Self-recovery method and system for intelligent converged terminal software gray release

Abstract

The application discloses a kind of intelligent fusion terminal software grey release self-healing method and system, the method includes in software library according to real-time selection probability selects software program as target software program, after the software program is selected as target software program, according to preset index, the performance of the network of district area using target software program is evaluated, and the real-time selection probability of corresponding software program in software library is adjusted according to the result of performance evaluation.The method and system are directed to the application characteristics of intelligent fusion terminal software, new and old software programs are summarized into software library, and the grey release and self-healing function of new and old software are automatically realized using specially designed dynamic probability adjustment method, which has good practicability in practical application.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power communication, in particular to a self-healing method and system for intelligent fusion terminal software gray release. BACKGROUND

[0002] The intelligent fusion terminal is a new type of substation monitoring device, which has multiple functions and can be expanded according to needs. Correspondingly, with the development of technology, the programs corresponding to each function are also updated gradually.

[0003] For example, one of the functions of the intelligent fusion terminal is to obtain the topology of the substation and detect the equipment or lines in the substation by checking the topology in real time.

[0004] The principle of obtaining the topology of the substation is that the intelligent sensing units such as electric energy meters and intelligent switches in the substation send signals in a specific way, the intelligent fusion terminal accessed in the substation captures and decodes the signals, and finally calculates the position information and state information of each intelligent sensing unit in the substation through an algorithm. In this process, the form of current signal sent by the intelligent sensing unit (signal transmission system, signal transmission method, etc.), the capture and decoding method of the intelligent fusion terminal for the current signal, and the final topology calculation algorithm are realized through a specially designed topology monitoring program.

[0005] The specific performance of the topology monitoring program is reflected in the response rate, etc. Due to the complexity of the substation structure, it is difficult for the topology monitoring program to completely judge its specific performance through simulation, and the topology monitoring program needs to be actually applied to the substation and verified for a certain period of time to completely test its specific performance. In this process, the problem of updating method of the topology monitoring program is involved.

[0006] Unlike the updating of conventional software, if the updating method is replaced, there is a great risk of directly replacing the topology monitoring program before verifying the performance of the new topology monitoring program, and the topology monitoring program does not have the function of local device gray test. Therefore, the updating method needs to be designed for programs such as the topology monitoring program in the intelligent fusion terminal. SUMMARY

[0007] The present application provides a self-healing method for intelligent fusion terminal software gray release, which is based on the application characteristics of intelligent fusion terminal software. The new and old software programs are collected into a software library, and the gray release and self-healing functions of the new and old software are automatically realized by using a specially designed dynamic probability adjustment method, which has good practicality in actual application.

[0008] Correspondingly, the present application provides a self-healing method for intelligent fusion terminal software gray release, which comprises:

[0009] Software library Comprising A software program, each of the software programs Has a corresponding real-time selection probability , , ;

[0010] The intelligent fusion terminal periodically selects a software program in the software library According to the real-time selection probability As a target software program And broadcasts the information of the target software program In the network, the information of the target software program Used to inform the intelligent sensing unit and the intelligent fusion terminal in the network to run according to the target software program , ;

[0011] The real-time selection probability of each of the software programs Based on the adjustment process, the adjustment process includes:

[0012] When the software program Is added to the software library , an initial real-time selection probability value is given , and the real-time selection probability of the remaining software programs in the software library Is adjusted synchronously;

[0013] After the software program Is selected as a target software program By the intelligent fusion terminal, the intelligent fusion terminal performs performance evaluation on the network applying the target software program According to the preset index;

[0014] According to the results of the performance evaluation, the real-time selection probability of the software program Is adjusted And the real-time selection probability of the remaining software programs in the software library Is adjusted synchronously;

[0015] The real-time selection probability of the software program The value range of . .

[0016] Optional implementation, in the software library The number of copies of each software program Is , the software library Together a software program, each of the software programs a real-time selection probability .

