the embodiment provided in this specification, the obtaining of at least one to-be-processed service model and at least one to-be-processed service data is input through an interactive command, taking the input of one to-be-processed service model and a plurality of to-be-processed service data as an example, a user inputs an interactive command at a front end: the method comprises the steps of obtaining a to-be-processed service model and to-be-processed service data carried in an interactive instruction by cloud side equipment after the interactive instruction is received by the cloud side equipment, wherein the to-be-processed service model comprises 'example. Mapl', '8230,' \ 8230, and 'ins2. Data' \ 8230, and 'ins10. Data'. The to-be-processed service model and the to-be-processed service data carried in the interactive instruction are obtained by the example.

Taking the example of inputting a plurality of service models to be processed and service data to be processed corresponding to each service model to be processed, a user inputs' modelexample1.Mapl data1.Dat at the front end; model example2. Male data2.Dat \8230;, wherein example1.Map is service data to be processed corresponding to the service model 1 to be processed, and dat 1.Dat is service data to be processed corresponding to example1. Map; example2.Mapl is pending business model 2, and data2.Dat is pending business data \8230 \ 8230;. Corresponding to example1. Map2. After receiving the interactive instruction, the cloud-side device obtains a to-be-processed service model and to-be-processed service data carried in the interactive instruction.

Step 206: and generating a plurality of business model instances according to the at least one business model to be processed and the at least one business data to be processed based on a preset modeling language.

The preset modeling language is specifically used for combining the service model to be processed and the service data to be processed to generate corresponding service model software, and at least one service model to be processed and at least one service data to be processed can be combined to generate corresponding service model instances through the preset modeling language.

Examples of the predetermined Modeling languages include AMPL (Artificial Programming Language), GAMS (the general Algebriac Modeling System), and AIMMS. AMPL is a modeling language for describing and solving large-scale complex mathematical problems, and supports most solvers on the market; GAMS is an advanced modeling system for mathematical planning and optimization, and can solve numerical solutions of various nonlinear optimization problems; the AIMMS is an optimization-based language development environment that can be connected to solvers (linear, integer, non-linear, etc.) for different mathematical plans. The business model to be processed and the business data to be processed can be combined together through a preset modeling language to generate a corresponding business model instance.

Taking the AMPL as an example, the AMPL itself cannot directly solve the service problem, and has a function similar to a compiler, converts the received service model to be processed and the service data to be processed into an nl file, and after reading in the model file and the data file, calls other solvers capable of solving various mathematical programming problems to solve the nl file, thereby obtaining a solution result.

In practical applications, as described in the above steps, the modeling language has a characteristic of separating a model from data, and can generate a plurality of different model instances according to one model and a plurality of sets of data, and also can receive a plurality of sets of models and data. In the following, the two cases will be explained separately.

In a specific embodiment provided in this specification, acquiring a to-be-processed service model and a plurality of to-be-processed service data, and correspondingly generating a plurality of service model instances according to the at least one to-be-processed service model and the at least one to-be-processed service data based on a preset modeling language includes:

and generating a plurality of business model instances according to the business model to be processed and the plurality of business data to be processed based on a preset modeling language.

Following the above example of steps, taking the service to be processed as the dock allocation for ship docking as an example for explanation, the service model to be processed is abstracted to "3ax +2by =0", the corresponding plurality of service data to be processed is "a =1, b =1", "a =1, b =2", "a =1, b =3", "a =2, b =1", "a =3, b =1", each service data to be processed can be brought into the service model to be processed, and service model instances "3x +2y =0", "3x +4y =0", "3x +6y =0", "6x +2y =0", "9x +2y =0" are obtained.

In another specific embodiment provided in this specification, acquiring a plurality of to-be-processed service models and to-be-processed service data corresponding to each to-be-processed service model, and correspondingly generating a plurality of service model instances according to the at least one to-be-processed service model and the at least one to-be-processed service data based on a preset modeling language includes:

and generating a plurality of business model instances according to each business model to be processed and the business data to be processed corresponding to each business model to be processed based on a preset modeling language.

Following the above example of steps, the pending service model 1 is "3ax +2by =0", and the corresponding pending service data1 is "a =1, b =1"; the to-be-processed business model 2 is' cx ² +3y =0", and its corresponding to-be-processed service data2 is" c =3"; the pending service model 3 is "(ay) ² -bz)/cz =0", and the corresponding to-be-processed service data 3 is" a =3, b =2, c =5", and a service model instance 1"3x +2y =0 "can be generated according to the to-be-processed service data1 and the to-be-processed service model 1; service model instance 2 and 3x can be generated according to the service data2 to be processed and the service model 2 to be processed ² +3y =0"; business model instance 3' (3 y) can be generated according to the business data 3 to be processed and the business model 3 to be processed ² -2z)/5z=0”。

Step 208: and determining an initial solver, and adjusting solving parameters of the initial solver in the plurality of service model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information to obtain a target solver.

