A Big Data-Based Learnable Onboard Computer AI Test and Maintenance System and Method

Abstract

This invention provides an onboard computer AI testing and maintenance system and method based on big data learning. It includes a hardware data acquisition module and a software data testing, analysis, and fault diagnosis module. The hardware data acquisition module comprises a power supply control unit, a timing control unit, and a data acquisition unit. The power supply control unit provides power to the onboard computer intelligent AI testing and maintenance system; the timing control unit primarily handles cross-clock domain issues across eight channels; and the data acquisition unit performs data sampling. The software data testing, analysis, and fault diagnosis module includes a main control program component, a data testing and analysis component, and a data fault diagnosis AI component. The main control program component is used for the overall system operation and the scheduling and coordination of various functions; the data testing and analysis component mainly performs batch analysis of the acquired data; and the data fault diagnosis AI component mainly performs deep learning on the test big data. Compared with existing technologies, this invention has the significant advantage of high integration.

Description

Technical Field

[0001] This invention belongs to the field of onboard computer data test batch processing, specifically, it relates to an onboard computer AI test and maintenance system and method based on big data learning, and in particular, an onboard computer intelligent AI test and maintenance system based on big data learning. Background Technology

[0002] The onboard computer is a key component of the missile system and the core of the seeker's information processing. With the continuous development and updating of electronic technology and integration processes, the onboard computer is becoming smaller and smaller, and the integration of components is becoming higher and higher. The efficiency of the original testing system can no longer meet the current mass production scale.

[0003] For example, patent document CN117315408A discloses a method and system for constructing a fault criterion model based on an AI target detection model. The method includes: acquiring historical operational data of the AI ​​target detection model, including historical fault data, to provide basic information for establishing the fault criterion model. Secondly, the historical fault data of the AI ​​target detection model is cleaned and integrated into a data fault set to ensure data quality and usability. Then, operational parameters are extracted from the data fault set, and key functional parameters are further screened. Subsequently, a functional model is established and fitting experiments are conducted to improve model performance. During this process, the functional dataset is processed to determine the optimal functional data, thereby improving the model's accuracy. Finally, the fault criterion model of the AI ​​target detection model is determined through the functional model, providing support for the reliability of the fault criterion model's establishment.

[0004] For example, patent document CN119762052A discloses a digital AI maintenance support system for armored equipment, including a fault information receiving module, an AI solution generation module, an intelligent auxiliary execution module, and a management feedback module. The fault information receiving module is used to receive fault information of armored equipment, the AI ​​solution generation module generates maintenance solutions based on the fault information, the intelligent auxiliary execution module outputs auxiliary information for the maintenance solutions, and the management feedback module manages data information during the maintenance process. This system can intelligently analyze and generate maintenance solutions based on fault information and control maintenance equipment to perform maintenance, thereby improving the automated maintenance capabilities of armored equipment.

[0005] However, existing testing systems have the following drawbacks:

[0006] 1. The size of the onboard computer has been reduced to about one-third of its original size. In the original design of the onboard computer data processing system, the tooling fixtures used for hardware testing were fastened with screws. Each disassembly and assembly required manual labor, which was not only inefficient but also could not meet the needs of mass production.

[0007] 2. The data processing of the onboard computer testing system adopts a linear processing mode for a single product. The system must complete the testing of all data such as frequency, spectrum, AD / DA of the first product before starting the testing process for the second product. During this process, when testing a certain data of the first product, other acquisition and testing modules are in standby mode, resulting in a serious waste of computing resources. Especially when facing a large number of test samples, this processing mode greatly extends the overall testing time.

[0008] 3. The data interpretation, analysis, organization, and summarization work in the onboard computer testing system mainly relies on manual labor. In practice, staff need to open the data test forms for each product one by one, manually judge whether the data meets the standards, and then carry out subsequent organization and summarization. In the past, when each batch consisted of only a dozen or so products, this manual processing method was still manageable. However, it is clearly inadequate to cope with the current situation where the number of batch products has increased significantly.

[0009] 4. The onboard computer testing system has accumulated a massive amount of test data during operation. However, the repair work for faulty products still mainly relies on manual operation. Summary of the Invention

[0010] To address the shortcomings of existing technologies, the purpose of this invention is to provide an onboard computer AI testing and maintenance system based on big data and capable of learning.

