Detection device for carried ammunition to be launched and mass big data analysis method and system

Abstract

The invention relates to the technical field of ammunition detection, in particular to a portable to-be-launched ammunition detection device and a mass big data analysis method and system. The quality monitoring system for the carried to-be-launched ammunition for the mountain combat is divided into an upper computer and a lower computer. The upper computer comprises a missile-borne test device, a lower computer device in a cockpit, a sound-light alarm device, a GPS / Beidou antenna and an LORA antenna, the missile-borne test device is installed in a modified missile body, a nine-axis sensor is tightly attached to a fuze part so that the state information of the fuze can be conveniently sensed, and a temperature and humidity sensor is also installed in the missile body; the GPS / Beidou / Beidou antenna is connected with lower computer equipment in a cockpit to measure the speed and position information of a chariot at any time, the lORA antenna is connected with the lower computer to provide communication between missile-borne test equipment, and the lower computer provides speed and time calibration information for the missile-borne test equipment. And the missile-borne test equipment immediately backs up the measurement data of the sensor to the lower computer through the LORA antenna. The upper computer realizes online monitoring and evaluates acousto-optic early warning in real time; and the lower computer carries out statistical analysis on the data offline, realizes dynamic fault tree and Bayesian reliability classification prediction, and provides theoretical analysis and engineering support for an ammunition damage failure model.

Description

Technical Field

[0001] The present invention relates to the technical field of ammunition detection, in particular to a portable ready-to-fire ammunition detection device, and a quality big data analysis method and system. Background Art

[0002] At present, the research on ammunition transportation safety assessment methods is relatively scattered and lacks systematic theoretical research. In addition, the existing research is not mature enough in practical application. The main manifestations are that the model selection is too simplified and it is impossible to objectively evaluate complex practical problems; some of the selected evaluation methods are too subjective and it is difficult to obtain objective and effective evaluation results; some evaluation methods have certain limitations, so that the corresponding research results have not been able to play a good role in practice. There is also a lack of system and depth in the exploration of failure modes and hazard mechanisms, which has led to the failure to provide strong support for the development of ammunition protection, guarantee, and assessment theories and technologies, and further research is needed. The following problems need to be solved urgently: (1) Analyze the factors affecting the safety and reliability of ammunition transportation, identify the weak links in the safety of ammunition, and discuss the impact of shock and vibration on the safety and reliability of ammunition; (2) Under the conditions of high altitude and cold mountainous areas, understand the impact of typical situations encountered during the transportation of ammunition on the safety of ammunition; (3) According to the safety and reliability requirements of ammunition and the mechanical environment conditions of ammunition, conduct damage model analysis. In summary, under the high altitude and cold environment, the damage caused by vibration impact to the fuze of ammunition carried by ammunition is mainly investigated. The stress influence of vibration impact load on the fuze can be evaluated to predict the vibration quality of ammunition transportation. Summary of the invention

[0003] The portable ammunition is carried uniformly according to the carrying capacity of personnel, weapons and combat vehicles, the tactical and technical performance of weapons, and the combat needs of a certain period of time. The amount of ammunition carried is an important measure to enable the troops to maintain combat readiness and have a certain combat capability. Ammunition is flammable and explosive. When it is exposed to high temperature, friction, vibration, collision, exposure and other factors, it is very easy to cause unsafe accidents. Real-time safety monitoring and environmental monitoring of portable ammunition is one of the important strategies to ensure the safety of portable ammunition and good performance of ammunition. At present, the troops have a large number of ammunition and many types, insufficient storage information on the plateau, weak carrying monitoring capabilities, and backward informationization. There is an urgent need to improve the monitoring and management of portable ammunition. The types of portable ammunition and dynamic and static management are comprehensively considered. The main influencing parameters mainly include four aspects: temperature, humidity, posture, impact and collision environmental stress information. The weak links of the portable ammunition in high altitude and cold areas are tested and demonstrated, and the reliability decline curve is fitted and loaded into the physical state acquisition terminal based on the typical ammunition as the prototype, so as to design a quality monitoring system for portable ammunition in high altitude and cold mountain combat. The system mainly includes sensor modules, microprocessors, communication modules, and life and safety assessment methods to realize intelligent monitoring of portable ammunition, and can grasp and record the carrying information, environmental stress information and other data of portable ammunition at any time, so as to realize real-time data collection and dynamic system management. Therefore, the environmental stress experienced by the combat portable ammunition is monitored throughout the process, so that the ammunition management personnel of the troops can judge whether the ammunition can continue to be carried and fired in time, which can greatly improve the reliability of the use of ammunition technology and combat application.

[0004] The ammunition on standby refers to the ammunition that is carried with tanks, armored vehicles, self-propelled artillery, mortars and other combat vehicles after the technical inspection before shooting, and is ready for firing. The climatic conditions of high-altitude and cold mountain operations are harsh, the geographical environment is complex, the combat conditions are difficult, and the firepower is frequently used, which puts higher requirements on the reliable combat effectiveness of ammunition. In the ammunition support tasks of high-altitude and cold field training and special combat exercises in recent years, it was found that border operations have the characteristics of diverse deployment opportunities and various types of action styles. The ammunition on standby without the protection of internal and external packaging is more susceptible to environmental factors after tactical maneuvers, causing changes in ammunition performance, and ammunition shooting accidents such as close bullets, chamber retention, misfires, non-explosion, half-explosion bullets, chamber explosions, and muzzle explosions may occur, and even accidental shooting and explosions may cause casualties, which seriously affect the combat effectiveness of ammunition and restrict the generation of combat effectiveness.

[0005] Taking ammunition as the research object, through the analysis of test data and the reasons for the failure of previous ammunition, the components or parts with the highest failure frequency, that is, the weak links of ammunition, are found. Then, the failure modes of the key components, assemblies and components of the typical ammunition are simulated, the reliability decline curve is fitted and loaded into the physical state acquisition terminal based on the prototype of the typical ammunition, and the ammunition is tactically maneuvered with the combat vehicle with the full ammunition on standby. The monitoring device is composed of a data acquisition module, a microprocessor, a communication module, a storage module, a power management module and a host computer. It has the function of real-time monitoring of the physical state of ammunition and recording the environmental stress of the combat carrying and standby and the corresponding time when the alarm exceeds the failure threshold. During the carrying and combat, it can not only send the sound and light alarm information of the reliability decline of ammunition to the combat personnel, but also save the physical state information of the failure threshold, which is convenient for professional and technical personnel to conduct a comprehensive analysis of the weak links of the ammunition on the vehicle and the reliability design of the ammunition on standby, and provide strong theoretical guidance and data support for the use of weapons and ammunition, performance and combat effectiveness.

[0006] The present invention solves the deficiencies of the prior art and provides a device and system specifically for monitoring portable ammunition ready to be fired. The following technical solutions are adopted to solve the deficiencies of the prior art: The portable ready-to-fire ammunition detection device comprises an upper computer and a lower computer; the upper computer is connected to the ready-to-fire ammunition body, the lower computer is connected to the cockpit, and the upper computer and the lower computer are connected through antenna signals; the upper computer is divided into an upper fixed frame assembly and a lower fixed frame assembly, and the upper fixed frame assembly and the lower fixed frame assembly are hollow cylindrical; the upper fixed frame assembly is provided with a body threaded section, and the upper fixed frame assembly and the lower fixed frame assembly are connected by threaded sleeves; the upper computer is connected to the ready-to-fire ammunition body through the body threaded section.