After the above steps, a plurality of service model instances can be determined, that is, one solver can be selected to solve each model instance, so as to determine target solution parameters corresponding to the solver in the plurality of service model instances according to a preset parameter adjusting algorithm and parameter adjusting configuration information, and the target solution parameters are used to set the solver, thereby obtaining a final target solver.

The initial solver specifically refers to a solver not configured with appropriate solving parameters, and the target solver specifically refers to a solver configured with appropriate solving parameters.

In practical application, there are many types of solvers, some solvers can only process linear programming problems, some solvers can only process nonlinear programming problems, some solvers can process linear programming problems and nonlinear programming problems, and after a plurality of service model instances are obtained, a proper solver needs to be selected to process the plurality of service model instances. The initial solver can be set by a user, and can also be automatically selected by a parameter adjusting method provided by the specification.

Specifically, determining the initial solver includes:

receiving a solver selection instruction of a user, and determining an initial solver in a solver set according to a solver identifier carried in the solver selection instruction; or

And obtaining the model type of the service model to be processed, and determining an initial solver in the solver set according to the model type of the service model to be processed.

When a user has corresponding parameter adjusting knowledge, a proper solver can be selected by himself, specifically, the user can select a proper initial solver at the end side, send a solver selection instruction to the cloud-side device based on the initial solver, the cloud-side device receives the solver selection instruction of the user, and determines the initial solver in the solver set according to a solver identifier carried in the solver selection instruction.

When a user does not have corresponding parameter adjusting knowledge and cannot judge which solver is selected, a suitable initial solver can be automatically selected for the user through the parameter adjusting method provided by the specification. Specifically, selecting an initial solver from a solver set according to the model type of the service model to be processed, and further determining the initial solver in the solver set according to the model type of each service model to be processed, including:

determining a solver that processes each model type in the set of solvers as an initial solver.

In practical application, the model type specifically refers to a problem type of each to-be-processed model for solving a business problem, for example, if the business problem is a linear programming problem, the model type is a linear programming problem; if the service problem is a mixed integer linear programming problem, the model type is the mixed integer linear programming problem; if the business problem is a mixed integer nonlinear programming problem, the model type is a mixed integer nonlinear programming problem and the like.

The solver set specifically refers to a set of all solvers for a user to use, and it should be noted that the solver set herein refers to a solver for which the user has a use right, for example, there are 20 solvers on the market, and if the user a only has authorization information of 10 solvers, for the user a, the corresponding solver set includes the 10 solvers; if user B has authorization information of 5 types of solvers, the corresponding solver set of user B comprises the 5 types of solvers.

In practical application, each solver has different processing capabilities, and the types of models that can be processed are different, for example, some solvers can only process linear programming problems, some solvers can only process mixed integer nonlinear programming problems, some solvers can only process nonlinear programming problems, and some solvers can process both linear programming problems and nonlinear programming problems; in addition, a solver set of one user may also have a plurality of solvers of model types at the same time, for example, similarly, solving a linear programming problem may include a solver 1, a solver 2, and a solver 3, where the computing resources of each solver are different, and preferably, a solver with a smaller computing resource is selected as an initial solver.

In practical application, if the model type has a linear programming problem and a nonlinear programming problem, the selected initial solver has the capability of simultaneously processing the linear programming problem and the nonlinear programming problem; if the model type only has the linear programming problem, the selected initial solver only needs to have the capability of processing the linear programming problem, and not only the initial solver which only can process the linear programming problem can be selected, but also the initial solver which can process the linear programming problem and other problems can be selected. Preferably, there are multiple solvers that can handle the linear programming problem, and in order to improve the calculation efficiency, the solver that occupies the least amount of calculation resources may be selected as the initial solver, and in addition, the available solver may be fed back to the front end, and a user selects one initial solver from the multiple available solvers.

In the solver set, at least one solver exists, the types of models which can be processed by each solver are different, and some solvers can only process linear programming problems; some solvers can not only process the linear programming problem, but also process the nonlinear programming problem; based on the above, a proper initial solver can be determined according to each model type, and if a plurality of models to be processed are all linear programming problems, a solver for processing the linear programming problems can be selected; if the plurality of models to be processed include both linear programming problems and nonlinear programming problems, a solver capable of processing both linear programming problems and nonlinear programming problems needs to be selected.

After the initial solver is determined, solution parameters in the initial solver need to be adjusted, so that the solver has better processing performance for a plurality of service model instances. Furthermore, the solving hyper-parameters in the initial solver need to be adjusted, so as to generate the target solver corresponding to the initial solver.

