CN109426912A - Air control system optimization method, system, device and electronic equipment - Google Patents
Air control system optimization method, system, device and electronic equipment Download PDFInfo
- Publication number
- CN109426912A CN109426912A CN201710768969.7A CN201710768969A CN109426912A CN 109426912 A CN109426912 A CN 109426912A CN 201710768969 A CN201710768969 A CN 201710768969A CN 109426912 A CN109426912 A CN 109426912A
- Authority
- CN
- China
- Prior art keywords
- optimization
- air control
- control system
- performance monitoring
- performance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Feedback Control In General (AREA)
- Air Conditioning Control Device (AREA)
Abstract
This specification embodiment discloses air control system optimization method, system, device and electronic equipment, the air control system optimization method includes: to be monitored according to the data by air control event that air control system is handled to air control system performance, according to performance monitoring data, using scheduled optimization algorithm, parameter optimization globally is carried out to multiple machine learning models in air control system and policing rule.
Description
Technical field
This specification is related to field of computer technology more particularly to air control system optimization method, system, device and electronics
Equipment.
Background technique
Modern air control system has been developed as the complication system comprising multiple machine learning models and policing rule.In wind
During O&M after controlling system building process and coming into operation, generally require to optimize air control system, to be adapted to
Actual business demand.
In the prior art, artificial experience often is based on by research staff to the optimization of air control system, for individual machines
Device learning model individually carries out model optimization.
Based on the prior art, more effective air control system optimization scheme is needed.
Summary of the invention
This specification embodiment provides air control system optimization method, system, device and electronic equipment, following for solving
Technical problem: more effective air control system optimization scheme is needed.
In order to solve the above technical problems, this specification embodiment is achieved in that
The air control system optimization method that this specification embodiment provides, comprising:
According to the data by air control event that air control system is handled, the business for evaluating the air control system performance is calculated
Index；
According to the operational indicator, performance monitoring is carried out to the air control system；
The multiple machine learning models for include to the air control system according to the performance monitoring data and/or strategy rule
Then carry out parameter optimization, wherein optimised each parameter is default corresponding between the performance monitoring data according to it
What relationship determined.
The air control system optimization system that this specification embodiment provides, comprising:
Performance evaluation module is calculated according to the data by air control event that air control system is handled for evaluating the air control
The operational indicator of system performance；
Performance monitoring module carries out performance monitoring to the air control system according to the operational indicator；
Performance optimization module, according to the performance monitoring data, the multiple machine learning moulds for including to the air control system
Type and/or policing rule carry out parameter optimization, wherein optimised each parameter is according to itself and the performance monitoring data
Between default corresponding relationship determine.
The air control system optimization apparatus that this specification embodiment provides, comprising:
Computing module is calculated according to the data by air control event that air control system is handled for evaluating the air control system
The operational indicator of performance；
Monitoring module carries out performance monitoring to the air control system according to the operational indicator；
Optimization module, according to the performance monitoring data, multiple machine learning models for include to the air control system and/
Or policing rule carry out parameter optimization, wherein optimised each parameter be according to it between the performance monitoring data
What default corresponding relationship determined.
The a kind of electronic equipment that this specification embodiment provides, comprising:
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes so that at least one described processor can:
According to the data by air control event that air control system is handled, the business for evaluating the air control system performance is calculated
Index；
According to the operational indicator, performance monitoring is carried out to the air control system；
The multiple machine learning models for include to the air control system according to the performance monitoring data and/or strategy rule
Then carry out parameter optimization, wherein optimised each parameter is default corresponding between the performance monitoring data according to it
What relationship determined.
At least one above-mentioned technical solution that this specification embodiment uses can reach following the utility model has the advantages that can be by wind
The multiple machine learning models and policing rule for including in control system are considered as entirety, globally integrally carry out parameter optimization to this,
This air control system optimization scheme is more effective.
Detailed description of the invention
In order to illustrate more clearly of this specification embodiment or technical solution in the prior art, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
The some embodiments recorded in this specification, for those of ordinary skill in the art, in not making the creative labor property
Under the premise of, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is the overall architecture schematic diagram that the scheme of this specification is related under a kind of practical application scene；
Fig. 2 is a kind of flow diagram for air control system optimization method that this specification embodiment provides；
Fig. 3 is a kind of flow diagram for Bayesian Optimization Algorithm that this specification embodiment provides；
Fig. 4 is a kind of structural schematic diagram for air control system optimization system that this specification embodiment provides；
Fig. 5 be this specification embodiment provide in a kind of specific embodiment, in Fig. 4 system detailed construction signal
Figure.
Specific embodiment
This specification embodiment provides air control system optimization method, system, device and electronic equipment.
In order to make those skilled in the art more fully understand the technical solution in this specification, below in conjunction with this explanation
Attached drawing in book embodiment is clearly and completely described the technical solution in this specification embodiment, it is clear that described
Embodiment be merely a part but not all of the embodiments of the present application.Based on this specification embodiment, this field
Those of ordinary skill's every other embodiment obtained without creative efforts, all should belong to the application
The range of protection.
Fig. 1 is the general frame schematic diagram that the scheme of this specification is related under a kind of practical application scene.General frame
(a) equipment, the optimization system place equipment for the air control system where specifically including that air control system.In practical applications, wind
Control system and optimization system may also be in same equipment, then can use general frame (b) in this case.