[0017] An optional embodiment, when the software program is added to the software library , an initial real-time selection probability value is given to the software program .

[0018] An optional embodiment, when the software program is added to the software library , the number of copies of the software program is expanded to , and the initial real-time selection probability value of the software program .

[0019] An optional embodiment, the real-time selection probability of the software program is adjusted according to the result of the performance evaluation .

[0020] When the result of the performance evaluation does not meet the preset requirement, the number of copies of the software program in the software library is reduced.

[0021] An optional embodiment, a balance number formula is constructed , wherein, is an integer value greater than 1, is the number of copies of the software program that needs to be reduced in the software library when the result of the performance evaluation of the network of the transformer area does not meet the preset requirement.

[0022] An optional embodiment, the real-time selection probability of the software program is adjusted according to the result of the performance evaluation .

[0023] When the result of the performance evaluation meets the preset requirement, and the cumulative number of times that the software program is selected as the target software program by the intelligent fusion terminal reaches a preset value, the number of copies of the software program in the software library is increased, and the cumulative number of times that the software program is selected as the target software program by the intelligent fusion terminal is cleared.

[0024] ​In an optional implementation, the intelligent fusion terminal applies the target software program according to preset indicators. The performance evaluation of the distribution network includes:

[0025] Based on the software library Evaluation indicators ;

[0026] Build the target software program Formula for calculating the total score ,in, , , For target software program Regarding evaluation indicators The rating values, different evaluation indicators They have the same range of rating values;

[0027] The total score is used as the preset indicator to apply the target software program. The performance of the distribution network was evaluated.

[0028] An optional implementation, in the software library In the context of the real-time selection probability Sort in descending order, excluding the target software program. Then, in order of order The software program's score is used as an evaluation criterion for applying the target software program. The performance of the distribution network was evaluated.

[0029] It is a preset positive integer.

[0030] In an optional implementation, the preset index is for the target software program. Descriptive evaluation metrics.

[0031] Accordingly, the present invention provides a self-healing system for gray-scale release of intelligent converged terminal software, used to implement the aforementioned self-healing method for gray-scale release of intelligent converged terminal software, including a transformer substation network;

[0032] The transformer substation network includes intelligent converged terminal software and intelligent sensing units, with one or more intelligent converged terminal software units and one or more intelligent sensing units respectively arranged in the transformer substation network according to a preset topology.

[0033] In conclusion, the application discloses a self-recovery method and system for intelligent fusion terminal software gray release, realizes automatic operation of a whole process of software program deployment, verification, optimization and elimination, liberates operation and maintenance personnel from heavy version monitoring, decision and rollback operation, reduces labor cost and operation risk, ensures that a power supply area network runs on optimal software most of the time, improves work efficiency of the power supply area network, provides a complete and data-driven autonomous management framework for the whole process of software version introduction, verification, promotion, monitoring, recession and elimination, maximizes the value of software assets, and invalid or harmful versions can be cleaned in time, and an internal competition and survival of the fittest software development environment is generated, and software programs provided by developers or algorithms will continue to compete in a real and dynamic environment, promote the spiral rise of the whole system function and performance, and lay a foundation for building an intelligent power grid software system with long-term vitality and adaptability. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 The self-recovery method for intelligent fusion terminal software gray release is shown in the flowchart of the embodiment of the application.

[0035] Figure 2 The adjustment flowchart is shown in the flowchart of the embodiment of the application.

[0036] Figure 3 The self-recovery system structure diagram for intelligent fusion terminal software gray release is shown in the structure diagram of the embodiment of the application. DETAILED DESCRIPTION

[0037] In order to further illustrate the technical means and effects taken by the application to achieve the predetermined purpose, the specific embodiments, structures, features and effects of the application are described in detail below in combination with the drawings and preferred embodiments. In the following description, different "an embodiment" or "embodiments" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0038] Figure 1 The self-recovery method for intelligent fusion terminal software gray release is shown in the flowchart of the embodiment of the application.

