Method for detecting reliability of intelligent terminal software

Abstract

The invention provides an intelligent terminal software reliability detection method, which comprises the following steps: acquiring intelligent terminal software reliability detection related parameters, the related parameters comprise terminal detection parameters, detection parameters between the master station and the terminals, detection parameters between the terminals and the electricity meters, electricity meter related test parameters, master station terminal electricity meter full-system detection parameters and field environment simulation test parameters; relevant parameters are divided into K categories, a Gaussian mixture model is constructed, and parameters for testing the reliability of the intelligent terminal software are adjusted; and obtaining the adjusted reliability test parameters, testing to obtain a test result, and evaluating the reliability of the intelligent terminal software according to the test result. The service architecture is constructed around six key dimensions, the operation robustness, the communication stability, the data reporting strategy, the meter reading efficiency, the protocol support and the compatibility of the intelligent terminal software and the performance of the system in various scenes are comprehensively evaluated, and the reliability and the maintainability of the intelligent terminal software are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent terminal software detection, and in particular to a method for detecting the reliability of intelligent terminal software in a smart grid. Background Art

[0002] With the in-depth development of the State Grid Corporation of China's smart grid construction, smart energy meters and electricity consumption information collection systems, as key components of the smart grid, have a significant impact on the stable operation of the power grid and the vital interests of power users due to the success rate and accuracy of their data collection.

[0003] Although the intelligent terminal has passed the routine functional tests of the metrology center laboratory during actual field operation, there are still problems such as reduced data collection success rate due to insufficient software reliability.

[0004] At present, the detection scheme of intelligent terminals mainly focuses on the conventional functions of the terminals, while ignoring the detection of software reliability, such as operating system stability, key data integrity, remote communication continuity, and software upgrade capabilities. In addition, due to the complexity of the on-site operating environment, the existing detection methods are difficult to simulate on-site problems, resulting in the inability to effectively evaluate the terminal exception handling mechanism. At the same time, there is also a lack of effective means to detect the long-term operation stability of various types of equipment in the system. Summary of the invention

[0005] The present invention relates to a method for detecting the reliability of software of an intelligent terminal, which aims to ensure that the operating system of the intelligent terminal will not freeze, key data will not be damaged, remote communication will not be disconnected, fatal faults can be recovered, and the operating system and software are upgradeable; the performance of the intelligent terminal in various actual operation scenarios is simulated and tested to improve its software reliability.

[0006] The technical solution of the present invention is:

[0007] The present invention provides a method for detecting the reliability of intelligent terminal software, comprising:

[0008] Obtaining parameters related to reliability detection of intelligent terminal software, including terminal detection parameters, detection parameters between the master station and the terminal, detection parameters between the terminal and the electric meter, electric meter related test parameters, master station terminal electric meter full system detection parameters and field environment simulation test parameters;

[0009] Classifying the relevant parameters to obtain K parameter categories, and constructing a Gaussian mixture model based on the K parameter categories;

[0010] According to the Gaussian mixture model, adjusting the parameters of the intelligent terminal software reliability test;

[0011] The adjusted reliability test parameters are obtained, and the reliability test of the intelligent terminal software is performed to obtain the test results, and the reliability of the intelligent terminal software is evaluated according to the test results.

[0012] Furthermore, the Gaussian mixture model includes K Gaussian distributions, and each Gaussian distribution corresponds to a parameter category.

[0013] Furthermore, constructing the Gaussian mixture model includes:

[0014] The number of iterations P and the expected return success rate Q are preset, and the mean vector, covariance matrix and mixing weight of each Gaussian distribution are randomly initialized.

[0015] Furthermore, constructing the Gaussian mixture model also includes:

[0016] The posterior probability that each detection parameter belongs to each Gaussian distribution is calculated according to the current Gaussian mixture model, and the posterior probability is used as the responsibility degree.

[0017] Furthermore, constructing the Gaussian mixture model also includes:

[0018] The mean vector, covariance matrix and mixing weight of each Gaussian distribution are updated based on the responsibility to obtain a new Gaussian mixture model, and the benefit of the new Gaussian mixture model relative to the current Gaussian mixture model is calculated.

[0019] Furthermore, constructing the Gaussian mixture model also includes:

[0020] If the profit is greater than the expected profit success rate Q or the number of iterations P is not reached, the new Gaussian mixture model is used as the current Gaussian mixture model, and the step of calculating the posterior probability of each detection parameter belonging to each Gaussian distribution according to the current Gaussian mixture model is returned;

[0021] Otherwise, the new Gaussian mixture model is used as the final Gaussian mixture model.