Adjusting solution parameters of the initial solver in the plurality of service model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information to obtain a target solver, wherein the method comprises the following steps:

solving the initial solver in a plurality of service model instances according to a preset parameter adjusting algorithm to obtain a solving result;

adjusting initial solving parameters of the initial solver according to the parameter adjusting configuration information and the solving result to obtain target solving parameters;

and generating a target solver according to the target solving parameters.

The preset parameter adjusting algorithm specifically refers to an algorithm used in a parameter adjusting process of the solver according to a service model instance, and includes an IteratedLocalSearch algorithm, a modebasedsurrogate algorithm, an iteratedlacing algorithm, and the like. After the preset parameter adjusting algorithm is determined, solving is carried out on the basis of the initial solver in a plurality of service model instances according to the parameter adjusting algorithm to obtain corresponding solving results, solving parameters of the initial solver are adjusted according to parameter adjusting configuration information and solving results input by a user to obtain adjusted solving parameters, and the solver is set by using new solving parameters. And then, continuously solving the plurality of service model instances by using a new solver, and iterating the solving parameters of the solver according to the solving results.

In the embodiments provided in this specification, the parameter adjuster iterates the combination of solution parameters, generates new reference solution parameters according to a preset parameter adjustment algorithm, the solver solves a plurality of service model instances according to the reference solution parameters, the solution result is returned to the parameter adjuster, the parameter adjuster evaluates whether the set of reference solution parameters is target solution parameters, generates the next set of reference solution parameters, and continues iteration until the target solution parameters are selected.

Specifically, solving the initial solver in a plurality of service model instances according to a preset parameter adjustment algorithm to obtain a solution result includes:

generating reference solving parameters according to a preset parameter adjusting algorithm;

and solving the plurality of business model instances based on the reference solving parameters to obtain a solving result.

In practical application, a parameter adjuster generates a set of reference solving parameters according to a preset parameter adjusting algorithm, and solves a plurality of service model examples according to the set of reference solving parameters through a solver to obtain solving results, wherein in the examples provided in the specification, the solving results comprise solving time length, solving occupied resources and the like.

Further, adjusting the initial solution parameters of the initial solver according to the parameter adjusting configuration information and the solution result to obtain target solution parameters, including:

evaluating the solving result and the reference solving parameter based on the parameter adjusting configuration information, and obtaining an evaluation result;

determining whether the reference solving parameter is a target parameter according to the evaluation result;

if so, determining the reference solving parameter as a target solving parameter;

if not, continuing to execute the operation of generating the reference solving parameters according to the preset parameter adjusting algorithm.

After the solving result is obtained, the reference solving parameter can be evaluated according to the parameter adjusting configuration information and the solving result, whether the reference solving parameter is the final target parameter or not is determined according to the evaluating result, if yes, the reference solving parameter is the target solving parameter, and if not, iterative training is continued.

For example, setting a parameter adjusting target in the parameter adjusting configuration information to be the shortest time, setting a reference solving parameter 1, obtaining a solving time duration t1 in a solving result according to the reference solving parameter 1, setting a reference solving parameter 2, obtaining a solving time duration t2 in a solving result according to the reference solving parameter 2, comparing t1 and t2, and if t1 is small, taking the reference solving parameter 1 as a candidate reference solving parameter. At the moment, a reference solving parameter 3 is set again, according to the solving time length t3 in the solving result obtained by the reference solving parameter 3, the t3 is compared with the t1, if the t3 is small, the reference solving parameter 3 is used as a candidate reference solving parameter, if the t1 is small, the reference solving parameter 1 is continuously used as a candidate reference solving parameter, and after multiple iterations, the final reference solving parameter is selected as a target solving parameter.

Furthermore, in practical application, multiple rounds of iteration can be set through parameter adjusting configuration information, for example, 5 rounds of iteration are set, and after 5 rounds of iteration are carried out, the selected solving parameters are target solving parameters; or setting the same reference solving parameter to be unchanged after 5 iterations, and determining the reference solving parameter as the target solving parameter; or setting the parameter adjusting time length to be 10 minutes, and then selecting the reference solving parameter as the target solving parameter after the parameter adjusting time length is reached. In the embodiments provided in the present specification, the stop condition of the multiple iterations is not limited, and the setting in actual application is used as the standard.

After the target solution parameters meeting the parameter adjustment configuration information are obtained, a solver can be set according to the target solution parameters, so that a final target solver is generated, and the target solver can process the corresponding to-be-processed service model in the adjustment of the solution parameters.