[0011] According to the present invention, an onboard computer AI testing and maintenance system based on big data learning is provided, comprising: a hardware data acquisition module and a software data testing, analysis and fault diagnosis module; the software data testing, analysis and fault diagnosis module is capable of inputting onboard data acquired by the hardware data acquisition module in parallel, and then analyzing, interpreting and summarizing the onboard data, and performing preliminary repairs on defective products.

[0012] Preferably, the hardware data acquisition module is used to test the onboard computer product and acquire onboard data; the hardware data acquisition module includes: a power supply control unit, a timing control unit, and a data acquisition unit.

[0013] Preferably, the power supply control unit is used to supply power to the onboard computer intelligent AI test and maintenance system; the power supply control unit includes: a ripple suppression component, which can effectively reduce the impact of ripple and noise on subsequent accuracy data and rate data acquisition, while improving the long-term stability of the system; the power supply control unit adopts an insert-type sliding groove design, with balanced insertion and removal forces, avoiding the bending, shrinkage, and breakage of the connector core pin due to uneven insertion and removal forces or insertion and removal shaking and twisting; the power supply control unit adopts a two-stage buffer overcurrent protection design. Specifically, when the load current exceeds the first warning line, the internal voltage regulator continues to work, and the onboard computer data processing system enters a half-load working mode; when the load current exceeds the second current warning line, the internal voltage regulator is interrupted, and the onboard computer data processing system stops working.

[0014] Preferably, the timing control unit is capable of handling cross-clock domains of 8 channels; the timing control unit employs multiple measures to solve the problems of metastability, clock skew and jitter, including: asynchronous FIFO buffering, bilateral handshake protocol confirmation, and clock domain crossover.

[0015] Preferably, the data acquisition unit is capable of sampling analog digital data, digital analog data, and temperature and humidity data; the data acquisition unit includes: a noise suppression component, a digital processing component, and a temperature drift compensation component, and the data acquisition unit can effectively improve the reliability of the acquired data; under the control of the data acquisition unit, the analog digital data, digital analog data, and temperature and humidity data are respectively transmitted to the noise suppression component, and after receiving the data from the noise suppression component, the digital processing component summarizes and sends it to the temperature drift compensation component.

[0016] Preferably, the software data testing and analysis fault diagnosis module is used to analyze, interpret, and summarize sampled data, and to perform preliminary repairs on defective products. The software data testing and analysis fault diagnosis module includes: a main control program component, a data testing and analysis component, and a data fault diagnosis AI component. After receiving parallel input data, the main control program component sends processed product data to the data testing and analysis component and sends an analysis of the fault cause and a preliminary repair suggestion to the data fault diagnosis AI component. After receiving processed product data from the main control program component, the data testing and analysis component returns correct data to the main control program component, simultaneously sends correct product data to the data fault diagnosis AI component, and obtains incorrect product data from the data fault diagnosis AI component. The data fault diagnosis AI component receives the analysis of the fault cause and the preliminary repair suggestion from the main control program component, receives correct product data from the data testing and analysis component, and sends a request to the data testing and analysis component to obtain incorrect product data.

[0017] Preferably, the main control program component is used for the overall operation of the system and the scheduling and coordination of various functions; the main control program component has the function of communicating and transmitting data with the onboard computer via CAN bus, and can test the working status of received instructions to determine whether the instruction reception is normal; the main control program component is responsible for global resource allocation, can monitor system resources in real time, and can perform computationally intensive data analysis when resources are surplus, which can effectively improve testing efficiency;

[0018] Preferably, the data testing and analysis component is capable of batch analysis, interpretation, and integration of collected data; the data testing and analysis component is capable of cascading calls to multiple software programs, including Matlab and Excel.