[0039] Specifically, the embodiment of the application provides a self-recovery method for intelligent fusion terminal software gray release, which comprises the following steps.

[0040] S101: software library preparation;

[0041] Software library Comprises A software program, each software program Has a corresponding real-time selection probability , , ;

[0042] Specifically, multiple software programs within the same software library perform the same function. These programs exhibit code differences, and consequently, their performance varies. The intelligent fusion terminal software gray-scale release self-healing method of this invention, in one aspect, uses real-time selection probability to characterize the performance of the software program (the specific characterization method will be further explained later). On the other hand, the real-time selection probability also characterizes the probability of a software program being selected for use (the specific characterization method will be further explained later). The superposition of these two functions achieves a positive functional superposition, meaning that higher-performing software programs are used more frequently, thus realizing the gray-scale release function; lower-performing software programs are automatically eliminated, thus realizing the self-healing recovery function.

[0043] S102: The intelligent converged terminal periodically appears in the software library. A software program is randomly selected according to the real-time selection probability. As the target software program And broadcast the target software program in the network area. Information;

[0044] Specifically, the target software program The information is used to notify the intelligent sensing units and the intelligent fusion terminal in the transformer substation network to follow the target software program. run, .

[0045] Specifically, the intelligent sensing units and intelligent fusion terminals in the distribution network will be used in the next target software program. Before broadcasting any information, the target software program for the current broadcast shall be followed. The information operation, that is, in the target software program For a period of time after the information broadcast, the corresponding target software program was used in the area network. run.

[0046] Steps S101 and S102 are the actual application methods of the intelligent converged terminal software in this embodiment of the invention, in which a smart converged terminal initiates the target software program. The broadcast notifies all relevant devices in the area network to comply with the target software program. Execution, in which the target software program The selection is done through random selection from the software library. Select from the software library Software programs in with corresponding real-time selection probability, in multiple selections, the final statistical selection result distribution tends to software program In software library The real-time selection probability distribution, the gray release self-healing function concerned by the embodiment of the application is realized by dynamically adjusting the real-time selection probability of the software program .

[0047] Specifically, the real-time selection probability of each software program is adjusted based on the adjustment process.

[0048] Figure 2 The adjustment process flowchart of the embodiment of the application.

[0049] The adjustment process includes:

[0050] S201: when the software program is added to the software library , an initial real-time selection probability value is given, and the real-time selection probability of the remaining software programs in the software library is adjusted synchronously;

[0051] Specifically, a new software program needs to be given an initial real-time selection probability value when newly added to the software library . It should be noted that in the same software library, the real-time selection probability value given to each software program when newly added to the software library needs to be fine-tuned, and problems such as probability inflation need to be avoided in the subsequent real-time selection probability value adjustment process, so as to ensure the feasibility of the method.

[0052] Specifically, the embodiment of the application deepens the design of the probability related to the software program in the software library.

[0053] Specifically, the real-time selection probability of the software program in the software library is converted into frequency count.

[0054] Specifically, in the software library , the number of copies of each software program is , the software library includes software programs, and the real-time selection probability of each software program ;

[0055] Correspondingly, when a new software program is added to the software library , the software program​ The number of copies is expanded to One, the software program Initial real-time selection probability value .

[0056] On the one hand, through this implementation method, firstly for the software library In fact, there is no need to further modify the software programs involved. The real-time selection probability is defined, and the number of software programs in the software library can be used to characterize the real-time selection probability of a software program; the software programs in the software library... The selection of target software programs is carried out in a completely random manner, without the need for additional probability distribution calculations. At the same time, since it is necessary to ensure that the sum of the real-time selection probabilities of all software programs in the software library is 1, if a simple probability calculation method is used, in actual statistics, after a new software program is added to the software library, the real-time selection probability of each software program does not need to be actively calculated and adjusted, but exists objectively by frequency counting.