The data processing method provided by the present specification receives a parameter adjustment instruction, where the parameter adjustment instruction carries parameter adjustment configuration information; acquiring at least one service model to be processed and at least one service data to be processed; generating a plurality of business model instances according to the at least one business model to be processed and the at least one business data to be processed based on a preset modeling language; determining an initial solver, and adjusting solving parameters of the initial solver in the plurality of service model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information to obtain a target solver. By means of the present method, it is possible to,

according to the method provided by the embodiment of the specification, the parameter adjusting method is combined with the modeling language, the characteristic that the model and the data in the modeling language are separated is fully utilized, the at least one service model to be processed and the at least one service data to be processed are generated into a plurality of service model examples, the preset parameter adjusting algorithm is called to adjust the parameters of the solver on the plurality of model examples, the final target solver is obtained, the overall parameter adjusting time is shortened, the solving parameters do not need to be set by a user, the use threshold of the user is lowered, and the use experience of the user is improved.

Secondly, the process of determining the solver can be automatically determined according to the model type of the service model to be processed, so that the process of human participation is further reduced, a user can operate the solver without learning a large amount of related knowledge, and the use experience of the user is improved.

The following describes the data processing method further by taking the application of the data processing method provided in this specification in the unit combination prediction as an example, with reference to fig. 3. Fig. 3 shows a flowchart of a processing procedure of a data processing method according to an embodiment of the present specification, which specifically includes the following steps.

Step 302: receiving a parameter adjusting instruction aiming at a solver, wherein the parameter adjusting instruction carries parameter adjusting configuration information.

Step 304: and responding to a data uploading instruction of a user, and displaying a data uploading page.

Step 306: and receiving a unit combination prediction service model uploaded by the user based on the data uploading page and a plurality of to-be-processed service data corresponding to the unit combination prediction service model.

Step 308: and generating a service model instance corresponding to a plurality of unit combination services according to the unit combination prediction service model and the plurality of to-be-processed service data.

Step 310: and determining an initial solver in the model type solver set of the unit combination prediction service model.

Step 312: and solving the initial solver in a plurality of service model instances according to a preset parameter adjusting algorithm to obtain a solving result.

Step 314: and carrying out iterative adjustment on the initial solving parameters of the initial solver according to the parameter adjusting configuration information and the solving result to obtain target solving parameters.

Step 316: and generating a target solver according to the target solving parameters.

The method provided by the embodiment of the description combines the parameter adjusting method with the modeling language, fully utilizes the characteristic of model and data separation in the modeling language, generates a plurality of service model instances according to one service model to be processed and a plurality of service data to be processed, and calls the preset parameter adjusting algorithm to adjust the parameters of the solver on the multi-model instances to obtain the final target solver, thereby reducing the overall parameter adjusting time, avoiding the need of setting solving parameters by a user, lowering the use threshold of the user and improving the use experience of the user.

The following describes the data processing method further by taking the application of the data processing method provided in this specification to a plurality of service scenarios as an example with reference to fig. 4. Fig. 4 shows a flowchart of a processing procedure of a data processing method according to an embodiment of the present specification, which specifically includes the following steps.

Step 402: receiving a parameter adjusting instruction aiming at a solver, wherein the parameter adjusting instruction carries parameter adjusting configuration information.

In this embodiment, a parameter adjustment instruction sent by a user at a front end is received, where the parameter adjustment instruction is specifically an interactive command, and the user inputs the interactive parameter adjustment instruction at the front end, where the parameter adjustment instruction carries parameter configuration information.

Step 404: and responding to a data uploading instruction of a user, and displaying a data uploading page.

In this embodiment, a data upload instruction sent by a user at a front end is received, a corresponding data upload page is displayed for the user at the front end, a to-be-processed service model portion and a to-be-processed service data portion are displayed in the data upload page, the user can upload corresponding information in the two portions respectively, upload a to-be-processed service model in the to-be-processed service model portion, and upload to-be-processed service data corresponding to the to-be-processed service model in the to-be-processed service data portion.

Step 406: and the receiving user uploads a plurality of groups of information to be processed based on the data uploading page, wherein each group of information to be processed comprises a service model to be processed and service data to be processed.

In this embodiment, a user uploads a plurality of sets of information to be processed in a data upload page at a front end, where each set of information to be processed includes a service model to be processed and service data to be processed, and further, a set of information to be processed may include a service model to be processed and service data to be processed; the set of to-be-processed information may also include one to-be-processed service model and a plurality of to-be-processed service data.

Step 408: and generating corresponding business model examples according to each group of to-be-processed business models and to-be-processed business data, and obtaining a plurality of business model examples.

In this embodiment, each to-be-processed service model and to-be-processed service data corresponding to each to-be-processed service model are compiled into a corresponding service model instance according to a preset modeling language AMPL.