[0007] The upper fixed frame assembly is provided with a clamping block, the body of the ready-to-fire ammunition is provided with a clamping groove, and the clamping block is buckledly connected with the clamping groove.

[0008] The upper fixed frame assembly is threadedly connected to a support sleeve, and the support sleeve is threadedly adjusted to a telescopic position with the upper fixed frame assembly; the other end of the support sleeve is in contact with the inner wall of the ready-to-fire ammunition.

[0009] The upper fixed frame assembly is provided with a plurality of support sleeves, and the plurality of support sleeves are evenly distributed along the circumferential direction of the outer edge of the upper fixed frame assembly.

[0010] The support sleeve is provided with a threading hole and a cable bayonet; the threading hole passes through the support sleeve, and the cable bayonet is communicated with the threading hole.

[0011] The portable ammunition detection system comprises the portable ammunition detection device, a GPS Beidou module, a wireless signal module, an acoustic and light alarm module, a Tyep-c data transmission module and a sensor module; the lower computer of the portable ammunition detection device is provided with a GPS Beidou module, a wireless signal module, an acoustic and light alarm module and a Tyep-c data transmission module; the GPS Beidou module is used for vehicle positioning, the acoustic and light alarm module is used for data abnormality alarm, and the Tyep-c data transmission module is used for data transmission and downloading of the lower computer; the upper computer of the portable ammunition detection device is provided with a sensor module and a Tyep-c data transmission module, and the sensor module is divided into an MCU processor, a nine-axis sensor, and a temperature and humidity sensor; the MCU processor is connected with the nine-axis sensor and the temperature and humidity sensor for projectile data collection; the Tyep-c data transmission module is used for data transmission and downloading of the upper computer; the wireless signal module is used for wireless data connection between the upper computer and the lower computer, and transmits the sensor module data of the projectile upper computer to the lower computer.

[0012] The upper computer sensor module completes the measurement of attitude angle, acceleration, temperature and humidity, and transmits the data to the lower computer through the LORA Internet of Things antenna, and leaves a type-c interface for direct reading of the data collected by the upper and lower computers; the current missile attitude angle, acceleration, temperature and humidity data are monitored in real time through wireless communication. When the above real-time monitoring data reaches the threshold, the sound and light alarm module is triggered to give an alarm prompt.

[0013] The quality big data analysis method of the portable ready-to-fire ammunition detection system includes the following steps: S1. Data preparation: All data files need to be read from the host computer (1) and the slave computer (2); S2. Data cleaning: Clean abnormal data, re-screen, sort, and remove duplicates, and delete abnormal or missing values; S3. Descriptive statistical analysis: Use big data models for data analysis; S4. Data visualization: Use the drawing functions of the big data model to draw histograms, scatter plots, and pie charts of the data; S5, Probability distribution: Calculate the probability density function PDF and cumulative distribution function CDF of the data; S6. Hypothesis test: Use the KW hypothesis test function to perform hypothesis test on the data to test whether there is a significant difference between the data; S7, regression analysis: The regression analysis function used establishes a linear or nonlinear relationship model between data and calculates the parameters of the model; S8, Model verification and optimization: Based on the analysis results, the model is optimized and verified using an iterative optimization function; S9. Prediction analysis: perform Bayesian reliability calculations, predict status, and analyze qualitative and quantitative results; save the analysis results output to a file or a graphical user interface GUI.

[0014] The beneficial effect of the present invention is that the quality monitoring system of the portable ammunition for combat in high-altitude and cold mountainous areas is divided into two parts: the upper computer and the lower computer. The upper computer realizes online monitoring and real-time evaluation of sound and light warning; the lower computer performs statistical analysis on the data offline, realizes dynamic fault tree and Bayesian reliability classification prediction, and provides theoretical analysis and engineering support for the ammunition damage failure model.

[0015] The upper computer includes missile-borne test equipment, lower computer equipment in the cockpit, sound and light alarm equipment, GPS / Beidou antenna and LORA antenna. The missile-borne test equipment is installed in the modified warhead, in which the nine-axis sensor is close to the fuze part to facilitate the sensing of the fuze status information, and the temperature and humidity sensor is also installed in the warhead; the GPS / Beidou / Beidou antenna is connected to the lower computer equipment in the cockpit to measure the speed and position information of the tank at any time, and the LORA antenna is connected to the lower computer to provide communication between the missile-borne test equipment. The lower computer provides speed and time calibration information for the missile-borne test equipment. The missile-borne test equipment backs up the sensor measurement data to the lower computer in real time through the LORA antenna. The data of the missile-borne test equipment and the lower computer equipment in the cockpit can be read through the type-C interface. The data of the lower computer equipment in the cockpit can be read at any time by connecting the USB flash drive through an external Type-C data cable, while the missile-borne test equipment can only read the data after the mission is completed or when it is convenient to disassemble the modified bomb. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of the structure of the detection device for carrying ready-to-fire ammunition; Figure 2 This is a schematic diagram of the structure of the carrying ready-to-fire ammunition detection system; Figure 3 It is a schematic diagram of the combination of the detection device for carrying ready-to-fire ammunition and the projectile body; Figure 4 It is a schematic diagram of the structure of the upper fixed frame assembly; Figure 5 Schematic diagram of the supporting casing structure. DETAILED DESCRIPTION

[0017] The portable ready-to-fire ammunition detection device comprises an upper computer 1 and a lower computer 2; the upper computer 1 is connected to the ready-to-fire ammunition body, the lower computer 2 is connected to the cockpit, and the upper computer 1 and the lower computer 2 are connected through antenna signals; the upper computer 1 is divided into an upper fixed frame assembly 101 and a lower fixed frame assembly 102, and the upper fixed frame assembly 101 and the lower fixed frame assembly 102 are hollow cylindrical; the upper fixed frame assembly 101 is provided with a body threaded section 108, and the upper fixed frame assembly 101 and the lower fixed frame assembly 102 are threadedly connected; the upper computer 1 is connected to the ready-to-fire ammunition body through the body threaded section 108.

[0018] The upper fixed frame assembly 101 is provided with a clamping block 103, and the body of the ready-to-fire ammunition is provided with a clamping groove 104, and the clamping block 103 is snap-connected with the clamping groove 104. The upper fixed frame assembly 101 is threadedly connected with a support sleeve 105, and the support sleeve 105 is adjusted to a telescopic position with the upper fixed frame assembly 101 through a thread; the other end of the support sleeve 105 is in contact with the inner wall of the ready-to-fire ammunition. The upper fixed frame assembly 101 is provided with a plurality of support sleeves 105, and the plurality of support sleeves 105 are evenly distributed along the circumferential direction of the outer edge of the upper fixed frame assembly 101. The support sleeve 105 is provided with a threading hole 106 and a cable bayonet 107; the threading hole 106 passes through the support sleeve 105, and the cable bayonet 107 is connected with the threading hole 106. It is convenient to connect various devices and sensor cables. Among them, various devices and sensors are powered by batteries. Batteries can be set in the upper fixed frame assembly 101 and the lower fixed frame assembly 102. Battery installation and wiring are conventional technologies in this field and will not be repeated here.