[0019] Preferably, the data test and analysis component invokes Excel through the C# main program, operates on Excel, automatically fills in the pass / fail criteria for various indicators, performs preliminary interpretation of the test data, filters out data of unqualified products, and records the numbers;

[0020] Preferably, the data fault diagnosis AI component is capable of deep learning of test big data, preliminary troubleshooting of data faults, and early condition-based maintenance suggestions for non-emerging faults; the data fault diagnosis AI component adopts a supervised learning mode; the data fault diagnosis AI component uses an algorithm combining logistic regression and decision trees, and can iteratively maintain learning; the data fault diagnosis AI component first optimizes the time of AI database retrieval through algorithms, and understands the user's query intent through natural language processing; the data fault diagnosis AI component includes a set of query decision instructions; the data fault diagnosis AI component first performs data integration, transformation, and anomaly detection through intelligent AI, and then analyzes abnormal data through regression algorithms to establish a fault tree including probability distribution, helping users make maintenance decisions; the fault tree includes: excessive AD error, AD device failure, AD device power supply abnormality, power supply ripple suppression module failure, digital signal processing algorithm abnormality, DC / DC abnormality, power chip abnormality, and voltage regulator abnormality; the data fault diagnosis AI component can issue early warnings and provide condition-based maintenance suggestions for abnormal and excessive data.

[0021] Preferably, under the control of the data fault diagnosis AI component, the input data undergoes natural language processing, keyword extraction, and a determination of whether it is legal natural language. If it is legal natural language, keyword retrieval is performed and mapped to the database primary key, while a logic tree is built. If the determination passes, a recursive logic processing mode is entered. In this mode, iterative learning is performed to adjust the branch weights and update the decision instruction set, and an action is taken. Then, the correctness of the action is determined again. If the action is correct, iterative learning is performed again to adjust the branch weights and update the decision instruction set. If the action is incorrect, the interpretation of the query is determined again. If the interpretation is determined to be incorrect before the recursive logic processing mode, the interpretation of the query is re-determined. If it is determined to be illegal natural language after the initial keyword extraction, the process ends directly.

[0022] This invention provides a missile-based computer AI testing and maintenance method based on big data learning, comprising:

[0023] Step S1: First, turn on the power switch, then perform a self-test of the device current, and finally start the power supply control unit and timing control unit to work;

[0024] Step S2: First, start the main control software, then run the one-click test function, and finally the hardware data acquisition module starts working, transmitting the acquired data to the main control program component through the controller LAN communication.

[0025] Step S3: Instruct the main control program component in step S2 to send the collected data to the software data test analysis fault diagnosis module, and call the data interpretation parameter settings;

[0026] Step S4: The software data testing and analysis fault diagnosis module analyzes and interprets the data according to the preset parameters, summarizes and outputs the qualified product data, draws the batch product data envelopment diagram, and finally feeds the qualified product data to the AI ​​for learning, and sends the unqualified products to the data fault diagnosis AI component.

[0027] Step S5: The data fault diagnosis AI component will combine past cases to analyze the causes of the fault phenomena, automatically generate a fault tree, troubleshoot and repair according to the fault tree, and finally give maintenance suggestions.

[0028] Step S6: The data fault diagnosis AI component analyzes the qualified product data, marks the data that does not exceed the calibrated correct data range as qualified, but is significantly different from other products in the same batch, analyzes the data, and provides repair suggestions based on the situation, combined with previous cases.

[0029] Step S7: Issue a data test report based on the condition-based maintenance recommendations in Step S6.

[0030] Preferably, step S4 includes:

[0031] Step S401: Perform data testing and analysis steps;

[0032] Step S402: Perform the data integration step;

[0033] Step S403: Perform the data output step;

[0034] Further, step S401 specifically refers to the following steps: assuming the first batch of weapons and equipment consists of 500 sets, firstly, the Excel file is opened through the C# main program and the relevant initialization settings are completed; then, the built-in IF function of Excel is used to set the implementation example to judge the validity of the data; finally, the For loop program is used to execute the operation 500 times, thereby realizing the judgment of all test data of these 500 sets of weapons and equipment and carrying out analysis and screening work.

[0035] Further, step S402 specifically refers to the following steps: First, the 500 Excel files of the 500 sets of products in step S401 are opened one by one using the corresponding code. Then, the key data is copied and summarized into a data summary table. Finally, the envelope and data distribution of the entire batch of test data are analyzed.

[0036] Further, step S403 specifically refers to drawing an envelope diagram from the data in the summary table compiled in step S402 using C# code.