[0057] On the other hand, compared with directly assigning an initial real-time selection probability value, using frequency counting to represent the probability value can better ensure the statistical stability between different software programs in the software library, enabling the software library to have a certain self-regulating ability and avoiding problems such as probability imbalance and probability inflation.

[0058] For example, if multiple software programs are added to the software library in a short period of time, and if an initial real-time selection probability value is directly assigned, each newly added software program will greatly affect the real-time selection probability of other software programs. As a result, the distribution of the real-time selection probability of software programs in the software library is no longer determined by their performance, which affects the normal operation logic of the method in this embodiment of the invention.

[0059] By using frequency counting to calculate the selection probability in real time, all software programs in the software library have a nearly identical starting point for representing performance effects, and newly added software programs will not excessively affect the performance evaluation of other existing software programs.

[0060] S202: In the software program Selected as the target software program by the intelligent fusion terminal Then, the intelligent fusion terminal applies the target software program according to preset indicators. The performance of the distribution network was evaluated.

[0061] In the software program When evaluating the performance in the transformer area network, appropriate preset indexes are selected according to the functions of the software programs to evaluate, so as to distinguish the performance differences between different software programs.

[0062] Specifically, when evaluating the performance of the software programs , the evaluation logic related to the functions of the software programs needs to be defined. Generally, the technical effects of the software programs are divided into two types: quantifiable and unquantifiable.

[0063] For some software programs with quantifiable technical effects, such as topology monitoring programs, the technical effects mainly include transformer area network topology structure reconstruction generation time, fault response speed, and average response time of each node device. These quantifiable technical effects can be accurately represented by numerical values.

[0064] For some software programs with unquantifiable technical effects, such as fault detection programs, descriptive evaluation indexes are needed for functions such as whether to normally identify fault positions, whether to issue repair work orders, and whether the generation of repair work orders is accurate.

[0065] Specifically, for quantifiable technical effects, the intelligent fusion terminal evaluates the performance of the transformer area network to which the target software program is applied according to preset indexes, which includes:

[0066] establishing evaluation indexes according to the software library;

[0067] constructing a score total value calculation formula for the target software program , wherein, , , , is the score value of the target software program in the evaluation index , different evaluation indexes have the same score value range;

[0068] using the score total value as the preset index to evaluate the performance of the transformer area network to which the target software program is applied.

[0069] Specifically, for unquantifiable technical effects, the preset index is a descriptive evaluation index for the target software program .

[0070] S203: Adjusting the real-time selection probability of the software program according to the results of the performance evaluation. Adjusting the real-time selection probability of the software program in the software library in the software library

[0071] in the software library in the software library in the software library .

[0072] Specifically, in this step, the real-time selection probability of the software program is adjusted according to the result of the performance evaluation and the real-time selection probability of the rest of the software programs in the software library is adjusted synchronously. The core purpose of adjusting the real-time selection probability of the rest of the software programs in the software library is to reflect the actual effect of the software program by using the real-time selection probability; in simple terms, the real-time selection probability of the software program with better performance effect will be increased accordingly, and the real-time selection probability of the software program with poor performance effect will be decreased accordingly, so that the software program with better performance effect will have a greater possibility of being selected as the target software program , and the probability of the software program with poor performance effect being selected as the target software program will be reduced to zero. This adjustment logic is also the core logic of the software gray release self-healing method of the embodiment.

[0073] Specifically, in long-term application, since the software programs in the software library all have a probability of being selected, if the effect of the software program is good (the performance evaluation result meets the preset requirement), the real-time selection probability of the software program will be increased, and if the effect of the software program is poor (the performance evaluation result does not meet the preset requirement), the real-time selection probability of the software program will be decreased. In addition, since the performance evaluation of most types of software programs is not constant, the technical effect will fluctuate. Through the implementation logic of this step, the performance fluctuation will be reflected in the form of real-time selection probability during the long-term operation of the method, so as to spontaneously select the optimal software programs. In a long-term time dimension, the software programs are objectively evaluated, and the limitations of the traditional updating method are expanded.

