A multi-mode fusion adaptive test data intelligent acquisition method and system

Abstract

The present invention proposes a multi-mode fusion adaptive test data intelligent acquisition method and system, the method comprising the following steps: 1) adopting three different methods, namely, a timing acquisition mode, a variable value trigger acquisition mode and an event trigger acquisition mode, to collect test data; 2) calculating the normalization factor of the collected test data, and calculating the first evaluation index in combination with the preset weight; 3) calculating the system difference loss function between the collected experimental data and the expected data; 4) using the first evaluation index and the system difference loss function as comprehensive evaluation indicators of the acquisition scheme, evaluating the current acquisition scheme, and adjusting the acquisition scheme according to the evaluation results; 5) continuing to collect data according to the adjusted acquisition method, repeating steps 1-4 until an ideal acquisition scheme is reached.

Description

Technical Field

[0001] The present invention relates to the field of data acquisition, and in particular to a multi-mode fusion adaptive test data intelligent acquisition method and system. Background Art

[0002] Test data refers to all data used and generated in equipment test and appraisal activities. It is the "first-hand information" that records major activities such as equipment testing, training, research and development, and scientific research in a timely, accurate and true manner. It is high-value data that is of common concern to equipment appraisal finalization parties, demonstration and development parties, and use and improvement parties. It is an important basic strategic resource to support the development of equipment construction. "Test data engineering" with test data as the research object, test data activities as the research content, and the goal of achieving efficient management, reuse, sharing and application of test data has become one of the current research hotspots in the field of test and appraisal.

[0003] Different from ordinary technical research, test data engineering is 30% technology and 70% management. The selection of management mechanism profoundly affects the construction of technical solutions in test data construction; therefore, management mechanism research is the premise and foundation for carrying out test data engineering construction. However, due to the constraints of concepts, technology, system and other factors, test data has been managed separately and stored in a decentralized manner for a long time. The management processes and methods of key parts such as data planning, collection, handover, sharing and application are not standardized, which has become a key factor restricting the efficiency of test data engineering construction and hindering the value-added and efficiency improvement of test data.

[0004] Test data is a high-value data resource generated in equipment test and appraisal activities, and is an important foundation for supporting equipment construction and development. At present, test data has been in a state of decentralized management and decentralized storage for a long time, and key links such as data collection, communication, and sharing lack a standardized management mechanism, which has become a key factor restricting the efficiency of test data engineering construction. Therefore, it is urgent to study a new technical solution that can achieve efficient collection, dynamic management, and intelligent application of test data. Summary of the invention

[0005] In view of this, the present invention provides a multi-mode fusion adaptive test data intelligent collection method, which comprises the following steps:

[0006] Step 1: Use three different methods, namely, timing acquisition mode, variable value trigger acquisition mode and event trigger acquisition mode, to collect test data;

[0007] Step 2, calculating the normalization factor of the collected test data, and calculating the first evaluation index in combination with the preset weight;

[0008] Step 3, calculating the system difference loss function between the collected experimental data and the expected data;

[0009] Step 4: using the first evaluation index and the system difference loss function as comprehensive evaluation indexes of the acquisition scheme, evaluating the current acquisition scheme, and adjusting the acquisition scheme according to the evaluation results;

[0010] Step 5, continue to collect data according to the adjusted collection method, and repeat steps 1-4 until an ideal collection plan is reached.

[0011] In particular, three different methods were used to collect test data, including:

[0012] The timing acquisition mode is: according to a preset acquisition cycle, the test data is periodically acquired and the acquisition time stamp is stored;

[0013] The variable value trigger acquisition mode is: monitoring the preset data of interest, and when the data of interest changes, collecting test data once and storing the acquisition time stamp;

[0014] The event-triggered acquisition mode is: on the basis of the variable value triggered acquisition mode, a trigger condition for the change of the concerned data is set, and when the concerned data changes and meets the trigger condition, the test data is collected once and the acquisition time stamp is stored.