[0019] The portable ammunition detection system comprises the portable ammunition detection device, a GPS Beidou module, a wireless signal module, an acoustic and light alarm module, a Tyep-c data transmission module and a sensor module; the lower computer 2 of the portable ammunition detection device is provided with a GPS Beidou module, a wireless signal module, an acoustic and light alarm module and a Tyep-c data transmission module; the GPS Beidou module is used for vehicle positioning, the acoustic and light alarm module is used for data abnormality alarm, and the Tyep-c data transmission module is used for data transmission and downloading of the lower computer 2; the upper computer 1 of the portable ammunition detection device is provided with a sensor module and a Tyep-c data transmission module, and the sensor module is divided into an MCU processor, a nine-axis sensor, and a temperature and humidity sensor; the MCU processor is connected with the nine-axis sensor and the temperature and humidity sensor for projectile data collection; the Tyep-c data transmission module is used for data transmission and downloading of the upper computer 1; the wireless signal module is used for wireless data connection between the upper computer 1 and the lower computer 2, and transmits the sensor module data of the projectile upper computer 1 to the lower computer 2. The upper computer sensor module completes the measurement of attitude angle, acceleration, temperature and humidity, and transmits the data to the lower computer through the LORA Internet of Things antenna, and leaves a type-c interface for direct reading of the data collected by the upper and lower computers; the current missile attitude angle, acceleration, temperature and humidity data are monitored in real time through wireless communication. When the above real-time monitoring data reaches the threshold, the sound and light alarm module is triggered to give an alarm prompt.

[0020] The quality big data analysis method of the portable ready-to-fire ammunition detection system includes the following steps: S1. Data preparation: All data files need to be read from the upper computer 1 and the lower computer 2; S2. Data cleaning: Clean abnormal data, re-screen, sort, and remove duplicates, and delete abnormal or missing values; S3. Descriptive statistical analysis: Use big data models for data analysis; S4. Data visualization: Use the drawing functions of the big data model to draw histograms, scatter plots, and pie charts of the data; S5, Probability distribution: Calculate the probability density function PDF and cumulative distribution function CDF of the data; S6. Hypothesis test: Use the KW hypothesis test function to perform hypothesis test on the data to test whether there is a significant difference between the data; S7, regression analysis: The regression analysis function used establishes a linear or nonlinear relationship model between data and calculates the parameters of the model; S8, Model verification and optimization: Based on the analysis results, the model is optimized and verified using an iterative optimization function; S9. Prediction analysis: perform Bayesian reliability calculations, predict status, and analyze qualitative and quantitative results; save the analysis results output to a file or a graphical user interface GUI.

[0021] Sensor data specifically includes: A. Implement data acquisition technology for accelerometers, gyroscopes, magnetometers, and temperature sensors. In the system, the accelerometer is used to measure the acceleration of the carrier, the gyroscope is used to measure the rotational angular velocity of the carrier in space and then integrate to obtain the spatial angle, the magnetometer data can be used to correct the heading angle, and the temperature data is used to provide temperature compensation; B. Add signal demodulation and filtering processing technology to reduce the problem of large noise in the output data of the missile sensor and improve the accuracy of attitude solution; C. The solution includes the solution analysis of the sensor output, the initial calibration of the attitude, the correction of the magnetometer, the calculation of the quaternion or Euler angle, the attitude solution, etc.; D. The storage of sensor data includes the original value of the accelerometer, gyroscope, magnetometer, temperature sensor, the calibrated value, and the attitude angle and quaternion calculated by the attitude solution. The amount of data is large, and the data source disappears every time the system is powered off. As for the sensor calibration parameters, including multiple error data of accelerometers and gyroscopes, and multiple error data of magnetometers, these parameters have the characteristics of small space, no change within a certain period of time, and only one-time storage; E. Design of communication technology between the system and the calibration host computer. Send the sensor to the calibration host computer for analysis through the communication protocol.

[0022] The steps for calculating the acceleration probability density are: Step 1: Perform statistics such as mean, variance, skewness, etc. to calculate whether all data conform to the common distribution; Step 2: Calculate the probability distribution of acceleration. We find that the probability of acceleration conforms to the gamma distribution, and then calculate α and β; Step 3: Calculate the vibration damage model based on the fault tree to obtain the minimum damage time prediction; Step 4: Perform Bayesian reliability calculation to obtain the current damage prediction and improve the evaluation accuracy.

[0023] The Bayesian reliability-based analysis method: Step 1: Randomly select m samples from the sample population, monitor the performance degradation of these m samples and record the performance degradation data of each sample at each moment. Step 2: Determine the functional distribution of the performance degradation data in the sample, and obtain the mean and sample standard deviation of each parameter in the model. The estimated distribution of this parameter can be used as the prior distribution of the model parameter. Step 3: Using the Bayesian method, combined with the performance degradation data on site, the distribution of the parameter is given as the posterior distribution function. Step 4: According to the failure threshold specified by the product, combined with the reliability analysis model, the reliability of the performance degradation at each moment is obtained to update the on-site information. Step 5: Select a suitable life distribution, and use the regression analysis method to obtain the estimated value of the unknown parameters in the product life distribution, so as to obtain the parameters of the reliability model.

[0024] System Function Requirements A. Main program control cycle: no more than 20ms; data acquisition frequency: 10ms; B. Status information should include: temperature and humidity data, vibration data, acceleration data and other status information, and the system can realize self-checking; C. Power supply requirements: battery standby ≥ 15 days; D. Dimension requirements: The structural dimensions shall not exceed 450mm×350mm×200mm under the premise of meeting the requirements of vibration resistance and heat dissipation; E. Data is stored locally and can be stored for more than 15 days; F. Operating temperature range: -40~85℃.

[0025] (3) Key technical indicators A. Technical indicators of nine-axis attitude sensor: Gyroscope range: greater than 200° / sec; gyroscope accuracy: less than 0.1° / sec; accelerometer range: more than ±10g; Accelerometer accuracy: less than 0.lg; reserved interface: USB or RS485;; operating temperature range: -40~85℃; Protection level: no less than IP53.

[0026] B. Temperature and humidity sensor technical indicators: Measurable temperature range: -40~80℃; temperature measurement accuracy: no more than 3.0%; relative humidity range: 0~100%; humidity measurement accuracy: no more than 3.0%.

[0027] C. Technical indicators of sound and light alarm: Decibel: not less than 100dB (adjustable); Working mode: support signal trigger alarm; Protection level: IP53 or above; Alarm tone: adjustable; Working temperature range: -40~80℃.