[0037] This invention provides an onboard computer AI testing and maintenance device based on big data learning, which significantly improves testing efficiency and performance through big data learning onboard computer AI testing and maintenance.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] 1. This invention utilizes tooling for batch product installation, which can significantly reduce testing time and improve testing efficiency. It effectively solves the drawbacks of traditional screw fastening of single sets of products and designs an insert-type sliding groove tooling, which features quick and convenient product insertion and removal, uniform force distribution, and supports hot-swap functionality, successfully avoiding the risk of misoperation during live insertion and removal.

[0040] 2. Through high-precision power supply and anti-interference circuit design, this invention adopts multiple hardware and software measures at the input and output ends, enabling the simultaneous acquisition of high-precision and high-speed data from multiple products, thus achieving a significant improvement in testing efficiency and performance.

[0041] 3. This invention uses intelligent AI to automatically analyze test data and can conveniently query various indicators of batch products in natural language, including data distribution and comparison with previous batches of product data. This not only improves data accuracy and shortens data processing time, but also assists in handling subsequent repair work of defective products.

[0042] 4. This invention adopts a multi-channel parallel working mode and uses multiple measures such as asynchronous FIFO buffering, bilateral handshake protocol confirmation, and clock domain crossover (CDC) to efficiently utilize idle resources and achieve simultaneous processing of multiple products.

[0043] 5. This invention has advantages such as high integration, fast testing, and intelligent data analysis, making it particularly suitable for testing large batches of products. Attached Figure Description

[0044] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0045] Figure 1 This is a schematic diagram of the architecture of the missile-borne computer AI test and maintenance system based on big data learning proposed in this invention.

[0046] Figure 2 This is a schematic diagram of the fixing fixture for the power supply control unit in the missile-borne computer AI test and maintenance system based on big data learning proposed in this invention.

[0047] Figure 3This is a schematic diagram of the software data testing and analysis fault diagnosis module in the missile-borne computer AI testing and maintenance system based on big data learning proposed in this invention.

[0048] Figure 4 This is a flowchart illustrating the missile-borne computer AI testing and maintenance system based on big data learning proposed in this invention.

[0049] Figure 5 This is a schematic diagram of the learning logic of the missile-borne computer AI test and maintenance system based on big data learning proposed in this invention.

[0050] Figure 6 This is a schematic diagram of the maintenance fault tree of the missile-borne computer AI test and maintenance system based on big data learning proposed in this invention.

[0051] The figure shows:

[0052] AD stands for analog-to-digital conversion;

[0053] DA stands for Digital-to-Analog Conversion;

[0054] DC stands for direct current. Detailed Implementation

[0055] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0056] This invention provides a missile-borne computer AI testing and maintenance system based on big data learning, comprising: a hardware data acquisition module and a software data testing, analysis, and fault diagnosis module; the software data testing, analysis, and fault diagnosis module is as shown in the attached diagram. Figure 3 As shown, the system can input the onboard data collected by the hardware data acquisition module in parallel, and then analyze, interpret, and summarize the onboard data, and perform preliminary repairs on defective products.

[0057] The hardware data acquisition module is used to test the onboard computer product and collect onboard data.

[0058] The hardware data acquisition module includes: a power supply control unit, a timing control unit, and a data acquisition unit;

[0059] The hardware data acquisition module, written in C#, is capable of acquiring all data indicators from the onboard computer.

[0060] The power supply control unit is used to supply power to the onboard computer intelligent AI test and maintenance system.

[0061] The power supply control unit includes a ripple suppression component, which can effectively reduce the impact of ripple and noise on subsequent accuracy data and rate data acquisition, while improving the stability of the system during long-term operation.

[0062] The power supply control unit adopts a two-stage buffer overcurrent protection design. Specifically, when the load current exceeds the first warning line, the internal voltage regulator (LDO) continues to work, and the onboard computer data processing system enters a half-load working mode. When the load current exceeds the second current warning line, the internal voltage regulator (LDO) is interrupted, and the onboard computer data processing system stops working.

[0063] The power supply control unit, as shown in the attached document Figure 2 As shown, the insertion-type sliding groove design ensures balanced insertion and removal forces, preventing the connector pins from warping, shrinking, or breaking due to uneven insertion and removal forces or shaking and twisting.