The workflow of above-mentioned overall architecture may include: that optimization system carries out performance monitoring to air control system, according to property
Energy monitoring data, the multiple machine learning models and/or policing rule progress parameter optimization for globally including to air control system.
Based on above-mentioned overall architecture, the scheme of this specification is described in detail below.
This specification embodiment provides a kind of air control system optimization method, as shown in Fig. 2, Fig. 2 is that the air control system is excellent
The flow diagram of change method, the process may comprise steps of:
S202: the data by air control event handled according to air control system are calculated for evaluating the air control system performance
Operational indicator.
In this specification embodiment, online transaction, Third-party payment such as may is that by air control event, believe from media
The events such as breath publication.
It may is that by the data of air control event to by the air control result or results of intermediate calculations of air control event, for example, wind
Control system gives the score calculated by air control event, and the score such as can reflect the possibility by air control event for risk case
Property；It is also possible to by the data of air control event: required mode input data when air control system is to by the progress air control of air control event,
For example, by the managing detailed catalogue of air control event itself, the user behavior characteristics or environmental information that are related to by air control event.
It can be precipitated to obtain or monitor in real time to obtain by historical data by the data of air control event.
To be the score by the data of air control event, air control system is for identification for risk case.It is assumed that score
Value range is 0 to 1, decision threshold 0.5, when the score is greater than 0.5, determines that corresponding by air control event is risk thing
Part determines that this by air control event is not risk case when being not more than 0.5 by the data of air control event.
Air control system performance can reflect the credibility or reliability of air control system.It can pass through institute in step S202
The operational indicator stated is evaluated.
The operational indicator can have one or more.It uses the example above and is illustrated, the operational indicator such as can be with
It is: risk case concentration (be correctly identified as risk case by air control event number/by air control total number of events amount), risk thing
Part accounting (be correctly identified as risk case by air control event number/be identified as risk case by air control event number
Amount), risk case accuracy of identification or strategy check the rate of bothering etc..
S204: according to the operational indicator, performance monitoring is carried out to the air control system.
Performance monitoring can be carried out using offline or online mode.It is assumed that the operational indicator for needing performance monitoring to be related to compared with
It is more, if then on-line monitoring may expend more system resource, so as to influence the normal operation of air control system, therefore,
In this case it is preferable to which performance monitoring can be carried out using offline mode.
S206: according to the performance monitoring data, the multiple machine learning models and/or plan for including to the air control system
Slightly rule carries out parameter optimization, wherein optimised each parameter is default between the performance monitoring data according to it
What corresponding relationship determined.
In this specification embodiment, air control system includes multiple machine learning models and/or policing rule.
For example, one is used to carry out online transaction event the air control system of air control, behavior decision model, friendship may be included
The two machine learning models of easy decision model, wherein behavior decision model is used for the use for determining to initiate the online transaction event
Whether the behavior at family meets its previous habit, and then determines whether the user is me；Decision model of trading is used for Behavior-based control
The judgement of decision model is as a result, further determine whether the online transaction event is legal.
Policing rule often can be used for auxiliary machinery learning model.For example, can be provided by policing rule: being directed to wind
Control system is when handling specifically by air control event, those specifically used machine learning models；For another example, it can be advised by strategy
It then provides: the output result of certain machine learning models is analyzed, filter out an output result or calculate one
Synthesis result, to the input as another machine learning model；Etc..
From the example above as can be seen that each machine learning model for including in air control system and each policing rule may
With certain relevance.In view of this relevance can will have at least the portion of relevance when carrying out air control model optimization
Point machine learning model and/or policing rule are considered as an entirety, globally parameter optimization are integrally carried out to this, to be conducive to
Obtain better effect of optimization.
Since each machine learning model or policing rule may include multiple optimizable parameters, every
When primary progress parameter optimization, multiple parameters from different machines learning model or policing rule may be optimized simultaneously.
It is described according to the operational indicator for step S204 in this specification embodiment, to the air control system into
Row performance monitoring, can specifically include: monitor whether the operational indicator meets pre-set level threshold value；Refer to according to the business
The data that mark monitoring obtains, determine the performance monitoring data to the air control system.Wherein, pre-set level threshold value is for judging industry
Whether business index meets expection.
It in practical applications, can be directly using the data obtained to operational indicator monitoring as performance monitoring data；It can also
Performance monitoring data is further calculated according to the data obtained to the monitoring of one or more operational indicator, for example,
The data obtained to the monitoring of each operational indicator are measured with score value respectively, then, after being weighted to each score value that measurement obtains,
It is mapped in " performance is qualified " or " performance is unqualified " both results, which can be used as performance monitoring data.
It is described according to the performance monitoring data for step S206 in this specification embodiment, to the air control system
The multiple machine learning models and/or policing rule that system includes carry out parameter optimization, can specifically include: when the performance monitoring
When data do not meet default capabilities threshold value, triggering is directed to the multiple machine learning models and/or strategy that the air control system includes
The parameter optimisation procedure of rule executes；
The parameter optimisation procedure may include: to calculate multiple parameters to be optimized using scheduled optimization algorithm
Optimization after value, value after the optimization is used in the corresponding machine learning model and/or policing rule.