[0074] Specifically, in combination with the foregoing implementation mode of representing the real-time selection probability by frequency counting, adjusting the real-time selection probability of the software program according to the result of the performance evaluation includes:

[0075] When the result of the performance evaluation does not meet the preset requirement, the number of copies of the software program in the software library is decreased.

[0076] In order to facilitate understanding and application, further, a balance quantity formula is constructed, wherein, For integer values ​​greater than 1 When the performance evaluation results of the substation network do not meet the preset requirements, it is necessary to refer to the software library. Reduced software programs The number of copies. The balancing formula is actually used to limit the number of times a new software program can "fail," when the performance evaluation results of the new software program do not meet the preset requirements. After that, the new software program will be removed from the software library.

[0077] Similarly, the software program is adjusted based on the results of the performance evaluation. Real-time selection probability include:

[0078] The performance evaluation results meet the preset requirements, and the software program Selected as the target software program by the intelligent fusion terminal When the cumulative number of times reaches a preset value, the software program is incremented. In the software library The number of copies, and the software program The cumulative number of times the software program was selected as the target by the intelligent fusion terminal is reset to zero.

[0079] Furthermore, the preset value for the cumulative number of times can be set as an arithmetic sequence, a geometric sequence, or an exponential sequence, which triggers the increase of the software program. In the software library The difficulty of copying the software gradually increases, thus ensuring that if a new software program has better performance, it can replace the original software program at a reasonable speed.

[0080] Furthermore, to prevent the counting frequency of the best-performing software program from growing indefinitely, and to ensure that newly added software programs with better performance can replace the original best-performing program, the software library... In the context of the real-time selection probability Sort in descending order, excluding the target software program. Then, in order of order The software program's score is used as an evaluation criterion for applying the target software program. The performance of the distribution network was evaluated. It is a preset positive integer.

[0081] Specifically, the embodiment can ensure that each software program can be compared with other software programs with better performance excluding itself, so that each software program can be compared with the rest of the software programs with better performance as the standard (rather than the number of frequency counts), and when a newly added software program has better performance, the growth of the frequency count of the software program with the best performance in the original software library will slow down, thereby accelerating the update rate of the software program with excellent performance.

[0082] Specifically, the reduction in the number of software program copies can be understood as the elimination of software programs with insufficient performance, and the increase in the number of software program copies can be understood as the weighting of the selection probability of software programs with better performance, so as to ensure that more time in the transformer network is spent in running software programs with better performance, and at the same time, software programs with better performance have more counting frequencies, so that if a software program with better performance is newly added, the software program with better performance needs to run for more time to replace the software program with the best performance in the original program library, which indirectly verifies the persistent performance of the new software program and achieves the purpose of gray update.

[0083] Figure 3 The structure diagram of the self-healing system for the gray release of the intelligent fusion terminal software of the embodiment of the application is used to realize the self-healing method for the gray release of the intelligent fusion terminal software, and comprises a transformer network.

[0084] The transformer network comprises intelligent fusion terminal software and intelligent sensing units, and one or more intelligent fusion terminal software and one or more intelligent sensing units are arranged in the transformer network according to a preset topology structure.

[0085] In summary, the embodiment of the application discloses a self-healing method and system for the gray release of intelligent fusion terminal software, realizes the automatic operation of the whole process of software program deployment, verification, optimization and elimination, liberates operation and maintenance personnel from heavy version monitoring, decision-making and rollback operations, reduces labor costs and operation risks, ensures that the transformer network runs on the optimal software most of the time, improves the working efficiency of the transformer network, provides a complete, data-driven autonomous management framework for the whole process of software version introduction, verification, promotion, monitoring, decline and elimination, maximizes the value of software assets, and invalid or harmful versions can be cleaned up in time, and an internal competition and survival of the fittest software development environment is generated, and the software programs provided by developers or algorithms will continue to compete in a real and dynamic environment, promote the spiral rise of the functions and performance of the whole system, and lay a foundation for building an intelligent power grid software system with long-term vitality and adaptability.