[0064] The timing control unit is capable of handling cross-clock domain operations for 8 channels;

[0065] The timing control unit employs multiple measures to address metastability, clock skew, and jitter issues. These measures include: asynchronous FIFO buffering (first-in, first-out mechanism), bilateral handshake protocol confirmation, and clock domain crossing (CDC).

[0066] The data acquisition unit is capable of performing analog-to-digital (AD) data sampling, digital-to-analog (DA) data sampling, and temperature and humidity data sampling.

[0067] The data acquisition unit includes: a noise suppression component, a digital processing component, and a temperature drift compensation component. The data acquisition unit can effectively improve the reliability of the acquired data.

[0068] Preferably, as shown in the appendix Figure 1 As shown, under the control of the data acquisition unit, analog-to-digital (AD) data, digital-to-analog (DA) data, and temperature and humidity data are transmitted to the noise suppression component, respectively. After receiving the data from the noise suppression component, the digital processing component summarizes and sends it to the temperature drift compensation component.

[0069] The software data testing and analysis fault diagnosis module is used to analyze, interpret, and summarize the sampled data, and to perform preliminary repairs on unqualified products.

[0070] The software data testing, analysis, and fault diagnosis module includes: a main control program component, a data testing and analysis component, and a data fault diagnosis AI component.

[0071] Preferably, as shown in the appendix Figure 1 As shown, after receiving the parallel input data, the main control program component sends the processed product data to the data test and analysis component, and sends the analysis of the fault cause and the preliminary repair suggestions to the data fault diagnosis and repair AI component.

[0072] After receiving and processing product data from the main control program component, the data test and analysis component returns correct data to the main control program component, and simultaneously sends correct product data to the data fault diagnosis AI component and obtains incorrect product data from the data fault diagnosis AI component.

[0073] The data fault diagnosis AI component receives and analyzes the cause of the fault from the main control program component and provides preliminary repair suggestions. It also receives correct product data from the data test and analysis component and feeds it to the AI, while sending a request to the data test and analysis component to obtain incorrect product data.

[0074] The main control program component is used for the overall operation of the system and the scheduling and coordination of various functions.

[0075] The main control program component has the function of communicating and transmitting data with the onboard computer via the CAN bus, and can test the working status of receiving instructions to determine whether the instruction reception is normal.

[0076] The main control program component is responsible for global resource allocation and can monitor system resources in real time. When resources are in surplus, it can perform computationally intensive data analysis, which can effectively improve testing efficiency.

[0077] The data testing and analysis component can analyze, interpret, and integrate the collected data in batches;

[0078] The data testing and analysis component can enable cascading calls to multiple software programs, including Matlab and Excel.

[0079] The data testing and analysis component invokes Excel through the C# main program, performs operations on Excel, automatically fills in the pass / fail criteria for various indicators, performs preliminary interpretation of the test data, filters out data of unqualified products, and records the numbers.

[0080] The data fault diagnosis and repair AI component is capable of deep learning of test big data, preliminary troubleshooting of data faults, and early condition-based maintenance suggestions for faults that have not yet occurred.

[0081] The data fault diagnosis AI component adopts a supervised learning mode;

[0082] The data fault diagnosis AI component uses an algorithm that combines logistic regression and decision tree, and can continuously learn through self-iteration.

[0083] The data fault diagnosis AI component first optimizes the time for AI to retrieve data from the database through algorithms, and understands the user's query intent through natural language processing (NLP). The data fault diagnosis AI component includes a set of query decision instructions.

[0084] Preferably, as shown in the appendix Figure 5 As shown, under the control of the data fault diagnosis AI component, the input data undergoes natural language processing (NLP) to extract keywords and determine whether it is legal natural language. If it is legal natural language, keyword retrieval is performed and mapped to the database primary key, while a logic tree is built. If the judgment is successful, the system enters a recursive logic processing mode. In this mode, iterative learning is performed to adjust the branch weights and update the decision instruction set, and an action is taken. The correctness of the action is then judged again. If the action is correct, iterative learning is performed again to adjust the branch weights and update the decision instruction set. If the action is incorrect, the system judges again whether the query is correctly understood. If the query is judged to be incorrect before the recursive logic processing mode, the system will also judge whether the query is correctly understood again. If the initial keyword extraction determines that the data is not legal natural language, the process ends directly.