For example it is assumed that operational indicator is risk case accuracy of identification, risk case accuracy of identification will directly be monitored and be obtained
Data as performance monitoring data, if the corresponding pre-set level threshold value of risk case accuracy of identification is 80%, preset property
Energy threshold value correspondingly may be 80%.So, when monitoring risk case accuracy of identification no more than 80%, ginseng can be triggered
Number optimization process executes.
If performance monitoring data is calculated according to the data obtained to the monitoring of one or more operational indicator.Then property
Energy threshold value can also be independent of pre-set level threshold value, alternatively, can be accordingly based upon one or more of operational indicators
Corresponding pre-set level threshold value presets performance threshold.
Further, in practical applications, parameter optimization mistake can also be carried out independent of the triggering of performance monitoring data
Cheng Zhihang, for example, can be executed with clocked flip parameter optimisation procedure.It should be noted that when by the way of timing optimization,
Optimised multiple parameters can be independent of performance monitoring data and determination, and there are many methods of determination, can refer in advance
It is fixed or selected etc. at random.
In this specification embodiment, value can be the process of an iteration after calculation optimization.For example, can be by business
Index is iterated operation as the objective function of optimization algorithm, and optimal objective function and its correspondence are iterated to calculate in each round
Parameter to be optimized value, when meeting scheduled termination condition, terminate iteration, obtain the excellent of each parameter to be optimized
Value after change.
In this specification embodiment, the Optimization Algorithms Library comprising one or more kinds of optimization algorithms can be constructed in advance,
And then parameter optimisation procedure can be executed based on the optimization algorithm in the Optimization Algorithms Library.It is possible to further pass through centainly
Function logic is configured and is safeguarded to Optimization Algorithms Library, for example, for the specified optimization algorithm to match of performance monitoring data, increasing
Add deduct few optimization algorithm, the existing optimization algorithm of upgrading, improves existing optimization algorithm and obtains new algorithm etc..
Each machine learning model or policing rule can directly or indirectly influence air control system performance in air control system, if
Individually to the carry out parameter optimization of some machine learning model, it may not be able to ascend air control service system, or even there are also reduction property
Energy.Based on such consideration, this specification embodiment globally to the multiple machine learning models for including in air control system and/or
Policing rule carries out parameter iteration optimization.
For example it is assumed that air control system includes machine learning model A, machine learning model B and policing rule C, for commenting
The operational indicator of valence air control system performance is d.If it is undesirable to monitor d, trigger parameter optimization process is executed.
Make a reservation for wherein 3 parameter a1, a2 and a3, wherein 2 the parameters b1 and b2 of B that parameter to be optimized includes: A, C's
Two of them parameter c1 and c2.It then can be above-mentioned by executing using a1, a2, a3, b1, b2, c1, c2 as a parameter sets
Parameter optimisation procedure globally optimizes all parameters in the parameter sets.
In this specification embodiment, scheduled optimization algorithm can be Bayesian Optimization Algorithm, Monte carlo algorithm or
The algorithm etc. that person is obtained based on both algorithm improvements.Certainly, these types of algorithm is exemplary, in practical applications, can
Globally other optimization algorithms that multiple parameters optimize may also be suitable for the scheme of this specification.
Further, described to utilize scheduled optimization algorithm by taking Bayesian Optimization Algorithm as an example, it calculates to be optimized multiple
Value after the optimization of the parameter, can specifically include:
Obtain the sample set of multiple sample points comprising having sampled and having observed, wherein i-th of sample point is counted as
{xi,f(xi), xiIndicate the variable being made of multiple parameters to be optimized in the value of i-th of sample point
(it is denoted as the value of variable x), f (xi) indicate to correspond to xiObjective function value, the objective function is for estimating the wind
The performance monitoring data of control system；Based on xiIt may be implemented simultaneously to optimize the multiple parameters of air control system；
According to the sample set, using objective function described in Bayesian Optimization Algorithm iteration optimization；
The value for calculating the good corresponding multiple parameters to be optimized of the objective function of iteration optimization, as
Value after the optimization of multiple parameters to be optimized.
Wherein, the variable can preferably be vector, and multiple parameters to be optimized correspond respectively to described
It is at least one-dimensional in vector.
In order to make it easy to understand, following concrete example brief description globally carries out parameter optimization using Bayesian Optimization Algorithm
Process.
Illustrate by taking one-dimensional variable as an example, it is assumed that f is objective function, f (xi) it is that variable x takes xiWhen the objective function that observes
Value, P (f) is the prior distribution of objective function f.
Assuming that we are sampled and observed t sample point, then sample set is denoted as D1:t={ x1:t,f
(x1:t), the Posterior distrbutionp of corresponding objective function f can be write as:
P(f|D1:t)∝P(D1:t|f)P(f)；
In every wheel interative computation, an evaluation function can be constructed according to the Posterior distrbutionp of objective function f, for selecting most
The value of the value of excellent x and its corresponding f, as newly-increased sample point.