[0015] In particular, the normalization factors include: a normalized value of a coverage factor V, a normalized value of a collection timeliness factor T, and a normalized value of an environmental factor H; wherein calculating the normalized value of the coverage factor V of the collected data includes:

[0016] f1(V)=exp·(-k1*V 2 );

[0017] Among them, f1(V) represents the coverage or quality of the collected data, V represents the coverage rate or repetition rate, and k1 represents the sensitivity parameter of the collection range;

[0018] Calculation of the standardized value of the collection timeliness factor T includes:

[0019] f2(T)=tanh(k2·(T-T0));

[0020] f2(T) represents the timeliness of data collection, T represents the delay time of data collection, T0 represents the target reference value of data collection timeliness; k2 represents the sensitivity parameter of data collection timeliness;

[0021] The standardized value for calculating the environmental factor H includes:

[0022]

[0023] f3(H) represents the impact of environmental factors on collection, H represents the network status index; k3 represents the sensitivity parameter of the collection environment factors.

[0024] In particular, the calculation of the first evaluation index is implemented by the following expression:

[0025] HI=w1·f1(V)+w2·f2(T)+w3·f3(H)+w4·log(G);

[0026] Among them, HI is the first evaluation index to be calculated, w1, w2, w3, and w4 are preset weights, log(G) is the total amount of collected data, and G is directly used for the amount of collected data.

[0027] In particular, calculating the systematic difference loss function between the collected data and the expected data specifically includes:

[0028]

[0029] Where v represents the system difference loss function, S gt , C gt are the key characteristic parameters of the topology of the expected system and the key characteristic parameters of the device heterogeneity, respectively; S p , C p They are the key characteristic parameters of the topological structure of the actual acquisition system and the key characteristic parameters of the device heterogeneity respectively.

[0030] In particular, the first evaluation index and the system difference loss function are used as evaluation indexes, and the parameters of the acquisition scheme are optimized and adjusted according to the calculation results, specifically including: if the first evaluation index is greater than a first threshold, or the system difference loss function is less than a second threshold, then the existing acquisition scheme is maintained; if the first evaluation index is less than the first threshold, or the system difference loss function is greater than the second threshold, then the acquisition mode is switched, the acquisition parameters are adjusted, or the preset weights of the comprehensive indicators are adjusted.

[0031] The present invention also discloses a multi-mode fusion adaptive test data intelligent acquisition system, comprising:

[0032] Experimental data acquisition module, used to collect experimental data using three different methods: timing acquisition mode, variable value trigger acquisition mode and event trigger acquisition mode;

[0033] A first evaluation index calculation module, used to calculate the normalization factor of the collected test data, and calculate the first evaluation index in combination with a preset weight;

[0034] A system difference loss function calculation module, used to calculate the system difference loss function between the collected experimental data and the expected data;

[0035] A collection scheme adjustment module is used to evaluate the current collection scheme by using the first evaluation index and the system difference loss function as comprehensive evaluation indexes of the collection scheme, and adjust the collection scheme according to the evaluation results;

[0036] The acquisition scheme generation module is used to continue to collect data according to the adjusted acquisition method, and repeatedly execute the experimental data acquisition module to the acquisition scheme adjustment module until an ideal acquisition scheme is generated.

[0037] Beneficial effects:

[0038] The technical solution proposed by the present invention can achieve the following technical effects:

[0039] By adopting three different collection modes, the diverse needs for test data collection can be met and the timeliness, continuity and integrity of the data can be improved.

[0040] The introduction of standardized factors and comprehensive evaluation indicators can dynamically evaluate the performance of the current acquisition scheme, and adaptively adjust the acquisition parameters according to the evaluation results to improve the quality of test data.

[0041] Combined with the system difference loss function, the acquisition plan can be further optimized to make it closer to the characteristics of the expected system and improve the representativeness and analytical value of the data.

[0042] By repeatedly executing the cycle of acquisition-evaluation-adjustment, an ideal adaptive test data acquisition plan can be finally generated.