[0028] First, the main control chip and each sensor chip need to be initialized, including basic settings such as system clock, general IO, and configuration of functional peripherals such as timer and USART. The latter mainly completes the setting of parameters such as working mode, range, and output data frequency of the accelerometer, gyroscope, and magnetoresistive sensor. After completing this part of the initialization work, the MCU and the sensor can work normally, and then the data interaction is realized through the I2C bus. In order to improve the accuracy of the system measurement value, different calibration methods need to be used to calibrate and compensate the output value of each sensor to obtain the corresponding calibration coefficient. Then write the calibration coefficient into the MCU's Flash for storage.

[0029] Each time the system starts, it will determine whether the sensor needs to be recalibrated. If it is determined that calibration is not necessary, the calibration parameters stored in the last calibration can be directly read from the Flash. Then the system time is initialized, and the calibrated angular velocity, acceleration and magnetic field strength data are fused using the extended Kalman filter algorithm to obtain the optimal estimate of the attitude angle. Finally, it is determined whether the system running time has reached the set host computer update time. If the update time is reached, the attitude angle data and the output value of the sensor are sent to the host computer software through the serial port for display, and the state of the LED indicator light is changed to indicate that a complete data transmission is successful; otherwise, the sensor measurement value is continued to be read for attitude solution until the update time is exceeded and the corresponding data is transmitted to the host computer software. After the normal display of the host computer software is realized, the system time is initialized continuously in a cycle, and the sensor data is acquired, corrected, and fused, and the attitude and sensor data are sent to the host computer, so as to realize the real-time solution and display of the carrier attitude, and alarm according to the designed threshold. The lower computer mainly performs tasks that the upper computer cannot complete, cooperates with the upper computer to complete complex model establishment and simulation, realizes dynamic fault number model and Bayesian reliability classification demonstration, and provides computational simulation for vibration fatigue damage of ammunition. A. Implement data acquisition technology for accelerometers, gyroscopes, magnetometers, and temperature sensors. In the system, the accelerometer is used to measure the acceleration of the carrier, the gyroscope is used to measure the rotational angular velocity of the carrier in space and then integrate to obtain the spatial angle, the magnetometer data can be used to correct the heading angle, and the temperature data is used to provide temperature compensation; B. Implement signal demodulation and filtering technology. Because the sensor output data has large noise, it affects the accuracy of attitude solution. Therefore, for this system, the core processor needs to have faster computing power so that the attitude solution data can be real-time and effective, and the design must be reasonable and reliable; C. Implement complex, real-time, and accurate solution technology design. The solution includes solution analysis of sensor output, initial calibration of attitude, correction of magnetometer, calculation of quaternion or Euler angle, attitude solution, etc.; D. Implement storage technology design for sensor data and calibration parameters. The storage of sensor data includes the original values ​​of accelerometers, gyroscopes, magnetometers, temperature sensors, calibrated values, attitude angles, quaternions, etc., which are large in size and disappear every time the system is powered off. As for sensor calibration parameters, including multiple error data of accelerometers and gyroscopes, and multiple error data of magnetometers, these parameters have the characteristics of small space, no change within a certain period of time, and only single storage. E. Design of communication technology between the system and the calibration host computer. The sensor is sent to the calibration host computer for analysis through the communication protocol.

[0030] Selection of main components Nine-axis sensor - Selected brand: Domestic Wuhan Yuansheng Technology YIS106; YIS106 is a high-precision attitude sensor that can measure the three-dimensional attitude angle, acceleration, angular velocity and magnetic field strength information of the carrier. YIS106 is a highly reliable three-axis MEMS gyroscope, three-axis MEMS accelerometer and three-axis magnetic sensor. Through the embedded YFusion® high-performance attitude fusion algorithm, high-precision sensor error compensation algorithm and strict factory test calibration, it can achieve 0.1° roll and pitch angle measurement accuracy and 0.5° non-reference heading angle and 1° magnetic reference heading angle measurement accuracy. YIS106 adopts an ultra-small size design of 9.3×9.5×3mm, supports SMT patches, supports UART and I2C interface communications, and synchronous signal input, which is convenient for integrated applications.

[0031] Temperature and humidity sensor - selected brand: domestic AHT20 temperature and humidity sensor AHT20 is a new generation of temperature and humidity sensor in China, setting new standards in size and intelligence. It is embedded with a dual-row flat leadless SMD package suitable for reflow soldering, and the sensor outputs a calibrated digital signal in standard I2C format. MEMS temperature and humidity sensor, ADC bit number is 20Bit, with the advantages of small size, high precision and low cost. The most notable feature is the volume of 3*3*1.0mm. Relative humidity accuracy RH=±2%, temperature accuracy T=±0.3°C. Relative humidity measurement range RH=0~100%, temperature measurement range T=-40~85°C. AHT20 has a polytetrafluoroethylene waterproof and dustproof film, allowing the sensor to be used in harsh environmental conditions (such as spray water and high contact dust). AHT20 has the characteristics of localization, small size, high precision and low cost, and its small size makes it easy to embed into products. AHT20 is equipped with a newly designed ASIC dedicated chip, an improved MEMS semiconductor capacitive humidity sensor element and a standard on-chip temperature sensor element. Its performance has been greatly improved and even exceeds the reliability level of the previous generation of sensors. The new generation of temperature and humidity sensors has been improved to make its performance more stable in harsh environments. Each sensor is calibrated and tested, and the product batch number is printed on the surface of the product. Due to the improvement and miniaturization of the sensor, it is more cost-effective, and ultimately all devices will benefit from cutting-edge energy-saving operation modes.

[0032] Sound and light alarm - Sound and light alarm: LTE-5051; Rated voltage: AC220 / DC24 / DC12V; Decibel: not less than 100dB (adjustable); Working mode: flashing warning, high-brightness LED always on / strobe, support signal trigger alarm; Protection level: waterproof and dustproof design, IP66 and above; Alarm tone: adjustable; Working temperature range: -40~80℃.

[0033] Microprocessor selection - Selection: Espressif ESP32-S3-MINI-1U. ESP32-S3 is an MCU chip with integrated 2.4GHz Wi-Fi and Bluetooth 5 (LE), supporting long-range mode (LongRange). The ESP32-S3 microcontroller is based on Cadence's Xtensa Dual-Core 32-bit LX7 architecture, has 45 GPIO interfaces, a main frequency of up to 240MHz, 384KB of ROM, 512KB of SRAM and 16KB of RTC SRAM on board, and supports digital peripheral interfaces of 4×SPI, 3×UART, 2×I²C, and 2×I²S. The analog peripheral interface uses a 20-channel 2×12-bit SAR analog / digital converter. In this project, we choose the ESP32-S3-DevKitC-1U-N8R8 module, Flash 8MB QD, PSRAM 8MB OT, and SPI voltage 3.3v.

[0034] Antenna selection - The onboard PCB antenna (ESP32-S3-MINI-1) is connected to an external antenna (ESP32-S3-MINI-1U) through a connector.

[0035] After measurement, the power consumption of the whole system is 50mA~60mA. When the capacity of the disposable lithium battery is 18000mAH, the nine-axis attitude sensor and temperature and humidity sensor in the missile can work continuously for about 12.5~15 days. When the battery capacity is 21000mAH, the nine-axis attitude sensor and temperature and humidity sensor in the missile can work continuously for about 14.5~17.5 days. After actual testing, the 19000mAH lithium battery can meet the design requirements.