For more detailed description algorithm flow, this specification embodiment additionally provides the signal of Bayesian Optimization Algorithm process
Figure, as shown in figure 3, detailed process includes:
Wherein, the pseudocode of Bayesian Optimization Algorithm can be described as follows:
For the description of algorithm above, specific execution process is as follows:
Step 1: t sample point of initialization.T difference of the above-mentioned variable for randomly choosing or being specified according to user takes
Value, is counted as x1:t；Wherein, xiIndicate above-mentioned variable in the value of i-th of sample point, f (xi) indicate to correspond to xiObjective function
Value, which such as can be used for estimating the performance monitoring data and/or operational indicator of air control system, by xiAnd f
(xi) the corresponding original training set conjunction D of composition1:t={ x1:t,f(x1:t)}。
Step 2: updating sample set.Sample set includes all sample points for having sampled and having observed, when having initialized
Cheng Shi, sample set are that original training set is closed；After each round interative computation, by newly-increased sample point { xt+1,f(xt+1) add
It is added in sample set, obtains updated sample set.
Step 3: judging whether to meet stop condition.Stop condition can be customized；For example, stop condition may is that into
The number of row iteration operation has reached customized maximum number of iterations；Alternatively, stop condition may is that current goal function exists
Maximum value f_max is not further added by sample set.
Step 4: if meeting stop condition, exporting the value of f_max and its corresponding x, process terminates.
Step 5: if being unsatisfactory for stop condition, using sample set D1:t={ x1:t,f(x1:t) fitted Gaussian process GP
(m (x), k (x, x ')), wherein m (x) is mean vector, and k (x, x ') is covariance function.
Step 6: the Posterior distrbutionp of calculating target function, i.e., the value of x corresponding to sample point each in sample set, meter
The distribution of calculation condition
Step 7: according to the Posterior distrbutionp of objective function, Calculation Estimation function u (x | D), calculate so that evaluation function takes most
The value of corresponding x, next observation point x as objective function when big value (or minimum value)t+1(that is, newly-increased sample point
X value)；Corresponding expression formula are as follows:
xt+1=argmaxxu(x|D)；
Wherein, evaluation function such as may is that UCB (x)=μ (x)+κ σ (x).
Step 8: by newly-increased sample point { xt+1,f(xt+1) it is back to step 2 update sample set.
Based on same thinking, this specification embodiment also provides a kind of air control system optimization system, as shown in figure 4, tool
Body may include:
Performance evaluation module 401 is calculated according to the data by air control event that air control system is handled for evaluating the wind
Control the operational indicator of system performance；
Performance monitoring module 402 carries out performance monitoring to the air control system according to the operational indicator；
Performance optimization module 403, according to the performance monitoring data, the multiple machine learning for including to the air control system
Model and/or policing rule carry out parameter optimization, wherein optimised each parameter is according to itself and the performance monitoring number
According to default corresponding relationship determine.
In this specification embodiment, the performance monitoring module includes performance monitoring submodule；Performance monitoring
Module carries out performance monitoring according to the operational indicator, to the air control system, can specifically include:
The performance monitoring submodule 412 monitors whether the operational indicator meets pre-set level threshold value；
According to the data obtained to operational indicator monitoring, the performance monitoring data to the air control system is determined.
In this specification embodiment, air control system risk case for identification；The operational indicator includes following
At least one: risk case concentration, risk case accounting, risk case accuracy of identification, strategy check the rate of bothering.
In this specification embodiment, the performance monitoring module 402 is also held comprising optimization triggering submodule 422 and optimization
Row submodule 423；The performance optimization module is according to the performance monitoring data, the multiple machines for including to the air control system
Learning model and/or policing rule carry out parameter optimization, can specifically include:
Optimization triggering submodule is not when the performance monitoring data meets default capabilities threshold value, by described excellent
Change the parameter optimization of multiple machine learning models and/or policing rule that triggering submodule triggering includes for the air control system
Process executes；
The parameter optimisation procedure includes:, using scheduled optimization algorithm, to be calculated to excellent by the performance optimization module
Value after the optimization for the multiple parameters changed；By the optimization implementation sub-module, by value after the optimization corresponding
It is used in the machine learning model and/or policing rule.
Further, the scheduled optimization algorithm includes: Bayesian Optimization Algorithm or Monte carlo algorithm.
Further, described to be calculated using scheduled optimization when the scheduled optimization algorithm is Bayesian Optimization Algorithm
Method calculates value after the optimization of multiple parameters to be optimized, can specifically include:
The performance optimization module 403 obtains the sample set of multiple sample points comprising having sampled and having observed, wherein
I-th of sample point is counted as { xi,f(xi), xiIt indicates by multiple parameters to be optimized in i-th of sample
The value for the variable that the value of point is constituted, f (xi) indicate to correspond to xiObjective function value, the objective function is for estimating
Count the performance monitoring data and/or operational indicator of the air control system；
According to the sample set, using objective function described in Bayesian Optimization Algorithm iteration optimization；
The value for calculating the good corresponding multiple parameters to be optimized of the objective function of iteration optimization, as
Value after the optimization of multiple parameters to be optimized.
Further, the variable is vector, and multiple parameters to be optimized correspond respectively in the vector
It is at least one-dimensional.
In this specification embodiment, the multiple machine learning models for including to the air control system and/or strategy
Rule carries out the mode of parameter optimization, further includes: the multiple machine learning models and/or plan for periodically including to the air control system
Slightly rule carries out parameter optimization.
This specification embodiment also provides the application example of an air control system optimization, and Fig. 5 is that this specification embodiment mentions
Supply in a kind of specific embodiment, the detailed construction of system in Fig. 4 specifically includes that data Layer, system are excellent in the schematic diagram
Change layer and application layer.