[0043] The intelligent acquisition system proposed in the present invention can effectively realize the efficient management, reuse and sharing of test data, and provide technical support for the construction of test data engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the multi-mode fusion adaptive test data intelligent collection method proposed in the present invention;

[0045] Figure 2 This is a schematic diagram of the multi-mode fusion adaptive test data intelligent acquisition system proposed in the present invention. DETAILED DESCRIPTION

[0046] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0047] The present invention provides a multi-mode fusion adaptive test data intelligent acquisition method, such as Figure 1 As shown, it includes the following steps:

[0048] Step 1: Use three different methods, namely, timing acquisition mode, variable value trigger acquisition mode and event trigger acquisition mode, to collect test data; use three different methods to collect test data, specifically including:

[0049] The timing acquisition mode is: according to a preset acquisition cycle, the test data is periodically acquired and the acquisition time stamp is stored;

[0050] The variable value trigger acquisition mode is: monitoring the preset data of interest, and when the data of interest changes, collecting test data once and storing the acquisition time stamp;

[0051] The event-triggered acquisition mode is: on the basis of the variable value triggered acquisition mode, a trigger condition for the change of the concerned data is set, and when the concerned data changes and meets the trigger condition, the test data is collected once and the acquisition time stamp is stored.

[0052] Step 2, calculate the normalization factor of the collected test data, and calculate the first evaluation index in combination with the preset weight; the normalization factor includes: the normalized value of the coverage factor V, the normalized value of the collection timeliness factor T and the normalized value of the environmental factor H; wherein, calculating the normalized value of the coverage factor V of the collected data includes:

[0053] f1(V)=exp(-k1*V 2 );

[0054] Among them, f1(V) represents the coverage or quality of the collected data, V represents the coverage rate or repetition rate, and k1 represents the sensitivity parameter of the collection range;

[0055] Calculation of the standardized value of the collection timeliness factor T includes:

[0056] f2(T)=tanh(k2·(T-T0));

[0057] f2(T) represents the timeliness of data collection, T represents the delay time of data collection, T0 represents the target reference value of data collection timeliness; k2 represents the sensitivity parameter of data collection timeliness;

[0058] The standardized value for calculating the environmental factor H includes:

[0059]

[0060] f3(H) represents the impact of environmental factors on collection, H represents the network status index; k3 represents the sensitivity parameter of the collection environmental factors. The calculation of the first evaluation index is achieved by the following expression:

[0061] HI=w1·f1(V)+w2·f2(T)+w3·f3(H)+w4·log(G);

[0062] Among them, HI is the first evaluation index to be calculated, w1, w2, w3, and w4 are preset weights, log(G) is the total amount of collected data, and G is directly used for the amount of collected data.

[0063] Step 3, calculating the system difference loss function between the collected experimental data and the expected data;

[0064] The system difference loss function between the collected data and the expected data is calculated by:

[0065]

[0066] Where v represents the system difference loss function, S gt , C gt are the key characteristic parameters of the topology of the expected system and the key characteristic parameters of the device heterogeneity, respectively; S p , C p They are the key characteristic parameters of the topological structure of the actual acquisition system and the key characteristic parameters of the device heterogeneity respectively.

[0067] Step 4: using the first evaluation index and the system difference loss function as comprehensive evaluation indexes of the acquisition scheme, evaluating the current acquisition scheme, and adjusting the acquisition scheme according to the evaluation results;

[0068] The first evaluation index and the system difference loss function are used as evaluation indexes, and the parameters of the acquisition scheme are optimized and adjusted according to the calculation results, specifically including: if the first evaluation index is greater than a first threshold, or the system difference loss function is less than a second threshold, then the existing acquisition scheme is maintained; if the first evaluation index is less than the first threshold, or the system difference loss function is greater than the second threshold, then the acquisition mode is switched, the acquisition parameters are adjusted, or the preset weights of the comprehensive indicators are adjusted.