[0036] The upper computer is realized by additive manufacturing 3D printing technology. The lower computer material ADC12 waterproof grade IP67. Aluminum alloy die-casting, frosted surface; using silicone sealing strip; has good sealing performance, waterproof, dustproof, anti-corrosion, protection grade up to IP67; operating temperature: -40℃~120℃; anti-RF interference, good machining performance.

[0037] The system includes a host computer, a slave system module, an alarm module and other modules. The system is based on an embedded microprocessor, and the modules include a microcomputer system with multi-channel A / D acquisition, encoder signal input and various types of digital signal output. During operation, it is necessary to collect multiple sensor signals and automatically record and store the collected data information. According to the monitoring situation, the current acceleration and vibration status data in the projectile are released to the host computer through wireless communication, and when the acceleration reaches the threshold, the host computer alarm is triggered and the alarm unit is triggered. The supporting equipment mainly consists of a power supply and a charger. The system is a special equipment for recording the key vibration, acceleration, temperature and humidity of a typical product. It is a key technical parameter of the equipment in the process. It can be used for data management of training results and provide a data basis for technical status analysis and fault diagnosis.

[0038] The technical indicators of the quality monitoring system for portable ammunition for combat in high altitude and cold mountainous areas meet the following requirements: (1) CPU chip. It has the characteristics of high performance, low power consumption, and real-time performance. It can meet the communication interface requirements of the data acquisition module and storage module, has communication functions, and can achieve coordinated control of the overall circuit of the system's lower computer; (2) Vibration and acceleration sensors. They have acceleration A / D output, programmable acceleration range selection, and communication functions; (3) Temperature and humidity sensor. The digital temperature and humidity sensor integrates temperature sensor and humidity sensor, has high transmission reliability, meets the needs of equipment temperature and humidity monitoring, and has low measurement power consumption; (4) Memory. Storage capacity ≥ 64Gb, capable of communicating with the microprocessor through an interface.

[0039] The system uploads the collected vibration, acceleration data and temperature and humidity data to the microprocessor respectively. The microprocessor uploads the data to the host computer through the bus or USB. The host computer processes, displays and stores the data. The storage module is used to store the collected data locally. The monitoring data processing is divided into two processing methods: real-time and offline. Real-time processing is an alarm when the safety threshold is exceeded. When the portable ammunition reaches the ammunition safety failure warning threshold due to long-distance carrying, long-term unsealing and high-altitude cold environmental stress, the device will issue an alarm reminder at the first time to prepare quality warning for the carrying and shooting of ammunition, so that the management department (detachment) can respond in time. Offline processing is the whole process of monitoring data analysis. The whole process of environmental stress information collection is carried out for ammunition that lacks high-altitude cold carrying tests and new equipment such as newly equipped artillery and tanks that carry ammunition, and the Bayesian reliability evaluation method is used to perform full-process dynamic simulation and carrying performance analysis of the portable ammunition.

[0040] The prior knowledge of this project is: The formula for calculating the probability of the safety floor event of carrying ready-to-fire ammunition in the field is: P=P1. Principles for selecting bottom event values: P 1 The value is taken within the base value of the bottom event probability, depending on whether it has undergone vibration testing or simulation. If it has undergone vibration testing or simulation, the value is smaller, and if it has not undergone vibration testing or simulation, the value is larger.

[0041] The likelihood function for the two-parameter Weibull distribution is as follows: By taking the derivative of the above formula and setting it to zero, we can obtain the following conditions that satisfy the maximum likelihood estimation: After arranging and simplifying, we get: The function is defined according to the maximum likelihood estimation as follows: Solve the maximum likelihood system of equations: Let initial value , its parameter iteration relationship is as follows: Among them: According to the maximum likelihood estimation definition, we get Substituting into the above formula, we can get: The maximum likelihood estimate will Substitute into the above formula, and then substitute again to get the corresponding distribution parameters of the Weibull distribution. The maximum likelihood estimate of .

[0042] Basic definition of naive Bayes classifier The naive Bayes classifier uses the following formula to describe the classification process: Given a feature vector, the sample belongs to the category The posterior probability of .

[0043] Yes Category The prior probability of Is the given sample belongs to the category Case characteristics The conditional probability of occurrence.

[0044] The Naive Bayes algorithm usually consists of the following steps: 1) Calculate prior probability: Based on the training data set, calculate the prior probability P(Ck) of each category Ck.

[0045] 2) Calculate the conditional probability: For each feature xi and each category Ck, calculate P(xi | Ck).

[0046] 3) Apply the Bayesian formula: For a new sample, apply the Bayesian formula to calculate the posterior probabilities of all possible categories.

[0047] 4) Classification decision: Select the category with the highest posterior probability as the predicted category of the sample.

[0048] In this project, we obtain a priori knowledge probability based on the description of the probability base value of the vibration bottom event; then, based on the data obtained from the hardware of the quality monitoring system, we obtain a posterior knowledge of the acceleration probability distribution, and establish a Bayesian classification condition prediction of the transport vibration damage threshold based on the Bayesian framework. First, the acceleration threshold is set as the failure condition, and the percentage of the threshold is used as the limiting condition. We use acceleration as the key indicator, with a range of ±16g, which is the maximum range of 32g. The set value here is 24g, then 12g (50%), 16.8g (70%), and 24g (100%) are used as immediate threshold alarm indicators, and then the damage cycle Bayesian prediction is used as the dynamic threshold indicator. Appropriate sampling data is selected as priori knowledge, and Bayesian probability calculation is performed to obtain a posterior knowledge, and the Bayesian and reliability prediction evaluation of the number of cyclic damage of the ammunition is performed on the lower computer, and then the actual impact of vibration acceleration on the ammunition is determined and warned.

[0049] Table 4-1 Threshold indicator data table To construct the Bayesian index system without loss of generality, we assume The sequence of the collected data sets is: Here j≤m, m is the multi-source data sequence collected by the sensor, xi is the i-th observation array with t as the sampling time, we take 1~k (k≤n, n is the total sampling array) data groups as prior data for statistical analysis, that is, Xkn={Xkn}, Table 4-2 Sensor data table In this project, we set the sampling frequency to t f=0.02ms, about 50 groups of data are collected per minute. When the equipment starts, because GPS / Beidou needs to find stars and locate, the temperature and humidity sensors need to convert the first 100 groups of temperature and humidity data. The data is not accurate enough. The time for the tank to circle the field is about 15 minutes, so the sampled data is about 45,000. We sample 180,000 data at a time, so we take 15 minutes of data, that is, 45,000 data, as our prior knowledge data set, and perform statistical analysis on these 45,000 data. We select the first 45,000 data in the sample sensor_data_20230721_030215 collected by the sensor as the prior sample for analysis.

[0050] The total probability formula is, if and constitutes a partition of the sample space, then the probability of the event is equal to and The probability of each event is multiplied by the sum of the conditional probabilities of the two events.