Step 410: and determining a solver capable of processing each model type in the solver set as an initial solver according to the model types of the plurality of service models to be processed.

In this embodiment, taking 3 types of models of a plurality of service models to be processed as examples, the types are linear programming, integer linear programming and nonlinear programming respectively. The solver set comprises 8 solvers, wherein the solver 2 and the solver 8 can simultaneously solve the problems of linear programming, integer linear programming and nonlinear programming. Further, comparing the consumption resource values of the solver 2 and the solver 8 during solving, wherein the consumption resource value solved by the solver 2 is higher than that of the solver 8, and therefore, the solver 8 is selected as an initial solver.

Step 412: and solving the initial solver in a plurality of service model instances according to a preset parameter adjusting algorithm to obtain a solving result.

In this embodiment, the parameter adjuster sets a reference solving parameter 1, and the solver 8 solves a plurality of service model instances according to the reference solving parameter 1 to obtain a solving result 1 corresponding to the reference solving parameter 1; and the parameter adjuster generates the reference solving parameters 2 again according to a preset parameter adjusting algorithm, and the solver 8 solves the plurality of service model instances according to the reference solving parameters 2 to obtain solving results 2 corresponding to the reference solving parameters 2.

Step 414: and iteratively adjusting the initial solving parameters of the initial solver according to the parameter adjusting configuration information and the solving result to obtain target solving parameters.

In this embodiment, according to the parameter-adjusting configuration information input by the user in advance, the solution result 1 and the solution result 2 are compared, the reference solution parameter corresponding to the solution result more meeting the requirement of the parameter-adjusting configuration information is selected, the operation of step 412 is continuously executed to obtain a solution result 3, the solution result 3 is compared, the reference solution parameter more meeting the parameter-adjusting configuration information is selected, and iterative adjustment is continuously performed until the stop condition in the parameter-adjusting configuration information is reached, so that the final target solution parameter is obtained.

Step 416: and generating a target solver according to the target solving parameters.

In this embodiment, after the target solution parameters are determined, a final target solver may be generated according to the target solution parameters and the initial solver.

The method provided by the embodiment of the description combines the parameter adjusting method with the modeling language, fully utilizes the characteristic of model and data separation in the modeling language, generates a plurality of service model instances according to a plurality of service models to be processed and a plurality of service data to be processed, and calls the preset parameter adjusting algorithm to adjust the parameters of the solver on the multi-model instances to obtain the final target solver, so that the overall parameter adjusting time is reduced, the user does not need to set solving parameters, the use threshold of the user is reduced, and the use experience of the user is improved.

Referring to fig. 5, fig. 5 shows a data processing method provided in an embodiment by this specification, where the method is applied to a cloud-side device, and in this embodiment, the data processing method is applied to a power scheduling service scenario, and is used to solve a power scheduling problem, and specifically includes:

step 502: receiving a parameter adjusting instruction sent by a user at an end-side device for the power dispatching service, wherein the parameter adjusting instruction carries parameter adjusting configuration information.

Specifically, a user inputs a parameter adjustment instruction for the power scheduling service at an end side, and after receiving the parameter adjustment instruction, the cloud-side device acquires the parameter adjustment configuration information carried in the parameter adjustment instruction.

Step 504: and acquiring a power cost model and at least one power cost data corresponding to the power dispatching service.

Specifically, in the service scene, the energy and power problem can be abstracted to an economic scheduling problem, the economic scheduling problem is an optimization problem of minimizing supply cost by meeting energy requirements and deciding output (power output) of different generators on a time section under the condition of meeting the operation constraint of a power system, and in a basic model of the economic scheduling problem, the output cost is a linear function related to the output on the assumption that only a thermal generator set is provided, so that the economic scheduling problem can be abstracted to a linear programming problem.

Based on the method, the generator combination is taken as G, the generator G belongs to G for example, and the fixed running cost of the generator G is f _g With varying cost ofu _g With predicted power demand d and decision variable being the output of each generator gp _g . This problem is abstracted as a model to compute the power cost of the generator set assemblycg（p _g ) At least one power cost data comprising a fixed cost f of operation of the generator g _g With varying cost ofu _g The predicted power demand is d and the decision variable is the output of each generator gp _g 。

Step 506: generating a plurality of power cost model instances from the power cost model and the at least one power cost data based on a preset modeling language.

Wherein generating a plurality of power cost model instances from the power cost model and the at least one power cost data based on a preset modeling language comprises:

generating a plurality of initial power cost model instances from the power cost model and the at least one power cost data based on a preset modeling language;

sending the plurality of initial power cost model instances to the end-side device;

and receiving an instance selection instruction sent by a user at the end-side equipment, and determining a power cost model instance in the plurality of initial power cost model instances based on the instance selection instruction.