In data Layer, including historical data backtracking module and data pick-up conversion load (Extract-Transform-
Load, ETL) unit/interface.Precipitating has each historical data received (that is, by air control event in historical data backtracking module
Data), for use in evaluation assignment index.After historical data needed for obtaining, historical data is taken out using ETL process unit
The operation such as sample, integration, is sent to corresponding business according to field by ETL process interface for historical data after treatment and refers to
Computational submodule is marked, operational indicator is calculated.
In system optimization layer, including performance evaluation module and performance optimization module.It wherein, include industry in performance evaluation module
Index computational submodule of being engaged in and data sub-module stored；The data sub-module stored is based on storage service index computational submodule
Obtained operational indicator, in order to which other modules can call operational indicator at any time.It include optimization control in performance optimization module
Device and Optimization Algorithms Library processed；It include various optimization algorithms in Optimization Algorithms Library, for example, Bayesian Optimization Algorithm, Monte Carlo
Markov algorithm etc.；Optimal controller is for configuring, upgrading or refined Hook Jeeves algorighm.Operational indicator is in output to performance monitoring
While module, also exports and give performance optimization module.The effect of performance optimization module is to the machine learning mould for having triggered optimization
Type or policing rule carry out parameter optimization.
In application layer, all kinds of machine learning models and/or policing rule of commencement of commercial operation on line are contained, it is whole or sexual
The machine learning model and policing rule that demand can be monitored are linked into the performance monitoring module.Wherein, machine learning model and/or
The mode that policing rule is connect with the performance monitoring module can be offline connection, be also possible to on-line joining process.Performance monitoring mould
Block mainly have of both task: one side task is, on demand (as daily or hourly) on line machine learning model or
The performance of policing rule is monitored and feeds back to external (such as personnel monitoring's equipment)；Another aspect task is executed to need
The optimization of the machine learning model or policing rule to be optimized, such as the parameter of optimization machine learning model.Further, herein
Performance monitoring module can be the monitoring part of broad sense, in addition to comprising for monitoring machine learning model and policing rule on line
Performance monitoring submodule outside, further comprise optimization triggering submodule and optimization implementation sub-module.Optimization triggering submodule
Trigger condition can trigger optimization instruction according to the result of monitoring, can also be optimized according to timing function clocked flip and be operated.Optimization
Implementation sub-module is then to adjust the parameter of each machine learning model or policing rule according to the parameter after optimization.
Based on same thinking, this specification embodiment additionally provides a kind of air control system optimization apparatus, comprising:
Computing module is calculated according to the data by air control event that air control system is handled for evaluating the air control system
The operational indicator of performance；
Monitoring module carries out performance monitoring to the air control system according to the operational indicator；
Optimization module, according to the performance monitoring data, multiple machine learning models for include to the air control system and/
Or policing rule carry out parameter optimization, wherein optimised each parameter be according to it between the performance monitoring data
What default corresponding relationship determined.
Based on same thinking, this specification embodiment additionally provides a kind of electronic equipment, comprising:
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes so that at least one described processor can:
According to the data by air control event of air control system, corresponding operational indicator is calculated；
The business output of air control system is obtained as a result, commenting according to the operational indicator business output result
Estimate；
According to the business output for completing assessment as a result, using predefined optimization algorithm, calculates and obtain the air control
Optimal Parameters in system.
It is above-mentioned that this specification specific embodiment is described.Other embodiments are in the scope of the appended claims
It is interior.In some cases, the movement recorded in detail in the claims or step can be come according to the sequence being different from embodiment
It executes and desired result still may be implemented.In addition, process depicted in the drawing not necessarily require show it is specific suitable
Sequence or consecutive order are just able to achieve desired result.In some embodiments, multitasking and parallel processing be also can
With or may be advantageous.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device,
For electronic equipment, nonvolatile computer storage media embodiment, since it is substantially similar to the method embodiment, so description
It is fairly simple, the relevent part can refer to the partial explaination of embodiments of method.
This specification embodiment provide device, system, electronic equipment, nonvolatile computer storage media and method be
Corresponding, therefore, device, system, electronic equipment, nonvolatile computer storage media also have and similar with corresponding method have
Beneficial technical effect, since the advantageous effects of method being described in detail above, it is right which is not described herein again
Answer the advantageous effects of device, system, electronic equipment, nonvolatile computer storage media.
In the 1990s, the improvement of a technology can be distinguished clearly be on hardware improvement (for example,
Improvement to circuit structures such as diode, transistor, switches) or software on improvement (improvement for method flow).So
And with the development of technology, the improvement of current many method flows can be considered as directly improving for hardware circuit.
Designer nearly all obtains corresponding hardware circuit by the way that improved method flow to be programmed into hardware circuit.Cause
This, it cannot be said that the improvement of a method flow cannot be realized with hardware entities module.For example, programmable logic device
(Programmable Logic Device, PLD) (such as field programmable gate array (Field Programmable Gate
Array, FPGA)) it is exactly such a integrated circuit, logic function determines device programming by user.By designer
Voluntarily programming comes a digital display circuit " integrated " on a piece of PLD, designs and makes without asking chip maker
Dedicated IC chip.Moreover, nowadays, substitution manually makes IC chip, this programming is also used instead mostly " is patrolled
Volume compiler (logic compiler) " software realizes that software compiler used is similar when it writes with program development,
And the source code before compiling also write by handy specific programming language, this is referred to as hardware description language
(Hardware Description Language, HDL), and HDL is also not only a kind of, but there are many kind, such as ABEL
(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description
Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL
(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby
Hardware Description Language) etc., VHDL (Very-High-Speed is most generally used at present
Integrated Circuit Hardware Description Language) and Verilog.Those skilled in the art also answer
This understands, it is only necessary to method flow slightly programming in logic and is programmed into integrated circuit with above-mentioned several hardware description languages,
The hardware circuit for realizing the logical method process can be readily available.