[0051] Conditional Probability is the probability of another event A occurring given that another event B occurs. Mathematically, conditional probability is represented by P(A|B), and the calculation formula is: here yes A and B The probability of simultaneous occurrence, It is an event B Probability of occurrence.

[0052] Bayesian inference transforms the conditional probability formula: Bayes' Formula is a formula for updating the probability estimate of random events. The basic form is: in, and yes A and B The marginal probability of and is the conditional probability.

[0053] In this project, we obtained a priori knowledge probability based on the description of the base value of the vibration bottom event probability, and then performed Bayesian probability calculation and fitted the reliability degradation curve to obtain a posterior knowledge, that is, to establish a Bayesian performance degradation dynamic prediction model.

[0054] The following factors should be considered when testing and verifying the ammunition transport quality monitoring and evaluation early warning methods: selection of typical ammunition, test location selection, test system, test road surface, vehicle, vehicle speed and loading conditions.

[0055] Ammunition generally consists of five parts: fuze, projectile, cartridge, propellant charge, and primer. When analyzing the safety of ammunition, the ammunition is divided into four major parts: fuze, projectile charge, propellant charge, and primer. Among them, the fuze is the most dangerous part of the five components, and is the focus of analysis when conducting safety analysis of ammunition. The analysis of the projectile is mainly to judge the safety performance of the projectile charge. Since the safety performance of the propellant charge and primer of each type of ammunition is similar, and there are rarely safety accidents such as explosions of ammunition due to the safety performance of the propellant charge and primer in daily service management, this project did not analyze the impact of these two factors too much.

[0056] The fuze of ammunition is a device that can sense the target or other predetermined information, such as time, air pressure, instructions, etc., and detonate the projectile or warhead in a timely manner. The fuze has four major functions: safety, release, target sensing, and detonation. The safety function ensures the safety of ammunition during storage, transportation, handling and launching. Fuzes are generally composed of a firing control system, a detonation sequence, a safety system and an energy device. The safety system includes a safety mechanism and an explosion-proof mechanism. Fuze safety is the function of the fuze to release the safety and explode under specified conditions in various environments such as production, service processing, loading, launching, and delayed release of the safety. As the name suggests, fuze safety is a guarantee of its own and even the safety performance of ammunition, and it is also a prerequisite for its normal combat effectiveness. In GJB373A-97 "Fuze Safety Design Guidelines", there are such requirements for the redundant insurance of the fuze safety system: "The fuze safety system should include at least two independent safety components, each of which should be able to prevent the fuze from accidentally disarming, and the stimulus for activating these at least two safety components should be obtained from different environments." From this perspective, the safety level of the fuze is an important indicator for ensuring the safety of ammunition.

[0057] The test system is completed by the quality monitoring system, which consists of the lower-level equipment at the driving level (GPS / Beidou antenna, LORA antenna), missile-borne test equipment and sound and light alarm equipment.

[0058] Test hardware components The hardware of the test system consists of missile-borne test equipment (nine-axis attitude sensor, temperature and humidity sensor), cockpit lower computer equipment, GPS / Beidou antenna, LORA antenna, sound and light alarm and power supply. The equipment monitors in real time and takes into account some simple analysis and processing functions, providing threshold evaluation and alarm functions. The lower computer mainly uses fault tree software and Bayesian reliability software developed based on jamovi to conduct secondary evaluation of test data, and conduct comprehensive analysis and processing.

[0059] Preprocess the data first.

[0060] (1) Since the selected accelerometer has a range of ±16g, the acceleration range is from -16g to +16g, and the amplitude can reach 32g. In this project, we examine the impact of the peak value of acceleration on damage, but not the direction of acceleration, so we take the absolute value of the acceleration data for analysis. In order to reduce the amount of calculation, we take the absolute value of the data and round it off without affecting the statistical law.

[0061] (2) Due to the preprocessing of the angular velocity data of the missile-borne test equipment, the angular velocity sensor has a roll angle and heading angle range of ±180° and a pitch angle range of ±90°. In order to reduce the amount of calculation, we take the absolute value of the data and round it off without affecting the statistical law.

[0062] The coordinate system of the nine-axis sensor is defined as shown in the figure below. Rotation around the X-axis corresponds to the roll angle, rotation around the Y-axis corresponds to the pitch angle, and rotation around the Z-axis corresponds to the heading angle. In accordance with the right-hand rule, the thumb points in the direction of the rotation axis, and the angle increases when rotating around the positive direction of rotation.

[0063] The projectile is a cylinder and cannot be placed strictly according to the coordinate system position in the vehicle. It can only be placed according to the regulations. The problem brought about by this in the research and development is that the relative directions of the X-axis, Y-axis and Z-axis cannot be determined. According to the installation position of the sensor, only the Z-axis can be determined. The upward direction along the projectile's single head is the positive direction of the Z-axis, which is also in line with the law when the projectile leaves the barrel. Then, according to the actual situation, the sensor will decompose and calculate the corresponding direction of the X and Y axes. Therefore, during the transportation process, we can determine the data situation based on the actual placement of the projectile and the data, and specifically determine whether the data corresponding to the sensor XYZ axis represents the vertical, horizontal and longitudinal data.

[0064] The installation coordinate system of the nine-axis sensor is shown in the figure. According to the installation position, the red direction is determined as the sensor Z-axis direction test for longitudinal data, the blue direction is determined as the sensor X-axis direction test for transverse data, and the green direction is determined as the sensor Y-axis direction test for vertical data.

[0065] Calculated according to the ammunition vibration damage model: Table 5-1 Ammunition vibration damage model G=[0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 22 27] Here, frequency of occurrence % = number of occurrences / total number of data; The ratio of the number of cycles of vibration damage to the ammunition that has occurred % = the number of occurrences ni / the number of cycles of vibration damage to the acceleration ammunition N. The following conclusions can be drawn from the above table: (1) Here our sampling time is tf=0.02s. The total collected data in this test is: 739200. The total test time is 739200×0.02÷60, which is 246.4 minutes. The acceleration is 9g, the total vibration time is 17.769 minutes, and the time percentage is 7.211444805. (2) In this 9g acceleration test, the number of vibration cycles is 53307, accounting for 0.001267803227083 of the total number of cycles of acceleration damage, which is 42,046,745.7892998000; (3) Based on the above proportional relationship (1) (2), we predict that when the acceleration is 9g, and the car continues to run in the current test field, it will take: 42,046,745.7892998000÷7.211444805×100*0.02÷60=3177039.25666616 minutes, which is calculated to be 134.9666 days. At this time, the ammunition reaches the maximum value of vibration fatigue damage and is completely damaged. In this test field, we use the data in the test field as Bayesian prior knowledge and perform Bayesian reliability prediction calculations. The expression is as follows: Sample X=[ G=[0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 2227], P (n | 0.0761999458874460.026149891774892 0.0094859307359310.072114448051948 0.0017234848484850.000374729437229 0.000150162337662 0.000063582251082 0.0000216450216450.000028409090909 0.000001352813853 0.000009469696970 0.000002705627706 0.000001352813853 0.000001352813853] P(xi)=1 / 28; ① In this project, the bottom event probability calculation formula for the vibration bottom event probability base value in the system resonance failure described in the "GJB / Z 179-2015 Fuze Fault Tree Bottom Event Data Manual" is as follows: P=P 1. Principles for selecting bottom event values: P 1 The value is taken within the base value of the bottom event probability, depending on whether it has undergone vibration testing or simulation. If it has undergone vibration testing or simulation, the value is smaller, and if it has not undergone vibration testing or simulation, the value is larger.