[0069] For example, if HI < 0.6 (the first evaluation index is low), the following adjustments can be made:

[0070] Switch the acquisition mode:

[0071] If you are currently using the timing acquisition mode, you can switch to the variable value trigger or event trigger mode to increase the timeliness of the data;

[0072] If you are currently using the variable value trigger or event trigger mode, you can switch to the timing acquisition mode to increase the continuity of the data;

[0073] Adjust acquisition parameters:

[0074] If the timing acquisition mode is used, the acquisition cycle can be shortened and the data acquisition frequency can be increased;

[0075] If you use the variable value trigger or event trigger mode, you can lower the trigger threshold and increase the trigger sensitivity;

[0076] Adjust the weight of comprehensive indicators:

[0077] Appropriately increase the weight w2 of the collection timeliness factor f2(T);

[0078] Appropriately reduce the weight fw3 of environmental factor f3(H);

[0079] If ν>0.4 (high system difference loss), the following adjustments can be made:

[0080] Switch the acquisition mode:

[0081] If you are currently using the timing acquisition mode, you can switch to the variable value trigger or event trigger mode to enhance the capture of key system parameters;

[0082] If you are currently using the variable value trigger or event trigger mode, you can add more trigger conditions to cover more system parameters;

[0083] Adjust acquisition parameters:

[0084] Adjust the trigger conditions to make them closer to the key characteristic parameters of the expected system;

[0085] Appropriately increase the resolution of collected data to obtain more detailed system parameter information;

[0086] Adjust the weight of comprehensive indicators:

[0087] Appropriately increase the weight w4 of the system difference loss function ν in HI;

[0088] Appropriately reduce the weight of collection coverage f1(V) and environmental factor f3(H);

[0089] Step 5, continue to collect data according to the adjusted collection method, and repeat steps 1-4 until an ideal collection plan is reached.

[0090] The present invention also discloses a multi-mode fusion adaptive test data intelligent acquisition system, comprising:

[0091] Experimental data acquisition module, used to collect experimental data using three different methods: timing acquisition mode, variable value trigger acquisition mode and event trigger acquisition mode;

[0092] A first evaluation index calculation module, used to calculate the normalization factor of the collected test data, and calculate the first evaluation index in combination with a preset weight;

[0093] A system difference loss function calculation module, used to calculate the system difference loss function between the collected experimental data and the expected data;

[0094] A collection scheme adjustment module is used to evaluate the current collection scheme by using the first evaluation index and the system difference loss function as comprehensive evaluation indexes of the collection scheme, and adjust the collection scheme according to the evaluation results;

[0095] The acquisition scheme generation module is used to continue to collect data according to the adjusted acquisition method, and repeatedly execute the experimental data acquisition module to the acquisition scheme adjustment module until an ideal acquisition scheme is generated.

[0096] The scheme in this embodiment corresponds one-to-one with the technical features in the previous method embodiment, so it will not be repeated here.

[0097] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0098] It is obvious to those skilled in the art that the embodiments of the present invention are not limited to the details of the above exemplary embodiments, and that the embodiments of the present invention can be implemented in other specific forms without departing from the spirit or basic features of the embodiments of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the embodiments of the present invention is limited by the attached claims rather than the above description, so it is intended to include all changes that fall within the meaning and scope of the equivalent elements of the claims in the embodiments of the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units, modules or devices stated in the system, device or terminal claims can also be implemented by the same unit, module or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0099] Finally, it should be noted that the above implementation modes are only used to illustrate the technical solutions of the embodiments of the present invention and are not intended to limit them. Although the embodiments of the present invention have been described in detail with reference to the above preferred implementation modes, those skilled in the art should understand that the technical solutions of the embodiments of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-mode fusion adaptive test data intelligent collection method, which comprises the following steps: Step 1: Use three different methods, namely, timing acquisition mode, variable value trigger acquisition mode and event trigger acquisition mode, to collect test data; Step 2, calculating the normalization factor of the collected test data, and calculating the first evaluation index in combination with the preset weight; Step 3, calculating the system difference loss function between the collected experimental data and the expected data; Step 4: using the first evaluation index and the system difference loss function as comprehensive evaluation indexes of the acquisition scheme, evaluating the current acquisition scheme, and adjusting the acquisition scheme according to the evaluation results; Step 5, continue to collect data according to the adjusted collection method, and repeat steps 1-4 until an ideal collection plan is reached.