Controller can be implemented in any suitable manner, for example, controller can take such as microprocessor or processing
The computer for the computer readable program code (such as software or firmware) that device and storage can be executed by (micro-) processor can
Read medium, logic gate, switch, specific integrated circuit (Application Specific Integrated Circuit,
ASIC), the form of programmable logic controller (PLC) and insertion microcontroller, the example of controller includes but is not limited to following microcontroller
Device: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320 are deposited
Memory controller is also implemented as a part of the control logic of memory.It is also known in the art that in addition to
Pure computer readable program code mode is realized other than controller, can be made completely by the way that method and step is carried out programming in logic
Controller is obtained to come in fact in the form of logic gate, switch, specific integrated circuit, programmable logic controller (PLC) and insertion microcontroller etc.
Existing identical function.Therefore this controller is considered a kind of hardware component, and to including for realizing various in it
The device of function can also be considered as the structure in hardware component.Or even, it can will be regarded for realizing the device of various functions
For either the software module of implementation method can be the structure in hardware component again.
System, device, module or the unit that above-described embodiment illustrates can specifically realize by computer chip or entity,
Or it is realized by the product with certain function.It is a kind of typically to realize that equipment is computer.Specifically, computer for example may be used
Think personal computer, laptop computer, cellular phone, camera phone, smart phone, personal digital assistant, media play
It is any in device, navigation equipment, electronic mail equipment, game console, tablet computer, wearable device or these equipment
The combination of equipment.
For convenience of description, it is divided into various units when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each unit can be realized in the same or multiple software and or hardware when specification one or more embodiment.
It should be understood by those skilled in the art that, this specification embodiment can provide as method, system or computer program
Product.Therefore, this specification embodiment can be used complete hardware embodiment, complete software embodiment or combine software and hardware
The form of the embodiment of aspect.Moreover, it wherein includes that computer is available that this specification embodiment, which can be used in one or more,
It is real in the computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) of program code
The form for the computer program product applied.
This specification is referring to the method, equipment (system) and computer program product according to this specification embodiment
Flowchart and/or the block diagram describes.It should be understood that can be realized by computer program instructions every in flowchart and/or the block diagram
The combination of process and/or box in one process and/or box and flowchart and/or the block diagram.It can provide these computers
Processor of the program instruction to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
To generate a machine, so that generating use by the instruction that computer or the processor of other programmable data processing devices execute
In the dress for realizing the function of specifying in one or more flows of the flowchart and/or one or more blocks of the block diagram
It sets.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
This specification can describe in the general context of computer-executable instructions executed by a computer, such as journey
Sequence module.Generally, program module include routines performing specific tasks or implementing specific abstract data types, programs, objects,
Component, data structure etc..Specification can also be practiced in a distributed computing environment, in these distributed computing environments,
By executing task by the connected remote processing devices of communication network.In a distributed computing environment, program module can
To be located in the local and remote computer storage media including storage equipment.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
The foregoing is merely this specification embodiments, are not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.
Claims (18)
1. a kind of air control system optimization method, comprising:
According to the data by air control event that air control system is handled, calculating refers to for evaluating the business of the air control system performance
Mark；
According to the operational indicator, performance monitoring is carried out to the air control system；
According to the performance monitoring data, multiple machine learning models for include to the air control system and/or policing rule into
Row parameter optimization, wherein optimised each parameter is the default corresponding relationship according to it between the performance monitoring data
Determining.
2. the method as described in claim 1, described according to the operational indicator, performance monitoring is carried out to the air control system,
It specifically includes:
Monitor whether the operational indicator meets pre-set level threshold value；
According to the data obtained to operational indicator monitoring, the performance monitoring data to the air control system is determined.
3. method according to claim 2, air control system risk case for identification；The operational indicator includes following
At least one: risk case concentration, risk case accounting, risk case accuracy of identification, strategy check the rate of bothering.
It is described according to the performance monitoring data 4. the method as described in claim 1, to the air control system include it is multiple
Machine learning model and/or policing rule carry out parameter optimization, specifically include:
When the performance monitoring data does not meet default capabilities threshold value, triggering is directed to multiple machines that the air control system includes
The parameter optimisation procedure of learning model and/or policing rule executes；
The parameter optimisation procedure includes: using scheduled optimization algorithm, after the optimization for calculating multiple parameters to be optimized
Value uses value after the optimization in the corresponding machine learning model and/or policing rule.
5. method as claimed in claim 4, the scheduled optimization algorithm includes: Bayesian Optimization Algorithm or Meng Teka
Lip river algorithm.