From the power cost model and the at least one power cost data, the contribution cost function may be reduced to a linear function related to the contribution, i.e. the contributioncg（p _g ）=f _g +u _g *p _g . Simultaneously setting the constraint conditions of each corresponding generator:

to constrain the maximum and minimum output limits of each generator while ensuring that the sum of the generated outputs of each generator meets the predicted power demand d.

Step 508: determining an initial solver, and adjusting solving parameters of the initial solver in the multiple power cost model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information to obtain a target solver.

Wherein, determining an initial solver comprises:

receiving a solver selection instruction sent by a user at the end-side equipment, and determining an initial solver in a solver set according to a solver identifier carried in the solver selection instruction; or

And determining an initial solver in the solver set according to the model type of the power cost model.

Further, determining an initial solver in a solver set according to the model type of the power cost model, including:

determining an initial solver in a solver set according to the model type based on a preset solver selection strategy; or

Determining a solver set to be screened in the solver set according to the model type, feeding back the solver set to be screened to the end-side equipment, receiving a selection instruction input by a user at the end-side equipment, and determining an initial solver in the solver set to be screened according to the selection instruction.

Specifically, if the user has the relevant parameter adjusting knowledge, a solver selection instruction can be directly sent out from the end-side device, the solver selection instruction carries a solver identifier, and a corresponding initial solver can be selected from the solver set through the solver identifier.

If the user does not have related parameter adjusting knowledge, the cloud-side device may select usable solvers from the solver set according to the model type of the power cost model to form a solver set to be screened, and the solver set to be screened may include solver attributes of each solver to be screened, such as information about resource occupation and solving speed for solving. And feeding back the solver set to be screened to the end-side equipment, wherein a user can select one solver from the solver set to be screened as an initial solver.

After the initial solver is determined, the parameter adjuster generates a group of reference solving parameters according to a preset parameter adjusting algorithm, the initial solver carries out solving on the power cost model based on the reference solving parameters and obtains solving results, the solving results are returned to the parameter adjuster, the parameter adjuster judges whether the solving results meet parameter adjusting configuration information or not and generates the next group of reference solving parameters, iterative adjustment is continued until the solving results generated by the generated reference solving parameters meet the parameter adjusting configuration information, and therefore final target solving parameters are obtained, and a target solver is obtained based on the target solving parameters.

According to the method provided by the embodiment of the description, the parameter adjusting method is combined with the modeling language, the characteristic that the model and the data in the modeling language are separated is fully utilized, a plurality of business model examples are generated according to one business model to be processed and a plurality of business data to be processed, a preset parameter adjusting algorithm is called to adjust the parameters of the solver on the multi-model examples, a final target solver is obtained, the overall parameter adjusting time is shortened, the solving parameters do not need to be set by a user, the use threshold of the user is lowered, and the use experience of the user is improved.

Referring to fig. 6, fig. 6 shows a data processing system provided in an embodiment of the present specification, where the system may include an end-side device 601 and a cloud-side device 602, where the end-side device 601 is configured to send a parameter adjustment instruction to the cloud-side device 602, and receive a target solver sent by the cloud-side device 602; the cloud-side device 602 is configured to perform an operation of adjusting a solution parameter of a solver on the received parameter adjustment instruction.

The end-side device 601 is configured to send a parameter adjustment instruction to the cloud-side device, where the parameter adjustment instruction carries parameter adjustment configuration information;

the cloud-side device 602 is configured to receive the parameter adjustment instruction, obtain at least one to-be-processed service model and at least one to-be-processed service data, generate a plurality of service model instances according to the at least one to-be-processed service model and the at least one to-be-processed service data based on a preset modeling language, determine an initial solver, adjust solution parameters of the initial solver in the plurality of service model instances according to a preset parameter adjustment algorithm and the parameter adjustment configuration information, obtain a target solver, and send the target solver to the end-side device.

The cloud-side device 602 may be a central cloud device of a distributed cloud architecture, the end-side device 601 may be an edge cloud device of the distributed cloud architecture, and the cloud-side device 602 and the end-side device 601 may be server-side devices such as a conventional server, a cloud server, or a server array, or may be terminal devices, which is not limited in this description embodiment. Moreover, the cloud-side device 602 provides super-strong computing and storage capabilities, and is far away from the user; and the deployment range of the end-side device 601 is large and is close to the user. The end-side device 601 is an extension of the cloud-side device 602, and can sink the computing capability of the cloud-side device 602 to the end-side device 601, and solve the service requirement that cannot be met in a centralized cloud computing mode through integration and cooperative management of end clouds.