[0085] The data fault diagnosis AI component first integrates, transforms, and detects anomalies through intelligent AI, and then analyzes the abnormal data through regression algorithms to establish a fault tree including probability distribution, helping users make maintenance decisions.

[0086] The fault tree, as shown in the appendix Figure 6 As shown, the following are possible causes: excessive AD error, AD device failure, abnormal power supply to AD device, failure of power supply ripple suppression module, abnormal digital signal processing algorithm, abnormal DC / DC converter, abnormal power supply chip, and abnormal voltage regulator.

[0087] The data fault detection AI component can issue early warnings for abnormal and excessive data and provide repair suggestions as appropriate.

[0088] This invention provides a missile-based computer AI testing and maintenance method based on big data learning, comprising:

[0089] Step S1: First, turn on the power switch, then perform a self-test of the device current, and finally start the power supply control unit and timing control unit to work;

[0090] Step S2: First, start the main control software, then run the "one-click test" function, and finally the hardware data acquisition module starts working, transmitting the acquired data to the main control program component through the controller LAN communication.

[0091] Step S3: Instruct the main control program component in step S2 to send the collected data to the software data test analysis fault diagnosis module, and call the data interpretation parameter settings;

[0092] Step S4: The software data testing and analysis fault diagnosis module analyzes and interprets the data according to the preset parameters, summarizes and outputs the qualified product data, draws the batch product data envelopment diagram, and finally feeds the qualified product data to the AI ​​for learning, and sends the unqualified products to the data fault diagnosis AI component.

[0093] Step S5: The data fault diagnosis AI component will combine past cases to analyze the causes of the fault phenomena, automatically generate a fault tree, troubleshoot and repair according to the fault tree, and finally give maintenance suggestions.

[0094] Step S6: The data fault diagnosis AI component analyzes the qualified product data, marks the data that does not exceed the calibrated correct data range as "qualified" but is significantly different from other products in the same batch, analyzes the data, and provides repair suggestions based on the situation, combined with previous cases.

[0095] Step S7: Issue a data test report based on the condition-based maintenance recommendations in Step S6.

[0096] Step S4 includes:

[0097] Step S401: Perform data testing and analysis steps;

[0098] Step S402: Perform the data integration step;

[0099] Step S403: Perform the data output step;

[0100] Further, step S401 specifically refers to the following steps: assuming the first batch of weapons and equipment consists of 500 sets, firstly, the Excel file is opened through the C# main program and the relevant initialization settings are completed; then, the implementation example is set using the built-in IF function of Excel to judge the validity of the data; finally, the operation is executed 500 times through a For loop program to judge all the test data of these 500 sets of weapons and equipment, and further analysis and screening work is carried out.

[0101] Further, step S402 specifically refers to first opening the 500 Excel files of the 500 sets of products in step S401 one by one through the corresponding code, then copying and summarizing the key data into a data master table, and finally analyzing the envelope and data distribution of the entire batch of test data.

[0102] Further, step S403 specifically refers to drawing an envelope diagram from the data in the summary table compiled in step S402 using C# code.

[0103] Preferably, as shown in the appendix Figure 4 As shown, in a specific scenario, the cabinet power is first turned on to start the equipment, followed by data acquisition. After acquisition, the interpretation conditions for data indicators are entered by clicking the "Configure" button, thus entering the data analysis stage. Clicking the "Analyze" button then uses Matlab software algorithms to perform in-depth data analysis. The analyzed data undergoes a qualification assessment. If the data is qualified, it directly enters the data integration stage; otherwise, if the data is unqualified, the erroneous data is stored in a designated directory for subsequent investigation and correction. After data integration is complete, clicking the "Summary" button uses Excel software to summarize all data into a single table, achieving centralized data management and display. Finally, a detailed data report and an intuitive envelope diagram are generated based on the summarized data, providing strong support for decision-making, and the entire process concludes.

[0104] The present invention also provides a missile-on-board computer AI testing and maintenance system based on big data learning. The missile-on-board computer AI testing and maintenance system based on big data learning can be implemented by executing the process steps of the missile-on-board computer AI testing and maintenance method based on big data learning. That is, those skilled in the art can understand the missile-on-board computer AI testing and maintenance method based on big data learning as a preferred embodiment of the missile-on-board computer AI testing and maintenance system based on big data learning.