[0022] Furthermore, the benefit of the new Gaussian mixture model relative to the current Gaussian mixture model is the difference in covariance matrices.

[0023] Further, the terminal detection parameters include terminal self-check parameters, version consistency check parameters and data integrity check parameters;

[0024] The detection parameters between the master station and the terminal include communication protocol consistency check parameters, data transmission accuracy check parameters and communication fault detection parameters;

[0025] The detection parameters between the terminal and the electric meter include electric meter access detection parameters, electric energy data collection detection parameters and electric energy metering abnormality detection parameters.

[0026] Further, the terminal detection parameters include: terminal operation stability percentage, terminal upgrade success rate, terminal meter reading success rate, terminal event reporting success rate and terminal online rate;

[0027] The detection parameters between the master station and the terminal include: master station communication success rate, master station communication performance, master station data reading success rate and master station parameter setting success rate;

[0028] The detection parameters between the terminal and the electric meter include: the terminal meter reading success rate and the terminal meter reading efficiency;

[0029] The meter-related test parameters include: clock deviation rate and meter freezing success rate;

[0030] The master station terminal meter full system detection parameters include: system data throughput rate and system data transmission performance;

[0031] The field environment simulation test parameters include: data integrity, acquisition success rate and data correctness.

[0032] Beneficial effects of the present invention:

[0033] The intelligent terminal software reliability detection method proposed in the present invention brings about significant beneficial effects in many aspects; first, it comprehensively detects the software reliability of the intelligent terminal, and through simulation and testing of various actual operation scenarios, ensures that the operating system will not crash, key data will not be damaged, remote communication will not be disconnected, and fatal faults are recoverable. Secondly, the present invention focuses on detecting the upgrade capabilities of the operating system and software, ensuring that the intelligent terminal can adapt to the upgrade needs of future technologies, thereby extending its service life and reducing maintenance costs. In addition, the detection method of the present invention also pays attention to the complexity of the on-site operating environment. By simulating the typical substation environment, continuously monitoring the collection tasks, data integrity, correctness and collection success rate, the terminal exception handling mechanism can be effectively evaluated, further enhancing the stability and adaptability of the system. This not only ensures the stable operation of the power grid, but also improves the satisfaction and trust of power users.

[0034] The intelligent terminal software reliability detection method of the present invention comprehensively detects all device objects in the system, as well as the communication and interaction between all device objects, to ensure the comprehensiveness of the acquisition system detection. The intelligent terminal software compatibility and reliability are comprehensively detected by simulating normal operation, abnormal operation, accelerated operation, and artificially implanting possible faults on site.

[0035] The reliability testing method based on the improved Gaussian mixture model (GMM) adopted in the present invention can optimize and adjust the use case scheme configuration of the stress test and the long-term stability test so as to achieve the best test effect.

[0036] Other features and advantages of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present invention.

[0038] Figure 1 A flow chart of a method for detecting the reliability of intelligent terminal software according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0039] The preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0040] Figure 1 A flow chart of a method for detecting the reliability of intelligent terminal software according to an embodiment of the present invention is shown.

[0041] like Figure 1 As shown, the present invention provides a method for detecting the reliability of intelligent terminal software, comprising:

[0042] Obtaining parameters related to reliability detection of intelligent terminal software, including terminal detection parameters, detection parameters between the master station and the terminal, detection parameters between the terminal and the electric meter, electric meter related test parameters, master station terminal electric meter full system detection parameters and field environment simulation test parameters;

[0043] Classifying the relevant parameters to obtain K parameter categories, and constructing a Gaussian mixture model based on the K parameter categories;

[0044] According to the Gaussian mixture model, adjusting the parameters of the intelligent terminal software reliability test;

[0045] The adjusted reliability test parameters are obtained, and the reliability test of the intelligent terminal software is performed to obtain the test results, and the reliability of the intelligent terminal software is evaluated according to the test results.