[0086] The above only describes preferred embodiments of the present application and is not used to limit the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A self-healing method for gray-scale release of intelligent fusion terminal software, characterized in that, include: Software Library include Each of the aforementioned software programs The number of copies is The software library The CCP includes Each of the software programs is described in a software program. It has a corresponding real-time selection probability , , ; The intelligent converged terminal periodically appears in the software library A software program is randomly selected according to the real-time selection probability. As the target software program And broadcast the target software program in the network area. The information, the target software program The information is used to notify the intelligent sensing units and the intelligent fusion terminal in the transformer network to follow the target software program. run, ; Each of the aforementioned software programs The real-time selection probability is adjusted based on an adjustment process, which includes: In the software program Add to the software library The initial real-time selection probability value is assigned in the middle time. And adjust the software library simultaneously. The real-time selection probability of the remaining software programs in the dataset; In the software program Selected as the target software program by the intelligent fusion terminal Then, the intelligent fusion terminal applies the target software program according to preset indicators. The performance of the distribution network was evaluated. The intelligent fusion terminal applies the target software program according to preset indicators. The performance evaluation of the distribution network includes: Based on the software library Evaluation indicators ; Build the target software program Formula for calculating the total score ,in, , , For target software program Regarding evaluation indicators The rating values, different evaluation indicators They have the same range of rating values; The total score is used as the preset indicator to apply the target software program. The performance of the distribution network was evaluated. Adjust the software program based on the performance evaluation results. real-time selection probability And adjust the software library simultaneously. The real-time selection probability of the remaining software programs in the dataset; The software program real-time selection probability The range of values ​​is ; Adjust the software program based on the performance evaluation results. real-time selection probability include: The performance evaluation results meet the preset requirements, and the software program The intelligent fusion terminal is selected as the target software program. When the cumulative number of times reaches a preset value, the software program is incremented. In the software library The number of copies, and the software program The cumulative number of times the software program has been selected as the target software program by the intelligent fusion terminal is reset to zero. When the performance evaluation results do not meet the preset requirements, reduce the software program. In the software library The number of copies.

2. The self-healing method for gray-scale release of intelligent fusion terminal software as described in claim 1, characterized in that, In the software program Add to the software library The initial real-time selection probability value is assigned in the middle time. include: In the software program Add to the software library During the process, the software program will be... The number of copies is expanded to One, the software program Initial real-time selection probability value .

3. The self-healing method for gray-scale release of intelligent fusion terminal software as described in claim 1, characterized in that, Constructing the equilibrium quantity formula ,in, For integer values ​​greater than 1 When the performance evaluation results of the substation network do not meet the preset requirements, it is necessary to refer to the software library. Reduced software programs The number of copies.

4. The self-healing method for gray-scale release of intelligent fusion terminal software as described in claim 1, characterized in that, In the software library In the context of the real-time selection probability Sort in descending order, excluding the target software program. Then, in order of order The software program's score is used as an evaluation criterion for applying the target software program. The performance of the distribution network was evaluated. It is a preset positive integer.

5. The self-healing method for gray-scale release of intelligent fusion terminal software as described in claim 1, characterized in that, The preset indicators are for the target software program. Descriptive evaluation metrics.

6. A self-healing system for gray-scale release of intelligent fusion terminal software, characterized in that, The method for implementing the self-healing method of gray-scale release of intelligent converged terminal software according to any one of claims 1 to 5 includes a transformer network; The transformer substation network includes intelligent converged terminal software and intelligent sensing units, with one or more intelligent converged terminal software units and one or more intelligent sensing units respectively arranged in the transformer substation network according to a preset topology.

Images