6. method as claimed in claim 4, described using pre- when the scheduled optimization algorithm is Bayesian Optimization Algorithm
Fixed optimization algorithm calculates value after the optimization of multiple parameters to be optimized, specifically includes:
Obtain the sample set of multiple sample points comprising having sampled and having observed, wherein i-th of sample point is counted as { xi,f
(xi), xiIndicate the value for the variable being made of multiple parameters to be optimized in the value of i-th of sample point,
f(xi) indicate to correspond to xiObjective function value, the objective function is used to estimate the performance monitoring number of the air control system
According to and/or operational indicator；
According to the sample set, using objective function described in Bayesian Optimization Algorithm iteration optimization；
The value for calculating the good corresponding multiple parameters to be optimized of the objective function of iteration optimization, as described
Value after the optimization of multiple parameters to be optimized.
7. method as claimed in claim 6, the variable is vector, multiple parameters to be optimized are corresponded respectively to
It is at least one-dimensional in the vector.
8. the method as described in claim 1, the method also includes: the multiple engineerings for periodically including to the air control system
It practises model and/or policing rule carries out parameter optimization.
9. a kind of air control system optimization system, comprising:
Performance evaluation module is calculated according to the data by air control event that air control system is handled for evaluating the air control system
The operational indicator of performance；
Performance monitoring module carries out performance monitoring to the air control system according to the operational indicator；
Performance optimization module, according to the performance monitoring data, multiple machine learning models for include to the air control system and/
Or policing rule carry out parameter optimization, wherein optimised each parameter be according to it between the performance monitoring data
What default corresponding relationship determined.
10. system as claimed in claim 9, the performance monitoring module includes performance monitoring submodule；The performance monitoring
Module carries out performance monitoring according to the operational indicator, to the air control system, specifically includes:
The performance monitoring submodule monitors whether the operational indicator meets pre-set level threshold value；
According to the data obtained to operational indicator monitoring, the performance monitoring data to the air control system is determined.
11. system as claimed in claim 10, air control system risk case for identification；The operational indicator include with
Lower at least one: risk case concentration, risk case accounting, risk case accuracy of identification, strategy check the rate of bothering.
12. system as claimed in claim 9, the performance monitoring module also includes that optimization triggering submodule and optimization execute son
Module；The performance optimization module is according to the performance monitoring data, the multiple machine learning moulds for including to the air control system
Type and/or policing rule carry out parameter optimization, specifically include:
For the optimization triggering submodule when the performance monitoring data does not meet default capabilities threshold value, triggering is directed to the wind
The parameter optimisation procedure for the multiple machine learning models and/or policing rule that control system includes executes；
The parameter optimisation procedure includes: the performance optimization module using scheduled optimization algorithm, is calculated to be optimized multiple
Value after the optimization of the parameter, the optimization implementation sub-module is by value after the optimization in the corresponding machine learning mould
It is used in type and/or policing rule.
13. system as claimed in claim 12, the scheduled optimization algorithm includes: Bayesian Optimization Algorithm or covers special
Carlow algorithm.
14. system as claimed in claim 12, when the scheduled optimization algorithm is Bayesian Optimization Algorithm, the utilization
Scheduled optimization algorithm calculates value after the optimization of multiple parameters to be optimized, specifically includes:
The performance optimization module obtains the sample set of multiple sample points comprising having sampled and having observed, wherein i-th of institute
It states sample point and is counted as { xi,f(xi), xiIndicate by multiple parameters to be optimized i-th of sample point value
The value of the variable of composition, f (xi) indicate to correspond to xiObjective function value, the objective function is for estimating the wind
The performance monitoring data and/or operational indicator of control system；
According to the sample set, using objective function described in Bayesian Optimization Algorithm iteration optimization；
The value for calculating the good corresponding multiple parameters to be optimized of the objective function of iteration optimization, as described
Value after the optimization of multiple parameters to be optimized.
15. system as claimed in claim 14, the variable is vector, and multiple parameters to be optimized respectively correspond
It is at least one-dimensional in the vector.
16. system as claimed in claim 9, the performance optimization module also: periodically to the air control system include it is multiple
Machine learning model and/or policing rule carry out parameter optimization.
17. a kind of air control system optimization apparatus, comprising:
Computing module is calculated according to the data by air control event that air control system is handled for evaluating the air control system performance
Operational indicator；
Monitoring module carries out performance monitoring to the air control system according to the operational indicator；
Optimization module, according to the performance monitoring data, the multiple machine learning models and/or plan for including to the air control system
Slightly rule carries out parameter optimization, wherein optimised each parameter is default between the performance monitoring data according to it
What corresponding relationship determined.