[0105] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A missile-based computer AI testing and maintenance system based on big data learning, characterized in that, include: Hardware data acquisition module; software data testing, analysis, and fault diagnosis module; The software data testing and analysis fault diagnosis module can input the onboard data collected by the hardware data acquisition module in parallel, and then analyze, interpret, and summarize the onboard data, and perform preliminary repairs on unqualified products. The hardware data acquisition module is used to test the onboard computer product and collect onboard data. The hardware data acquisition module includes: a power supply control unit, a timing control unit, and a data acquisition unit; The power supply control unit is used to supply power to the onboard computer intelligent AI test and maintenance system. The power supply control unit includes a ripple suppression component, which effectively reduces the impact of ripple and noise on subsequent accuracy and rate data acquisition, while improving the long-term stability of the system. The power supply control unit adopts an insert-type sliding groove design, ensuring balanced insertion and removal forces and preventing connector pin distortion, shrinkage, and breakage due to uneven insertion and removal forces or shaking / twisting. The power supply control unit employs a two-stage buffer overcurrent protection design. Specifically, when the load current exceeds the first warning line, the internal voltage regulator continues to operate, and the onboard computer data processing system enters a half-load operating mode. When the load current exceeds the second current warning line, the internal voltage regulator is interrupted, and the onboard computer data processing system stops operating.

2. The onboard computer AI testing and maintenance system based on big data learning as described in claim 1, characterized in that, The timing control unit is capable of handling cross-clock domain operations for 8 channels; The timing control unit employs multiple measures to address metastability, clock skew, and jitter issues. These measures include: asynchronous FIFO buffering, bilateral handshake protocol confirmation, and clock domain crossover.

3. The onboard computer AI testing and maintenance system based on big data learning as described in claim 1, characterized in that, The data acquisition unit is capable of performing analog-to-digital data sampling, digital-to-analog data sampling, and temperature and humidity data sampling. The data acquisition unit includes: a noise suppression component, a digital processing component, and a temperature drift compensation component. The data acquisition unit can effectively improve the reliability of the acquired data. Under the control of the data acquisition unit, analog digital data, digital analog data, and temperature and humidity data are transmitted to the noise suppression component. After receiving the data from the noise suppression component, the digital processing component summarizes and sends it to the temperature drift compensation component.

4. The onboard computer AI testing and maintenance system based on big data learning as described in claim 1, characterized in that, The software data testing and analysis fault diagnosis module is used to analyze, interpret, and summarize sampled data, and to perform preliminary repairs on defective products. The module includes a main control program component, a data testing and analysis component, and a data fault diagnosis AI component. After receiving parallel input data, the main control program component sends processed product data to the data testing and analysis component and sends analysis of fault causes and preliminary repair suggestions to the data fault diagnosis AI component. After receiving processed product data from the main control program component, the data testing and analysis component returns correct data to the main control program component, simultaneously feeding correct product data to the AI ​​component and obtaining incorrect product data from the AI ​​component. The data fault diagnosis AI component receives analysis of fault causes and preliminary repair suggestions from the main control program component, receives correct product data from the data testing and analysis component, feeds it to the AI ​​component, and sends a request to the data testing and analysis component to obtain incorrect product data. The main control program component is used for the overall operation of the system and the scheduling and coordination of various functions. The main control program component has the function of communicating and transmitting data with the onboard computer via CAN bus, and can test the working status of received instructions to determine whether the instruction reception is normal. The main control program component is responsible for global resource allocation, can monitor system resources in real time, and can perform computationally intensive data analysis when resources are surplus, which can effectively improve testing efficiency. The data testing and analysis component can analyze, interpret, and integrate the collected data in batches; the data testing and analysis component can realize the cascading call of multiple software programs, including Matlab and Excel. The data testing and analysis component invokes Excel through the C# main program, performs operations on Excel, automatically fills in the pass / fail criteria for various indicators, performs preliminary interpretation of the test data, filters out data of unqualified products, and records the numbers. The data fault diagnosis and repair AI component is capable of deep learning of test big data, preliminary troubleshooting of data faults, and early condition-based maintenance suggestions for non-emerging faults. The component employs a supervised learning model and uses an algorithm combining logistic regression and decision trees, enabling iterative learning. It first optimizes the time for AI database retrieval through algorithms and understands user query intent through natural language processing, including a set of query decision instructions. The component first integrates, transforms, and detects anomalies through intelligent AI, then analyzes abnormal data using regression algorithms to establish a fault tree including probability distributions, helping users make maintenance decisions. The fault tree includes: excessive AD error, AD device failure, AD device power supply abnormality, power ripple suppression module failure, digital signal processing algorithm abnormality, DC / DC abnormality, power chip abnormality, and voltage regulator abnormality. The component can issue early warnings and provide condition-based maintenance suggestions for abnormal and excessive data.