[0066] According to the national military standard, the failure probability is as follows: P (n) = 1 × 10 -4 As the transport vibration accumulates, according to Bayesian theory, we calculate as follows: During installation, the host computer rotates the missile body to seal the sensor inside the missile body, measures the temperature and humidity inside the missile body, acceleration, angular velocity, magnetic environment, etc., evaluates the status of the ammunition fuze in real time, and issues an evaluation alarm. According to the measured data analysis, the temperature environment stress shock inside the missile body is within the specified range from 22℃ to 50℃ for 2 hours. The humidity is maintained at the state when the missile body is installed because a closed space is formed after the missile body is rotated during installation. The magnetic field state forms an approximately shielded space inside the missile body, so the measured value is also within the range specified by the national military standard. The air pressure and other states are also within the specified range. Therefore, this project focuses on the safety and reliability factors that may be caused by the fatal defect state of the fuze caused by mechanical vibration during transportation.

[0067] (1) Since the selected accelerometer has a range of ±16g, the acceleration range is from -16g to +16g, and the amplitude can reach 32g. In this project, we examine the impact of the peak value of acceleration on damage, but not the direction of acceleration, so we take the absolute value of the acceleration data for analysis. In order to reduce the amount of calculation, we take the absolute value of the data and round it off without affecting the statistical law; (2) Due to the pre-processing of the angular velocity data of the missile-borne test equipment, the angular velocity sensor has a roll angle and heading angle range of ±180° and a pitch angle range of ±90°. In order to reduce the amount of calculation, we take the absolute value of the data and round it off without affecting the statistical law; (3) The vertical vibration value is the largest, followed by the lateral vibration and the smallest in the longitudinal vibration. The average value of the vibration acceleration in the three directions increases with the increase of vehicle speed. The magnitude of the vibration is most closely related to vehicle speed and road conditions. The higher the vehicle speed and the worse the road conditions, the greater the vibration. This project focuses on the characteristics of carrying ready-to-fire ammunition and combines the environmental conditions that may be encountered during the field transportation of military supplies specified in the national military standard GJB3493-1998 "Environmental Conditions for Military Supplies Transportation", including mechanical, climatic, chemically active substances, mechanically active substances, biological and electromagnetic, etc. The project analyzes the factors affecting the safety and reliability of field transportation, such as the environmental factors of carrying ready-to-fire ammunition, the transportation traffic environmental factors and the vehicle speed. The research shows that: (1) In most cases, the "Environmental Conditions for Military Material Transportation" actually stipulates a variety of influencing factors, but only some of them have an impact on the safety and reliability of field transportation. Many factors do not exist or are not too harsh in certain environments, or the conditions are moderate or the probability of occurrence is small, and will not have a significant impact; (2) During the storage and transportation of ammunition, the vibration of ammunition is closely related to the environmental conditions in which it is located. The ammunition exhibits dynamic vibration and impact characteristics due to the mutual influence and interference of various factors within the system; (3) During the field transportation process, field ammunition has different tolerances to different environmental factors. From practical experience, we know that vibration and shock have the greatest impact on the safety of field ammunition transportation, followed by high and low temperatures and high humidity. Chemically active substances, biological environmental conditions, solar radiation, etc. also have an impact on the ammunition packaging system during long-term transportation.

[0068] (4) As a transportation activity, ammunition field transportation is inevitably affected by transportation conditions. As an objective social activity, it is inevitably affected by objective factors such as nature, human social history and social transportation needs. In the transportation system, the driver is the core factor. Under the joint influence of vehicles, roads, environment and management factors, unsafe behavior of people and unsafe state of things are the direct causes of accidents, and management errors are the essential causes of accidents.

[0069] (5) Vehicle speed restricts the efficiency of transportation support, and at the same time profoundly affects the safety of the transportation environment and transportation traffic. Field transportation of ammunition must not only prevent the impact and vibration environment from affecting the quality and safety and reliability of ammunition, but also ensure that no traffic accidents occur under normal transportation conditions, otherwise major hazards or traffic accidents may occur. Vibration and shock have the greatest impact on the safety of ammunition transportation, and vehicle speed is closely related to shock and vibration. According to practical experience, when the transportation speed is high, the impact of sudden braking in an emergency will be stronger.

[0070] (6) Factors affecting the safety of ammunition transportation can be divided into two categories: transportation environment and transportation traffic. The main environmental conditions that affect the safety of the transportation environment are the shock and vibration in the mechanical environment, and the intensity of shock and vibration is closely related to the speed of the transportation vehicle.

[0071] Combined with the structural characteristics of ammunition, the damage analysis of ammunition caused by factors such as transport vibration and impact during field transportation was carried out. The research shows that: (1) Under the premise of ensuring the safety of the ammunition, the impact and vibration in the transportation environment are the main factors affecting the safety of ammunition transportation. (2) When the ammunition is impacted during transportation, although its speed only changes to a limited extent, it will generate a great acceleration due to the extremely short duration of the impact. The characteristics of the impact are: the impact force is short in duration and extremely large in magnitude. The duration of the impact and the amplitude of the acceleration to which the ammunition is subjected are the two most important parameters in mechanical shock. (3) During field transportation, vibration is also an inevitable phenomenon. When the vibration excitation on the ammunition exceeds the fatigue limit that the ammunition can withstand, the ammunition will be damaged or destroyed. (4) The consequences of the impact stress on the internal mechanisms and components of the ammunition may manifest as: mechanism damage or premature action, loosening of components after damage, loosening or breaking of the charge, damage to the electronic circuit, etc., which seriously affect the combat effectiveness of the ammunition.

[0072] Ammunition reliability evaluation method based on Bayesian The reliability assessment method based on Bayesian network is an effective way to solve the uncertainty problem of complex systems. The application of Bayesian network (BN) for fault diagnosis has many advantages as follows: (1) The Bayesian network method has a solid theoretical foundation. The Bayesian network is based on probabilistic reasoning, and the reasoning results are persuasive. Moreover, compared with the Bayesian method, the Bayesian network has greatly reduced the requirements for prior probability. At the same time, the conditional independence expressed by the Bayesian network is conducive to expressing the correlation between reliability; (2) The probabilistic reasoning algorithm of the Bayesian network has low requirements for existing information and can be used for reasoning under incomplete and uncertain information conditions; (3) Bayesian networks are more suitable for expressing equipment fault diagnosis problems. Using fault tree models to express fault diagnosis problems is a common method, and it can be used to study the optimization of diagnostic strategies. However, once a fault tree model is built, it is not easy to change, accept and process new information, conduct human-computer information interaction, and include relevant information that is not related to equipment failure but can be used to diagnose the failure into the fault diagnosis process. Bayesian networks can learn at any time through practical accumulation to improve network structure and parameters and enhance reliability assessment capabilities; (4) Bayesian networks have strong learning capabilities. Learning capabilities are an important feature of artificial intelligence. Bayesian network learning is the process of finding a network (DAG) that best matches a given training sample set, including learning the network structure G and learning the conditional probability table P. BN is very suitable for expressing evaluation decision problems, using a model that combines graphics and statistics to express the components and elements of the equipment and the test segments (including observations) that can be used to evaluate them and their relationships.