Corresponding to the above method embodiment, the present specification further provides a data processing apparatus embodiment, and fig. 7 shows a schematic structural diagram of a data processing apparatus provided in an embodiment of the present specification. As shown in fig. 7, the apparatus includes:

a receiving module 702, configured to receive a parameter adjustment instruction, where the parameter adjustment instruction carries parameter adjustment configuration information;

an obtaining module 704 configured to obtain at least one to-be-processed service model and at least one to-be-processed service data;

a generating module 706 configured to generate a plurality of business model instances according to the at least one to-be-processed business model and the at least one to-be-processed business data based on a preset modeling language;

an adjusting module 708, configured to determine an initial solver, and adjust solution parameters of the initial solver in the multiple service model instances according to a preset parameter tuning algorithm and the parameter tuning configuration information, to obtain a target solver.

Optionally, the adjusting module 708 is further configured to:

And obtaining the model type of each service model to be processed, and determining an initial solver in the solver set according to the model type of each service model to be processed.

Optionally, the adjusting module 708 is further configured to:

Optionally, the obtaining module 704 is further configured to:

Optionally, the generating module 706 is further configured to:

and generating a plurality of business model examples according to one business model to be processed and a plurality of business data to be processed based on a preset modeling language.

Optionally, the obtaining module 704 is further configured to:

Optionally, the generating module 706 is further configured to:

and generating a plurality of business model instances according to at least one to-be-processed business model and to-be-processed business data corresponding to each to-be-processed business model based on a preset modeling language.

Optionally, the obtaining module 704 is further configured to:

and receiving at least one to-be-processed business model and at least one to-be-processed business data uploaded by a user based on the data uploading page.

Optionally, the adjusting module 708 is further configured to:

and generating a target solver according to the target solving parameters.

Optionally, the adjusting module 708 is further configured to:

evaluating the solving result and the reference solving parameter based on the parameter adjusting configuration information, and obtaining an evaluating result;

The data processing apparatus provided in this specification receives a parameter adjustment instruction, where the parameter adjustment instruction carries parameter adjustment configuration information; acquiring at least one service model to be processed and at least one service data to be processed; generating a plurality of business model instances according to the at least one business model to be processed and the at least one business data to be processed based on a preset modeling language; determining an initial solver, and adjusting solving parameters of the initial solver in the plurality of service model instances according to a preset parameter adjusting algorithm and the parameter adjusting configuration information to obtain a target solver. By means of the present method, it is possible to,

by the device provided by the embodiment of the specification, the parameter adjusting method is combined with the modeling language, the characteristic that the model and the data in the modeling language are separated is fully utilized, at least one service model to be processed and at least one service data to be processed are generated into a plurality of service model instances, a preset parameter adjusting algorithm is called to adjust the parameters of the solver on the multi-model instances, a final target solver is obtained, the overall parameter adjusting time is shortened, the solving parameters do not need to be set by a user, the use threshold of the user is lowered, and the use experience of the user is improved.

The above is a schematic configuration of a data processing apparatus of the present embodiment. It should be noted that the technical solution of the data processing apparatus and the technical solution of the data processing method belong to the same concept, and details that are not described in detail in the technical solution of the data processing apparatus can be referred to the description of the technical solution of the data processing method.

FIG. 8 illustrates a block diagram of a computing device 800, according to one embodiment of the present description. The components of the computing device 800 include, but are not limited to, memory 810 and a processor 820. The processor 820 is coupled to the memory 810 via a bus 830, and the database 850 is used to store data.

Computing device 800 also includes access device 840, access device 840 enabling computing device 800 to communicate via one or more networks 860. Examples of such networks include a Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the internet. The Access device 840 may include one or more of any type of Network interface (e.g., a Network interface controller) that may be wired or wireless, such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, a worldwide interoperability for Microwave Access (Wi-MAX) interface, an ethernet interface, a Universal Serial Bus (USB) interface, a cellular Network interface, a bluetooth interface, near Field Communication (NFC), or wireless Communication.

In one embodiment of the present description, the above-described components of computing device 800, as well as other components not shown in FIG. 8, may also be connected to each other, such as by a bus. It should be understood that the block diagram of the computing device architecture shown in FIG. 8 is for purposes of example only and is not limiting as to the scope of the description. Those skilled in the art may add or replace other components as desired.

Computing device 800 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet, personal digital assistant, laptop, notebook, netbook, etc.), a mobile phone (e.g., smartphone), a wearable computing device (e.g., smartwatch, smartglasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or Personal Computer (PC). Computing device 800 may also be a mobile or stationary server.

Wherein the processor 820 is configured to execute computer-executable instructions that, when executed by the processor, implement the steps of the data processing method described above. The above is an illustrative scheme of a computing device of the present embodiment. It should be noted that the technical solution of the computing device and the technical solution of the data processing method belong to the same concept, and details that are not described in detail in the technical solution of the computing device can be referred to the description of the technical solution of the data processing method.