[0046] In one example, the Gaussian mixture model includes K Gaussian distributions, each Gaussian distribution corresponds to a parameter category, and constructing the Gaussian mixture model includes:

[0047] The number of iterations P and the expected success rate of return Q are preset, and the mean vector μ of each Gaussian distribution is randomly initialized k , covariance matrix θk and the mixing weight τ k ;

[0048] E step (expected step):

[0049] Calculate each detection parameter x according to the current Gaussian mixture model i The posterior probability of each Gaussian distribution is taken as the responsibility γ ik ;

[0050] Where N(x|μ,θ) is the probability density function of the multivariate Gaussian distribution;

[0051] M-step (maximization step):

[0052] Based on the responsibility degree, the mean vector, the covariance matrix and the mixing weight of each Gaussian distribution are updated to obtain a new Gaussian mixture model;

[0053] Mixed weights: where a k is the adjustment factor, and In a typical case,

[0054] Mean vector:

[0055] Covariance matrix:

[0056] The mixing weights must satisfy And each τ k >0;

[0057] Repeat steps E and M to calculate the benefit of the new Gaussian mixture model relative to the current Gaussian mixture model, ε = || θ j+1 -θ j ||;

[0058] If ε≤Q after P iterations, it means that the current model meets the maximum benefit;

[0059] When ε>Q after P iterations, adjust a k Repeat steps E and M until the requirement of ε≤Q is met.

[0060] According to the completed Gaussian mixture model, adjust the relevant parameters of stress test and long-term stability test.

[0061] The preferred K≤5; P=100; Q=0.001.

[0062] If the profit is greater than the expected profit success rate Q or the number of iterations P is not reached, the new Gaussian mixture model is used as the current Gaussian mixture model, and the step of calculating the posterior probability of each detection parameter belonging to each Gaussian distribution according to the current Gaussian mixture model is returned;

[0063] Otherwise, the new Gaussian mixture model is used as the final Gaussian mixture model.

[0064] In one example,

[0065] 1. Terminal detection, which is used to evaluate the operational robustness of the terminal under normal and abnormal conditions; the terminal detection parameters include: terminal operation stability percentage, terminal upgrade success rate, terminal meter reading success rate, terminal event reporting success rate and terminal online rate;

[0066] Among them, terminal operation stability percentage = actual operation time / preset operation time × 100%. ;

[0067] Terminal upgrade success rate = actual number of successful upgrades / preset number of upgrades × 100%;

[0068] Terminal meter reading success rate = actual number of successful meter reading data points / preset number of successful meter reading data points × 100%;

[0069] Terminal event reporting success rate = number of successfully reported events / number of actual events × 100%;

[0070] Terminal online rate = actual online time / (test end time - test start time) × 100%.

[0071] 2. Detection between the master station and the terminal, used to examine the stability of the communication channel and the master station parameter setting function; the detection parameters between the master station and the terminal include: master station communication success rate, master station communication performance, master station data reading success rate and master station parameter setting success rate;

[0072] Among them, the master station communication success rate = actual online time / (test end time-test start time)×100%;

[0073] Master station communication performance = actual communication rate / theoretical bandwidth × 100%;

[0074] The master station data reading success rate = the actual number of data points successfully read / the actual number of data points stored in the terminal × 100%;

[0075] Master station parameter setting success rate = actual number of successful settings / preset number of settings × 100%.

[0076] 3. Detection between the terminal and the meter, used to examine the meter reading strategy and HPLC meter reading concurrent mechanism; the detection parameters between the terminal and the meter include: terminal meter reading success rate and terminal meter reading efficiency;

[0077] Among them, the terminal meter reading success rate = the actual number of successful meter reading data points / the actual number of frozen data points of the meter × 100%;

[0078] Terminal meter reading efficiency = actual meter reading time / theoretical meter reading time × 100%.

[0079] 4. Electricity meter related tests, used to verify the terminal's support for multiple protocols of the electricity meter; the electricity meter related tests include: clock deviation rate and electricity meter freezing success rate;

[0080] Among them, clock deviation rate = clock deviation duration / (test end time - test start time) × 100%;

[0081] Meter freezing success rate = actual number of frozen data / theoretical number of frozen data × 100.

[0082] 5. Master station-terminal-electricity meter full system detection, used to examine the performance of the entire system in a specific scenario; the master station terminal electric meter full system detection parameters include: system data throughput rate and system data transmission performance;

[0083] Among them, the total data throughput rate = the total data volume (in bytes) / completion time × 100%;

[0084] System data transmission performance = actual transmission rate.

[0085] 6. On-site environment simulation test, used to continuously monitor data collection tasks by simulating typical substations; the on-site environment simulation test parameters include: data integrity, collection success rate and data correctness.