[0073] Verification of ammunition quality monitoring and assessment early warning method Combined with the causes of ammunition accidents, a typical ammunition fuze was selected and established. A certain type of 152mm bullet was used as the test object to establish an ammunition transport vibration failure model, determine the ammunition transport failure threshold, and develop a quality monitoring system to establish a model; according to Bayesian and reliability theories, the failure number model was used as Bayesian prior knowledge, and combined with the state information of the developed quality monitoring and early warning system to form prior knowledge for analysis, and the safety and reliability of the ammunition after transportation were predicted. The experiment showed that the theory was correct, the test method was feasible, and the prediction results were in line with the law of the ammunition failure model. This project focuses on the dynamic quality monitoring of portable ammunition in the field transportation environment of high-altitude and cold areas. It has realized the installation of the portable ammunition environmental information monitoring equipment in the portable ammunition fuze. The sensor senses and obtains the first-hand vibration impact angular velocity, acceleration, temperature and humidity status data of the fuze, and uses the cumulative damage theory and dynamic Bayesian reliability evaluation method to conduct safety warnings for portable ammunition in high-altitude and cold mountain operations. The innovation points of the project are mainly from two aspects: theoretical method innovation and engineering development.

[0074] This project applied the cumulative damage theory for the first time to the vibration fatigue test of portable ammunition fuzes, and obtained a life damage prediction method for vibration impact of portable ammunition fuzes during portable transportation; acceleration, temperature, humidity, and ammunition posture information were used as real-time thresholds for online evaluation and early warning, and fault tree, Bayesian, and reliability algorithm fusion were used offline to predict the safety and reliability of portable ammunition field transportation, filling the theoretical and engineering gaps in real-time safety risk early warning of portable ammunition field transportation; in the verification test, we found that the original data did not conform to the conventional normal distribution. After analyzing and processing the data, we found that the data after rounding the absolute value of the three-axis acceleration data conformed to the Gamma distribution.

Claims

1. A portable ammunition detection device, characterized in that The invention comprises an upper computer (1) and a lower computer (2); the upper computer (1) is connected to the body of the ready-to-fire ammunition, the lower computer (2) is connected to the cockpit, and the upper computer (1) and the lower computer (2) are connected via antenna signals; the upper computer (1) is divided into an upper fixed frame assembly (101) and a lower fixed frame assembly (102), and the upper fixed frame assembly (101) and the lower fixed frame assembly (102) are hollow cylindrical; the upper fixed frame assembly (101) is provided with a body threaded section (108), and the upper fixed frame assembly (101) and the lower fixed frame assembly (102) are threadedly connected; the upper computer (1) is connected to the body of the ready-to-fire ammunition via the body threaded section (108).

2. The portable ready-to-fire ammunition detection device according to claim 1, characterized in that The upper fixed frame assembly (101) is provided with a clamping block (103), the ready-to-fire ammunition body is provided with a clamping groove (104), and the clamping block (103) is connected to the clamping groove (104) by buckling.

3. The portable ready-to-fire ammunition detection device according to claim 1, characterized in that The upper fixed frame assembly (101) is threadedly connected to a support sleeve (105), and the support sleeve (105) is threadedly adjusted to a telescopic position with the upper fixed frame assembly (101); the other end of the support sleeve (105) is in contact with the inner wall of the ready-to-fire ammunition body.

4. The portable ready-to-fire ammunition detection device according to claim 3 is characterized in that The upper fixed frame assembly (101) is provided with a plurality of support sleeves (105), and the plurality of support sleeves (105) are evenly distributed along the circumferential direction of the outer edge of the upper fixed frame assembly (101).

5. The portable ready-to-fire ammunition detection device according to claim 3 or 4, characterized in that The support sleeve (105) is provided with a threading hole (106) and a cable bayonet (107); the threading hole (106) passes through the support sleeve (105), and the cable bayonet (107) is communicated with the threading hole (106).

6. The portable ammunition detection system is characterized by The invention comprises the portable ammunition detection device as claimed in claim 1, a GPS Beidou module, a wireless signal module, an audible and optic alarm module, a Tyep-c data transmission module and a sensor module; the lower computer (2) of the portable ammunition detection device is provided with a GPS Beidou module, a wireless signal module, an audible and optic alarm module and a Tyep-c data transmission module; the GPS Beidou module is used for vehicle positioning, the audible and optic alarm module is used for data abnormality alarm, and the Tyep-c data transmission module is used for data transmission and downloading of the lower computer (2); the upper computer (1) of the portable ammunition detection device is provided with a sensor module and a Tyep-c data transmission module, and the sensor module is divided into an MCU processor, a nine-axis sensor, and a temperature and humidity sensor; the MCU processor is connected with the nine-axis sensor and the temperature and humidity sensor for projectile data collection; the Tyep-c data transmission module is used for data transmission and downloading of the upper computer (1); the wireless signal module is used for wireless data connection between the upper computer (1) and the lower computer (2), and transmits the sensor module data of the projectile upper computer (1) to the lower computer (2).

7. The portable ready-to-fire ammunition detection system according to claim 6, characterized in that The upper computer sensor module completes the measurement of attitude angle, acceleration, temperature and humidity, and transmits the data to the lower computer through the LORA Internet of Things antenna, and leaves a type-c interface for direct reading of the data collected by the upper and lower computers; the current missile attitude angle, acceleration, temperature and humidity data are monitored in real time through wireless communication. When the above real-time monitoring data reaches the threshold, the sound and light alarm module is triggered to give an alarm prompt.

8. The quality big data analysis method of the portable ready-to-fire ammunition detection system according to claim 6 or 7 is characterized in that The steps include: S1. Data preparation: All data files need to be read from the host computer (1) and the slave computer (2); S2. Data cleaning: Clean abnormal data, re-screen, sort, and remove duplicates, and delete abnormal or missing values; S3. Descriptive statistical analysis: Use big data models for data analysis; S4. Data visualization: Use the drawing functions of the big data model to draw histograms, scatter plots, and pie charts of the data; S5, Probability distribution: Calculate the probability density function PDF and cumulative distribution function CDF of the data; S6. Hypothesis test: Use the KW hypothesis test function to perform hypothesis test on the data to test whether there is a significant difference between the data; S7, regression analysis: The regression analysis function used establishes a linear or nonlinear relationship model between data and calculates the parameters of the model; S8, Model verification and optimization: Based on the analysis results, the model is optimized and verified using an iterative optimization function; S9. Prediction analysis: perform Bayesian reliability calculations, predict status, and analyze qualitative and quantitative results; save the analysis results output to a file or a graphical user interface GUI.

Images