5. The onboard computer AI testing and maintenance system based on big data learning as described in claim 4, characterized in that, Under the control of the data fault diagnosis AI component, the input data undergoes natural language processing, keyword extraction, and a determination of whether it is legal natural language. If it is legal natural language, keyword retrieval is performed and mapped to the database primary key, while a logic tree is built. If the determination is successful, a recursive logic processing mode is entered. In this mode, iterative learning is performed to adjust the branch weights and update the decision instruction set, and an action is taken. The correctness of the action is then determined. If the action is correct, iterative learning is performed again to adjust the branch weights and update the decision instruction set. If the action is incorrect, the interpretation of the query is determined again. If the interpretation is determined to be incorrect before the recursive logic processing mode, the interpretation of the query is re-determined. If the initial keyword extraction determines that it is not legal natural language, the process ends directly.

6. A missile-based computer AI testing and maintenance method based on big data learning, employing the missile-based computer AI testing and maintenance system based on big data learning as described in any one of claims 1 to 5, characterized in that, include: Step S1: First, turn on the power switch, then perform a self-test of the device current, and finally start the power supply control unit and timing control unit to work; Step S2: First, start the main control software, then run the one-click test function, and finally the hardware data acquisition module starts working, transmitting the acquired data to the main control program component through the controller LAN communication. Step S3: Instruct the main control program component in step S2 to send the collected data to the software data test analysis fault diagnosis module, and call the data interpretation parameter settings; Step S4: The software data testing and analysis fault diagnosis module analyzes and interprets the data according to the preset parameters, summarizes and outputs the qualified product data, draws the batch product data envelopment diagram, and finally feeds the qualified product data to the AI ​​for learning, and sends the unqualified products to the data fault diagnosis AI component. Step S5: The data fault diagnosis AI component will combine past cases to analyze the causes of the fault phenomena, automatically generate a fault tree, troubleshoot and repair according to the fault tree, and finally give maintenance suggestions. Step S6: The data fault diagnosis AI component analyzes the qualified product data, marks the data that does not exceed the calibrated correct data range as qualified, but is significantly different from other products in the same batch, analyzes the data, and provides repair suggestions based on the situation, combined with previous cases. Step S7: Issue a data test report based on the condition-based maintenance recommendations in Step S6.

7. The onboard computer AI testing and maintenance method based on big data learning as described in claim 6, characterized in that, Step S4 includes: Step S401: Perform data testing and analysis steps; Step S402: Perform the data integration step; Step S403: Perform the data output step; Further, step S401 specifically refers to the following steps: assuming the first batch of weapons and equipment consists of 500 sets, firstly, the Excel file is opened through the C# main program and the relevant initialization settings are completed; then, the built-in IF function of Excel is used to set the implementation example to judge the validity of the data; finally, the For loop program is used to execute the operation 500 times, thereby realizing the judgment of all test data of these 500 sets of weapons and equipment and carrying out analysis and screening work. Further, step S402 specifically refers to the following steps: First, the 500 Excel files of the 500 sets of products in step S401 are opened one by one using the corresponding code. Then, the key data is copied and summarized into a data summary table. Finally, the envelope and data distribution of the entire batch of test data are analyzed. Further, step S403 specifically refers to drawing an envelope diagram from the data in the summary table compiled in step S402 using C# code.

8. A missile-based computer AI testing and maintenance device with big data learning capability, characterized in that, The missile-on-board computer AI test and maintenance system based on big data learning, as described in any one of claims 1 to 5, achieves a significant improvement in testing efficiency and performance through big data learning-based missile-on-board computer AI test and maintenance.

Images