[0086] Among them, data integrity = complete data transmitted within a period of time / all data transmitted within this period of time × 100%;

[0087] Collection success rate = number of data points successfully read by the master station / number of data points actually frozen by the meter × 100%;

[0088] Data accuracy = the number of correct data points successfully read by the master station within a period of time / the number of data points required to be read within the period of time × 100%.

[0089] The business architecture of the present invention is built around six key dimensions, including terminal detection, detection between the master station and the terminal, detection between the terminal and the meter, meter-related tests, master station-terminal-meter full system detection, and field environment simulation test. These detection dimensions comprehensively evaluate the operation robustness, communication stability, data reporting strategy, meter reading efficiency, protocol support and compatibility of the smart terminal software, as well as the performance of the system in various scenarios, to ensure the reliability and maintainability of the smart terminal software.

[0090] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A method for detecting the reliability of intelligent terminal software, characterized in that: include: Obtaining parameters related to reliability detection of intelligent terminal software, including terminal detection parameters, detection parameters between the master station and the terminal, detection parameters between the terminal and the electric meter, electric meter related test parameters, master station terminal electric meter full system detection parameters and field environment simulation test parameters; Classifying the relevant parameters to obtain K parameter categories, and constructing a Gaussian mixture model based on the K parameter categories; According to the Gaussian mixture model, adjusting the parameters of the intelligent terminal software reliability test; The adjusted reliability test parameters are obtained, and the reliability test of the intelligent terminal software is performed to obtain the test results, and the reliability of the intelligent terminal software is evaluated according to the test results.

2. The intelligent terminal software reliability detection method according to claim 1 is characterized in that: The Gaussian mixture model includes K Gaussian distributions, each Gaussian distribution corresponds to a parameter category.

3. The intelligent terminal software reliability detection method according to claim 2 is characterized in that: Constructing the Gaussian mixture model includes: The number of iterations P and the expected return success rate Q are preset, and the mean vector, covariance matrix and mixing weight of each Gaussian distribution are randomly initialized.

4. The intelligent terminal software reliability detection method according to claim 3 is characterized in that: Constructing the Gaussian mixture model also includes: The posterior probability that each detection parameter belongs to each Gaussian distribution is calculated according to the current Gaussian mixture model, and the posterior probability is used as the responsibility degree.

5. The intelligent terminal software reliability detection method according to claim 4 is characterized in that: Constructing the Gaussian mixture model also includes: The mean vector, covariance matrix and mixing weight of each Gaussian distribution are updated based on the responsibility to obtain a new Gaussian mixture model, and the benefit of the new Gaussian mixture model relative to the current Gaussian mixture model is calculated.

6. The intelligent terminal software reliability detection method according to claim 5 is characterized in that: Constructing the Gaussian mixture model also includes: If the profit is greater than the expected profit success rate Q or the number of iterations P is not reached, the new Gaussian mixture model is used as the current Gaussian mixture model, and the step of calculating the posterior probability of each detection parameter belonging to each Gaussian distribution according to the current Gaussian mixture model is returned; Otherwise, the new Gaussian mixture model is used as the final Gaussian mixture model.

7. The intelligent terminal software reliability detection method according to claim 4 is characterized in that: The gain of the new Gaussian mixture model relative to the current Gaussian mixture model is the difference in covariance matrices.

8. The intelligent terminal software reliability detection method according to any one of claims 1 to 7, characterized in that: The terminal detection parameters include terminal self-check parameters, version consistency check parameters and data integrity check parameters; The detection parameters between the master station and the terminal include communication protocol consistency check parameters, data transmission accuracy check parameters and communication fault detection parameters; The detection parameters between the terminal and the electric meter include electric meter access detection parameters, electric energy data collection detection parameters and electric energy metering abnormality detection parameters.

9. The intelligent terminal software reliability detection method according to claim 8, characterized in that: The terminal detection parameters include: terminal operation stability percentage, terminal upgrade success rate, terminal meter reading success rate, terminal event reporting success rate and terminal online rate; The detection parameters between the master station and the terminal include: master station communication success rate, master station communication performance, master station data reading success rate and master station parameter setting success rate; The detection parameters between the terminal and the electric meter include: the terminal meter reading success rate and the terminal meter reading efficiency; The meter-related test parameters include: clock deviation rate and meter freezing success rate; The master station terminal meter full system detection parameters include: system data throughput rate and system data transmission performance; The field environment simulation test parameters include: data integrity, acquisition success rate and data correctness.

Images