2. The multi-mode fusion adaptive test data intelligent collection method according to claim 1, characterized in that: Three different methods were used to collect test data, including: The timing acquisition mode is: according to a preset acquisition cycle, the test data is periodically acquired and the acquisition time stamp is stored; The variable value trigger acquisition mode is: monitoring the preset data of interest, and when the data of interest changes, collecting test data once and storing the acquisition time stamp; The event-triggered acquisition mode is: on the basis of the variable value triggered acquisition mode, a trigger condition for the change of the concerned data is set, and when the concerned data changes and meets the trigger condition, the test data is collected once and the acquisition time stamp is stored.

3. The multi-mode fusion adaptive test data intelligent collection method according to claim 2, characterized in that: The standardized factors include: a standardized value of a coverage factor V, a standardized value of a collection timeliness factor T, and a standardized value of an environmental factor H; wherein calculating the standardized value of the coverage factor V of the collected data includes: f1(V)=exp(-k1*V 2 ); Among them, f1(V) represents the coverage or quality of the collected data, V represents the coverage rate or repetition rate, and k1 represents the sensitivity parameter of the collection range; Calculation of the standardized value of the collection timeliness factor T includes: f2(T)=tanh(k2·(T-T0)); f2(T) represents the timeliness of data collection, T represents the delay time of data collection, T0 represents the target reference value of data collection timeliness; k2 represents the sensitivity parameter of data collection timeliness; The standardized value for calculating the environmental factor H includes: f3(H) represents the impact of environmental factors on collection, H represents the network status index; k3 represents the sensitivity parameter of the collection environment factors.

4. The multi-mode fusion adaptive test data intelligent collection method according to claim 3, characterized in that: The calculation of the first evaluation index is achieved by the following expression: HI=w1·f1(V)+w2·f2(T)+w3·f3(H)+w4·log(G); Among them, HI is the calculation of the first evaluation index, w1, w2, w3, w4 are preset weights, log(G) is the total amount of collected data, and G directly uses the amount of collected data.

5. The multi-mode fusion adaptive test data intelligent collection method according to claim 4, characterized in that: The system difference loss function between the collected data and the expected data is calculated by: Where ν represents the system difference loss function, S gt , C gt are the key characteristic parameters of the topology of the expected system and the key characteristic parameters of the device heterogeneity, respectively; S p , C p They are the key characteristic parameters of the topological structure of the actual acquisition system and the key characteristic parameters of the device heterogeneity respectively.

6. The multi-mode fusion adaptive test data intelligent collection method according to claim 5, characterized in that: The first evaluation index and the system difference loss function are used as evaluation indexes, and the parameters of the acquisition scheme are optimized and adjusted according to the calculation results, specifically including: if the first evaluation index is greater than a first threshold, or the system difference loss function is less than a second threshold, then the existing acquisition scheme is maintained; if the first evaluation index is less than the first threshold, or the system difference loss function is greater than the second threshold, then the acquisition mode is switched, the acquisition parameters are adjusted, or the preset weights of the comprehensive indicators are adjusted.

7. A multi-mode fusion adaptive test data intelligent acquisition system, characterized in that: include: Experimental data acquisition module, used to collect experimental data using three different methods: timing acquisition mode, variable value trigger acquisition mode and event trigger acquisition mode; A first evaluation index calculation module, used to calculate the normalization factor of the collected test data, and calculate the first evaluation index in combination with a preset weight; A system difference loss function calculation module, used to calculate the system difference loss function between the collected experimental data and the expected data; A collection scheme adjustment module is used to evaluate the current collection scheme by using the first evaluation index and the system difference loss function as comprehensive evaluation indexes of the collection scheme, and adjust the collection scheme according to the evaluation results; The acquisition scheme generation module is used to continue to acquire data according to the adjusted acquisition method, and repeatedly execute the experimental data acquisition module to the acquisition scheme adjustment module until an ideal acquisition scheme is generated.

Images