An embodiment of the present specification also provides a computer-readable storage medium storing computer-executable instructions, which when executed by a processor, implement the steps of the data processing method described above.

The above is an illustrative scheme of a computer-readable storage medium of the present embodiment. It should be noted that the technical solution of the storage medium belongs to the same concept as the technical solution of the data processing method, and details that are not described in detail in the technical solution of the storage medium can be referred to the description of the technical solution of the data processing method.

An embodiment of the present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the data processing method.

The above is an illustrative scheme of a computer program of the present embodiment. It should be noted that the technical solution of the computer program and the technical solution of the data processing method belong to the same concept, and details that are not described in detail in the technical solution of the computer program can be referred to the description of the technical solution of the data processing method.

The foregoing description of specific embodiments has been presented for purposes of illustration and description. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The computer instructions comprise computer program code which may be in the form of source code, object code, an executable file or some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-only Memory (ROM), random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, etc. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.

It should be noted that, for the sake of simplicity, the foregoing method embodiments are described as a series of acts, but those skilled in the art should understand that the present embodiment is not limited by the described acts, because some steps may be performed in other sequences or simultaneously according to the present embodiment. Further, those skilled in the art should also appreciate that the embodiments described in this specification are preferred embodiments and that acts and modules referred to are not necessarily required for an embodiment of the specification.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

The preferred embodiments of the present specification disclosed above are intended only to aid in the description of the specification. Alternative embodiments are not exhaustive and do not limit the invention to the precise embodiments described. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the embodiments and the practical application, to thereby enable others skilled in the art to best understand and utilize the embodiments. The specification is limited only by the claims and their full scope and equivalents.

Claims

1. A data processing method is applied to cloud-side equipment and comprises the following steps:

2. The method of claim 1, determining an initial solver comprising:

3. The method of claim 2, determining an initial solver in a set of solvers according to the model type of the business model to be processed, comprising:

and determining a solver of the processing model type in the solver set as an initial solver.

4. The method of claim 1, generating a plurality of business model instances from the at least one business model to be processed and the at least one business data to be processed based on a preset modeling language, comprising:

5. The method of claim 1, generating a plurality of business model instances from the at least one business model to be processed and the at least one business data to be processed based on a preset modeling language, comprising:

and generating a plurality of business model examples according to at least one business model to be processed and business data to be processed corresponding to each business model to be processed based on a preset modeling language.

6. The method of claim 1, obtaining at least one pending business model and at least one pending business data, comprising:

7. The method of claim 1, wherein adjusting solution parameters of the initial solver in the plurality of service model instances according to a preset parameter tuning algorithm and the parameter tuning configuration information to obtain a target solver comprises:

and generating a target solver according to the target solving parameters.

8. The method of claim 7, solving the initial solver in a plurality of service model instances according to a preset parameter tuning algorithm to obtain a solution result, comprising:

9. The method of claim 8, adjusting initial solution parameters of the initial solver according to the parameter configuration information and the solution result to obtain target solution parameters, comprising:

10. A data processing method is applied to cloud-side equipment and comprises the following steps:

receiving a parameter adjusting instruction sent by a user on a terminal side device aiming at a power dispatching service, wherein the parameter adjusting instruction carries parameter adjusting configuration information;

11. The method of claim 10, generating a plurality of instances of a power cost model from the power cost model and the at least one power cost data based on a preset modeling language, comprising:

an instance selection instruction sent by a user at the end-side device is received, and a power cost model instance is determined in the plurality of initial power cost model instances based on the instance selection instruction.

12. The method of claim 10, determining an initial solver, comprising:

13. The method of claim 12, determining an initial solver in a set of solvers based on the model type of the power cost model, comprising:

14. A data processing system comprising:

the cloud-side device is configured to receive the parameter adjustment instruction, acquire at least one to-be-processed service model and at least one to-be-processed service data, generate a plurality of service model instances according to the at least one to-be-processed service model and the at least one to-be-processed service data based on a preset modeling language, determine an initial solver, adjust solution parameters of the initial solver in the plurality of service model instances according to a preset parameter adjustment algorithm and the parameter adjustment configuration information, acquire a target solver, and send the target solver to the end-side device.

15. A data processing device is applied to cloud-side equipment and comprises:

16. A computing device, comprising:

a memory and a processor;

the memory is for storing computer-executable instructions and the processor is for executing the computer-executable instructions, which when executed by the processor implement the steps of the method of any one of claims 1-9 or 10-13.

17. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, perform the steps of the method of any one of claims 1-9 or 10-13.