18. a kind of electronic equipment, comprising:
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
Manage device execute so that at least one described processor can:
According to the data by air control event that air control system is handled, calculating refers to for evaluating the business of the air control system performance
Mark；
According to the operational indicator, performance monitoring is carried out to the air control system；
According to the performance monitoring data, multiple machine learning models for include to the air control system and/or policing rule into
Row parameter optimization, wherein optimised each parameter is the default corresponding relationship according to it between the performance monitoring data
Determining.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710768969.7A CN109426912B (en) | 2017-08-31 | 2017-08-31 | Wind control system optimization method, system and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710768969.7A CN109426912B (en) | 2017-08-31 | 2017-08-31 | Wind control system optimization method, system and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109426912A true CN109426912A (en) | 2019-03-05 |
| CN109426912B CN109426912B (en) | 2022-07-26 |
Family
ID=65505306
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710768969.7A Active CN109426912B (en) | 2017-08-31 | 2017-08-31 | Wind control system optimization method, system and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109426912B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110322153A (en) * | 2019-07-09 | 2019-10-11 | 中国工商银行股份有限公司 | Monitor event processing method and system |
| CN110378580A (en) * | 2019-07-03 | 2019-10-25 | 国家电网有限公司技术学院分公司 | A kind of electric network fault multi-agent system preferentially diagnostic method and device |
| CN111353850A (en) * | 2020-02-26 | 2020-06-30 | 支付宝(杭州)信息技术有限公司 | Risk identification strategy updating method and device and risk merchant identification method and device |
| CN112184235A (en) * | 2020-09-04 | 2021-01-05 | 支付宝(杭州)信息技术有限公司 | Wind control data changing method and device |
| CN113672940A (en) * | 2021-08-23 | 2021-11-19 | 京东科技信息技术有限公司 | Service data monitoring method, system, device and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140096519A (en) * | 2013-01-28 | 2014-08-06 | 네이버 주식회사 | Apparatus, method and computer readable recording medium of monitoring and preventing a false operation for a system optimization function |
| CN104765916A (en) * | 2015-03-31 | 2015-07-08 | 西南交通大学 | Dynamics performance parameter optimizing method of high-speed train |
| CN105654102A (en) * | 2014-11-10 | 2016-06-08 | 富士通株式会社 | Data processing device and data processing method |
| CN105989441A (en) * | 2015-02-11 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Model parameter adjustment method and device |
| CN107067157A (en) * | 2017-03-01 | 2017-08-18 | 北京奇艺世纪科技有限公司 | Business risk appraisal procedure, device and air control system |
-
2017
- 2017-08-31 CN CN201710768969.7A patent/CN109426912B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140096519A (en) * | 2013-01-28 | 2014-08-06 | 네이버 주식회사 | Apparatus, method and computer readable recording medium of monitoring and preventing a false operation for a system optimization function |
| CN105654102A (en) * | 2014-11-10 | 2016-06-08 | 富士通株式会社 | Data processing device and data processing method |
| CN105989441A (en) * | 2015-02-11 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Model parameter adjustment method and device |
| CN104765916A (en) * | 2015-03-31 | 2015-07-08 | 西南交通大学 | Dynamics performance parameter optimizing method of high-speed train |
| CN107067157A (en) * | 2017-03-01 | 2017-08-18 | 北京奇艺世纪科技有限公司 | Business risk appraisal procedure, device and air control system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110378580A (en) * | 2019-07-03 | 2019-10-25 | 国家电网有限公司技术学院分公司 | A kind of electric network fault multi-agent system preferentially diagnostic method and device |
| CN110322153A (en) * | 2019-07-09 | 2019-10-11 | 中国工商银行股份有限公司 | Monitor event processing method and system |
| CN111353850A (en) * | 2020-02-26 | 2020-06-30 | 支付宝(杭州)信息技术有限公司 | Risk identification strategy updating method and device and risk merchant identification method and device |
| CN112184235A (en) * | 2020-09-04 | 2021-01-05 | 支付宝(杭州)信息技术有限公司 | Wind control data changing method and device |
| CN113672940A (en) * | 2021-08-23 | 2021-11-19 | 京东科技信息技术有限公司 | Service data monitoring method, system, device and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109426912B (en) | 2022-07-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109426912A (en) | Air control system optimization method, system, device and electronic equipment | |
| CN107066365A (en) | The monitoring method and device of a kind of system exception | |
| CN110278175A (en) | Graph structure model training, the recognition methods of rubbish account, device and equipment | |
| US11222262B2 (en) | Non-Markovian control with gated end-to-end memory policy networks | |
| CN108460523A (en) | A kind of air control rule generating method and device | |
| CN110245047B (en) | Time sequence abnormality detection method, device and equipment | |
| CN109002949A (en) | A kind of method and device of air control strategy configuration and business air control | |
| CN111045814B (en) | Resource scheduling method and terminal equipment | |
| CN104346926B (en) | Running time Forecasting Methodology and device and related terminal device | |
| CN108038040A (en) | Computer cluster performance indicator detection method, electronic equipment and storage medium | |
| CN108665143A (en) | The appraisal procedure and device of air control model | |
| JP6869347B2 (en) | Risk control event automatic processing method and equipment | |
| CN109657890A (en) | A kind of risk for fraud of transferring accounts determines method and device | |
| CN109784599A (en) | A kind of method, device and equipment of model training, risk identification | |
| CN109670784A (en) | A kind of method, apparatus and system for informing the waiting time | |
| CN109003091A (en) | A kind of risk prevention system processing method, device and equipment | |
| CN110378400A (en) | A kind of model training method and device for image recognition | |
| CN110287229A (en) | A kind of data statistics processing method and device | |
| CN108460490A (en) | A kind of prediction technique, device and the equipment of business occurrence quantity | |
| CN110766184A (en) | Order quantity prediction method and device | |
| CN109584431A (en) | A kind of data processing method of priority queue, apparatus and system | |
| CN107038127A (en) | Application system and its buffer control method and device | |
| CN110245973A (en) | Data processing method and device | |
| CN104052625B (en) | A kind of energy-efficient deployment method and apparatus of cluster server | |
| CN116823521B (en) | Intelligent secondary water supply terminal control system based on AI algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201019 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20201019 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |