Product reliability improving method and system based on equipment failure data analysis

Abstract

The invention discloses a product reliability improvement method and system based on equipment failure data analysis, and particularly relates to the field of failure data, and the method comprises the steps: enabling an equipment parameter collection module to be embedded in a collaborative operation environment of equipment networking, and collecting the real-time data of each single equipment in the equipment networking; defining failure mode characteristics of the equipment, and establishing an equipment failure state recognition model; calculating the decision weight of each device in the device network in the decision model based on the core decision model; when it is identified that the state of a single device in the device network is invalid, a data transmission channel between the invalid device and the central control device is disconnected, and the data processing center sends analog data to the central control device; performing decision weight redistribution on normal equipment in the equipment network; and when the state of the failed equipment is recognized to be normal, resetting a decision-making weight in the core decision-making model of the central control equipment, and recovering a data transmission channel of the failed equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of failure data analysis, and more specifically, to a method and system for improving product reliability based on equipment failure data analysis. Background Art

[0002] With the rapid development of the Internet of Things (IoT) and smart devices, such as smart homes, autonomous driving, industrial automation, and smart manufacturing, more and more products rely on multiple devices working together to achieve efficient automated control and data processing.

[0003] However, during long-term operation, equipment often faces problems such as failure, performance degradation or failure. Especially in a complex equipment networking environment, the failure of a single device may affect the decision-making accuracy of the entire system, and even induce a chain of equipment failures, which in turn leads to the unavailability of the entire system. In order to ensure the stability and reliability of the system, it is necessary to identify and handle the failure status of the equipment in a timely manner, and adjust the decision-making and control strategies of the system in a targeted manner to avoid the direct impact of the failure of a single device on product performance.

[0004] Therefore, how to propose a universal solution based on analyzing equipment failure data to improve the reliability of related products is an urgent problem that needs to be solved.

[0005] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a product reliability improvement method based on equipment failure data analysis to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions: S1: Analyze the device networking structure of the product, embed the device parameter collection module into the collaborative working environment of the device networking, and collect the real-time data of each single device in the device networking; S2: Select the historical period when the product's function deviation occurred, analyze the historical data of the operation of the single device in the device network during the historical period, define the failure mode characteristics of the device, and establish a device failure state identification model; S3: Obtain the core decision model of the central control device in the device network, and calculate the decision weight of each device in the device network in the decision model based on the hierarchical analysis method; S4: When it is identified that a single device in the device network is in a failed state, the data transmission channel between the failed device and the central control device is disconnected, and the data processing center sends analog data to the central control device; S5: Based on the decision weight ratio of the failed device in the core decision model of the central control device, the decision weight of the normal devices in the device network is redistributed; S6: When it is identified that the state of the failed device has turned to normal, the device data simulation is suspended, and the decision-making power in the core decision-making model of the central control device is reset, and the data transmission channel between the failed device and the central processing device is restored.

[0008] In a preferred embodiment, in S1, the device networking structure of the product is analyzed, and the device parameter acquisition module is embedded in the collaborative working environment of the device networking. The real-time data of each single device in the device networking is collected, specifically including: Analyze the equipment operation architecture of the product, identify and mark the functional modules of each single device and its functional role positioning in the entire equipment network; Extract the functional role positioning tags of devices in the device network, eliminate the edge devices in the device network, obtain the communication interfaces and interaction methods of all devices in the collaborative working environment of the devices that only retain the information input function and the central control device, and embed the global device parameter acquisition module to collect device data in real time and upload it to the data processing center.

[0009] In a preferred embodiment, in S2, selecting a historical period in which a product has functional deviation, analyzing historical data of operation of a single device in a device network during the historical period, defining the failure mode characteristics of the device, and establishing a device failure state identification model specifically include: Obtain the historical operation log of the product, retrieve the log records where functional abnormalities occur, and mark the start and end time span of consecutive abnormal log records as the functional offset period; According to the actual length of each functional offset period, an offset comparison section is defined at a fixed ratio before and after the endpoint of the functional offset period, and historical operation data of all single devices in the device network during the functional offset period and the offset comparison section are extracted; Calculate the root mean square span of the historical operation data of each single device in the functional deviation period and the deviation control period, screen the single devices whose root mean square spans exceed the preset gradient, and calibrate the data of the screened single devices in the current functional deviation period as failure data; The failure data calibrated for each single device in all functional deviation periods are integrated to form a complete time series failure data set. The time series failure data set of each single device is subjected to fast Fourier transform to extract the frequency domain features of the failure data set. At the same time, the time series dimension features of the failure data set are extracted based on the time series analysis model. The frequency domain features and the time series dimension features are integrated into the failure mode features of the single device. The failure status recognition model of each single device in the equipment network is trained based on a deep neural network, and the trained model is deployed to the data processing center to perform status recognition on the collected real-time equipment data.

[0010] In a preferred embodiment, in S3, obtaining the core decision model of the central control device in the device network, and calculating the decision weight of each device in the device network in the decision model based on the hierarchical analysis method specifically include: Check whether the core decision model of the central control device has preset the decision weight of each single device in the device network; If so, the decision weights of individual devices are synchronously calibrated according to the preset weights of the core decision model; If not, the communication signals between each single device and the central control device in the device network and the response signals of the central control device within the set time window are collected, and a hierarchical model is established to calculate the decision weight of the single device. The specific steps are as follows: Set the top level of the hierarchical model to the decision weight of a single device; The second layer of the hierarchical structure model is set as the weight judgment standard, which includes the communication frequency between the single device and the central control device, the control conversion rate after communication, and the average response speed; Set the bottom layer of the hierarchical model to be a single device entity of device networking; The importance scores of the individual judgment criteria in the weight judgment criteria are set to form a judgment matrix, and the ratio of all elements in the judgment matrix to the total of the column is calculated to obtain a normalized matrix; Calculate the average value of each row of the normalized matrix to obtain each weight judgment standard coefficient, bring the communication frequency between each single device and the central control device, the control conversion rate after communication, and the average response speed into the weight judgment standard coefficient, and calculate the weighted score of each single device; The decision weight of each device in the device network in the decision model is calibrated based on the calculation result of the weighted score.

[0011] In a preferred embodiment, in S4, when it is identified that a single device in the device network is in a failed state, disconnecting the data transmission channel between the failed device and the central control device, and sending the simulation data from the data processing center to the central control device specifically includes: When the failure state identification model of a single device in the data processing center determines that the current device is in a failure state, the bus isolation mechanism is activated to disconnect the real-time data transmission channel between the failed device and the central control device; The data processing center analyzes the historical data trends of failed equipment and generates simulated equipment data in real time based on the time series prediction algorithm and sends it to the central control device.

[0012] In a preferred embodiment, in S5, based on the decision weight proportion of the failed device in the core decision model of the central control device, redistributing the decision weights of the normal devices in the device network specifically includes: Obtain the decision weight calibrated in the decision model for each device in the device network, and store the calibrated decision weight record as a decision weight allocation snapshot in the decision weight snapshot table of the data processing center; All decision weights of failed devices are recovered, and the decision weights of failed devices are distributed to other normal devices in an equal proportion load balancing manner according to the ratio of the decision weight of each normal device in the device network to the cumulative value of the decision weights of all normal devices.

[0013] In a preferred embodiment, in S6, when it is identified that the state of the failed device turns to normal, the device data simulation is suspended, and the decision right in the core decision model of the central control device is reset, and the data transmission channel between the failed device and the central processing device is restored, which specifically includes: When the failure state identification model of a single device in the data processing center determines that the failed device has returned to a normal state, the data center suspends data simulation of the failed device; According to the device identification of the failed device, retrieve the weight distribution snapshot of the device from the decision weight snapshot table of the data processing center, and reset the decision weight in the device network based on the weight distribution snapshot; After the decision weight is reset, the snapshot record in the decision weight snapshot table is deleted, and the data transmission channel connection between the failed device and the central processing device is restored.

[0014] On the other hand, the present invention provides a product reliability improvement system based on equipment failure data analysis, including an equipment parameter acquisition module, a failure state identification module, a simulation data transmission module, and a decision weight allocation module: Equipment parameter acquisition module: selects corresponding data acquisition technology according to the communication interface and interaction mode of the equipment in the collaborative operation environment, collects equipment data in real time and uploads it to the data processing center; Failure status identification module: Based on the historical failure data characteristics of all single devices in the equipment network during the historical period when the product function shifted, a failure status identification model is established to identify the status of the collected real-time equipment data; Simulation data transmission module: when the device is judged to be in failure state, it generates simulation device data in real time to replace the failure data and sends it to the central control device; Decision weight allocation module: Calibrate the decision weight of a single device through decision model preset or hierarchical analysis. When a device becomes invalid, record a snapshot of the decision weight of the current device networking status, and recover all decision weights of the invalid device and distribute them to other normal devices in a balanced manner.

[0015] The technical effects and advantages of the product reliability improvement method and system based on equipment failure data analysis of the present invention are as follows: By deeply analyzing the operation history data of individual devices in the device network, a device failure mode and state recognition model is established. By real-time monitoring of device status and timely identification of failed devices, dynamic adjustment of device decision weights can be achieved to ensure that the central control device can effectively manage the operation of normal devices. This method avoids the direct impact of device failure on product performance by simulating the behavior of failed devices and adjusting the system accordingly, while improving the reliability of device networking. It is of great significance in improving product reliability, reducing downtime and optimizing maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a method for improving product reliability based on equipment failure data analysis according to the present invention; Figure 2 The present invention is a structural schematic diagram of a product reliability improvement system based on equipment failure data analysis. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Example 1 Figure 1 The present invention provides a method for improving product reliability based on equipment failure data analysis, which comprises the following steps: S1: Analyze the device networking structure of the product, embed the device parameter collection module into the collaborative working environment of the device networking, and collect the real-time data of each single device in the device networking; S2: Select the historical period when the product's function deviation occurred, analyze the historical data of the operation of the single device in the device network during the historical period, define the failure mode characteristics of the device, and establish a device failure state identification model; S3: Obtain the core decision model of the central control device in the device network, and calculate the decision weight of each device in the device network in the decision model based on the hierarchical analysis method; S4: When it is identified that a single device in the device network is in a failed state, the data transmission channel between the failed device and the central control device is disconnected, and the data processing center sends analog data to the central control device; S5: Based on the decision weight ratio of the failed device in the core decision model of the central control device, the decision weight of the normal devices in the device network is redistributed; S6: When it is identified that the state of the failed device has turned to normal, the device data simulation is suspended, and the decision-making power in the core decision-making model of the central control device is reset, and the data transmission channel between the failed device and the central processing device is restored.

[0019] In S1, the device architecture of the product is first analyzed to clarify the position and functional modules of each single device in the entire device network. The functional roles in the device network include core control units, data acquisition devices, execution devices, etc. The role of each device in the network is determined by the functional requirements of the devices and their dependencies.

[0020] Core devices are usually devices that undertake the main computing and decision-making tasks, while edge devices may be auxiliary devices for execution or sensing. In device networking, edge devices do not directly participate in the decision-making and control process, so they can be excluded in this step. The following is a typical scenario: In an autonomous driving system, the device network can include radars, cameras, GPS sensors, central control units, etc. Through device architecture analysis, the functional role of each device is determined, and edge devices that are not directly involved in decision-making (such as auxiliary sensors) are eliminated. Core devices such as radars and cameras will work with the central control unit through high-speed communication interfaces. The device parameter acquisition module transmits data to the processing center in real time to analyze the vehicle status and make real-time decisions.

[0021] The embedded global device parameter acquisition module collects device data in real time and uploads it to the data processing center. The collaborative working environment between the device and the central control device needs to clarify the key parameters of each device, such as the communication protocol, data transmission rate, and data format, to ensure seamless collaboration between devices. For example, devices may exchange data through communication protocols such as HTTP and CoAP. The communication interface between each device and the central control device may include wireless communication, LAN connection, or other protocols to ensure that information can be transmitted stably in real time.

[0022] In S2, a historical period in which the product has functional deviation is selected, the historical data of the operation of individual devices in the device network during the historical period is analyzed, the failure mode characteristics of the equipment are defined, and an equipment failure state identification model is established.

[0023] Obtain the historical operation log of the product, retrieve the log records where functional abnormalities occur, and mark the start and end time span of consecutive abnormal log records as the functional offset period; According to the actual length of each functional offset period, an offset comparison segment is defined before and after the endpoint of the functional offset period at a fixed ratio (50% by default), and the historical operation data of all single devices in the device network during the functional offset period and the offset comparison segment are extracted.

[0024] Calculate the RMS spans of the historical operating data of the time series of each single device in the functional offset period and the offset control period, screen the single devices whose RMS spans exceed the preset gradient (for example, the span amplitude exceeds 120%), and calibrate the data of the screened single devices in the current functional offset period as failure data.

[0025] The failure data of each single device calibrated in all functional deviation periods are integrated to form a complete time series failure data set. The time series failure data set of each single device is subjected to fast Fourier transform to extract the frequency domain characteristics of the failure data set. The time series data of the device in the failure period is set as , its fast Fourier transform expression is: In the formula, represents the component at the kth frequency in the frequency domain signal, is the total number of time points of the time series data within the failure period, represents the nth point of the time domain signal, j is the imaginary unit, k is the index frequency, and from the frequency domain signal The frequency peak, spectrum width and energy distribution are extracted as the frequency domain features of the failure data set.

[0026] Based on the autoregressive moving average model (ARMA), the trend and periodicity characteristics of the time series of the failure data set are extracted, and the frequency domain characteristics and time series dimension characteristics are integrated into the failure mode characteristics of the single device.

[0027] The failure status recognition model of each single device in the equipment network is trained based on a deep neural network, and the trained model is deployed to the data processing center to perform status recognition on the collected real-time equipment data.

[0028] In S3, the core decision model of the central control device in the device network is obtained, and the decision weight of each device in the device network in the decision model is calculated based on the hierarchical analysis method.

[0029] Check whether the core decision model of the central control device has preset the decision weight of each single device in the device network; If so, the decision weights of individual devices are synchronously calibrated according to the preset weights of the core decision model; If not, the communication signals between each single device and the central control device in the device network and the response signals of the central control device within the set time window are collected, and a hierarchical model is established to calculate the decision weight of the single device. The following is a specific calculation example: Set the top level of the hierarchical model to the decision weight of a single device; The second layer of the hierarchical model is set as the weight judgment standard, which includes the communication frequency (CF) between the single device and the central control device, the control conversion rate (CCR) after communication, and the average response speed (RS); Set the bottom layer of the hierarchical model to be a single device entity of the device network; The importance scores of the individual judgment criteria in the weight judgment criteria are set to form a judgment matrix. Assuming that the relationship between CF and CCR is 3, the relationship between CF and RS is 5, and the relationship between CCR and RS is 3, then the judgment matrix is : ; To eliminate the dimensionality difference in the matrix, we first calculate the sum of each column and normalize the matrix by dividing each element in the matrix by the sum of that column, specifically: Sum of CF columns: 1 + 1 / 3 + 1 / 5 ≈ 1.533, CCR column sum: 3 + 1 + 1 / 3 ≈ 4.333, RS column sum: 5 + 3 + 1 ≈ 9, Its normalized matrix : ; Calculate the average value of each row of the normalized matrix to obtain the standard coefficients for each weight judgment , , .

[0030] The communication frequency between each single device and the central control device, the control conversion rate after communication, and the average response speed are converted to the same scale and then brought into the weight judgment standard coefficient to calculate the weighted score of each single device. For example, after the monitored data is converted into scores, the frequency score, control conversion rate score, and response speed score of the device are device A (8, 7, 6), device B (6, 8, 7), and device C (7, 6, 8), respectively. The weighted scores of devices A, B, and C are: S(A)=(8*0.634)+(7*0.260)+(6*0.106)=5.072+1.820+0.636=7.528, S(B)=(6*0.634)+(8*0.260)+(7*0.106)=3.804+2.080+0.742=6.626, S(C)=(7*0.634)+(6*0.260)+(8*0.106)=4.438+1.560+0.848=6.846, Based on the weighted score of each device, the decision weight of the device in the decision model can be determined, which is 0.35 for device A, 0.32 for device B, and 0.33 for device C.

[0031] In S4, when it is identified that a single device in the device network is in a failed state, the data transmission channel between the failed device and the central control device is disconnected, and the data processing center sends the simulation data to the central control device.

[0032] When the failure status identification model of a single device in the data processing center determines that the current device is in a failure state, the bus isolation mechanism is activated to disconnect the real-time data transmission channel between the failed device and the central control device, ensuring that the central control device no longer receives data from the device to prevent the failed data from causing the entire system to be unavailable.

[0033] The data processing center analyzes the historical data trends of the failed equipment, uses time series prediction algorithms (such as ARIMA model, LSTM neural network, Prophet algorithm or simple historical average method, etc.) to predict the future data of the failed equipment, and generates a set of simulated data to fill the missing data of the failed equipment. This simulated data will be sent to the central control device to ensure that the system can continue to operate and avoid operating errors caused by missing data.

[0034] In S5, based on the decision weight ratio of the failed device in the core decision model of the central control device, the decision weight of the normal devices in the device network is redistributed to ensure that when a device fails, the decision focus of the system is transferred to other normal device data.

[0035] The decision weight calibrated in the decision model of each device in the device network is obtained, and the calibrated decision weight record is stored as a decision weight allocation snapshot in the decision weight snapshot table of the data processing center.

[0036] All decision weights of failed devices are recovered, and the decision weights of failed devices are distributed in equal proportion according to the decision weights of other normal devices. Specifically, the decision weights of failed devices are distributed according to the proportion of the decision weights of normal devices to the cumulative value of the decision weights of all normal devices, ensuring that the loads of all normal devices are reasonably balanced.

[0037] In S6, when it is identified that the state of the failed device has turned to normal, the device data simulation is suspended, and the decision-making power in the core decision model of the central control device is reset, and the data transmission channel between the failed device and the central processing device is restored.

[0038] When the failure state identification model of a single device in the data processing center determines that the failed device has returned to normal state, the data center suspends data simulation of the failed device to prevent the simulated data from affecting the central processing device after the device recovers its weight.

[0039] According to the device identification of the failed device, the weight distribution snapshot of the device is retrieved from the decision weight snapshot table of the data processing center. The decision weight in the device network is reset based on the weight distribution snapshot to ensure that the decision weight of the failed device in the device network is restored to the state before the failure.

[0040] After the decision weight is reset, the snapshot record in the decision weight snapshot table is deleted, and it is determined that only the latest decision weight snapshot is retained in the snapshot table, and the data transmission channel connection between the failed device and the central processing device is restored.

[0041] Example 2 The difference between Example 2 of the present invention and Example 1 is that this example introduces a product reliability improvement system based on equipment failure data analysis.

[0042] Figure 2 A structural schematic diagram of a product reliability improvement method based on equipment failure data analysis of the present invention is given, and a product reliability improvement method based on equipment failure data analysis includes an equipment parameter acquisition module, a failure state identification module, a simulation data transmission module, and a decision weight allocation module: Equipment parameter acquisition module: selects corresponding data acquisition technology according to the communication interface and interaction mode of the equipment in the collaborative operation environment, collects equipment data in real time and uploads it to the data processing center; Failure status identification module: Based on the historical failure data characteristics of all single devices in the equipment network during the historical period when the product function shifted, a failure status identification model is established to identify the status of the collected real-time equipment data; Simulation data transmission module: when the device is judged to be in failure state, it generates simulation device data in real time to replace the failure data and sends it to the central control device; Decision weight allocation module: Calibrate the decision weight of a single device through decision model preset or hierarchical analysis. When the device becomes invalid, record the decision weight snapshot of the current device networking status, and recover all decision weights of the invalid device and distribute them to other normal devices in a balanced manner.

[0043] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0044] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0045] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0046] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0047] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0048] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0049] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0050] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0051] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0052] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A product reliability improvement method based on equipment failure data analysis, characterized in that: The steps include: S1: Analyze the device networking structure of the product, embed the device parameter collection module into the collaborative working environment of the device networking, and collect the real-time data of each single device in the device networking; S2: Select the historical period when the product's function deviation occurred, analyze the historical data of the operation of the single device in the device network during the historical period, define the failure mode characteristics of the device, and establish a device failure state identification model; S3: Obtain the core decision model of the central control device in the device network, and calculate the decision weight of each device in the device network in the decision model based on the hierarchical analysis method; S4: When it is identified that a single device in the device network is in a failed state, the data transmission channel between the failed device and the central control device is disconnected, and the data processing center sends analog data to the central control device; S5: Based on the decision weight ratio of the failed device in the core decision model of the central control device, the decision weight of the normal devices in the device network is redistributed; S6: When it is identified that the state of the failed device has turned to normal, the device data simulation is suspended, and the decision-making power in the core decision-making model of the central control device is reset, and the data transmission channel between the failed device and the central processing device is restored.

2. A method for improving product reliability based on equipment failure data analysis according to claim 1, characterized in that: In S1, the device networking structure of the product is analyzed, and the device parameter collection module is embedded in the collaborative working environment of the device networking. The real-time data of each single device in the device networking is collected, including: Analyze the equipment operation architecture of the product, identify and mark the functional modules of each single device and its functional role positioning in the entire equipment network; Extract the functional role positioning tags of devices in the device network, eliminate the edge devices in the device network, obtain the communication interfaces and interaction methods of all devices in the collaborative working environment of the devices that only retain the information input function and the central control device, and embed the global device parameter acquisition module to collect device data in real time and upload it to the data processing center.

3. A method for improving product reliability based on equipment failure data analysis according to claim 2, characterized in that: In S2, select the historical period when the product has functional deviation, analyze the historical data of the operation of the single device in the device network during the historical period, define the failure mode characteristics of the equipment, and establish the equipment failure state identification model, which specifically includes: Obtain the historical operation log of the product, retrieve the log records where functional abnormalities occur, and mark the start and end time span of consecutive abnormal log records as the functional offset period; According to the actual length of each functional offset period, an offset comparison section is defined at a fixed ratio before and after the endpoint of the functional offset period, and historical operation data of all single devices in the device network during the functional offset period and the offset comparison section are extracted; Calculate the root mean square span of the historical operation data of each single device in the functional deviation period and the deviation control period, screen the single devices whose root mean square spans exceed the preset gradient, and calibrate the data of the screened single devices in the current functional deviation period as failure data; The failure data calibrated for each single device in all functional deviation periods are integrated to form a complete time series failure data set. The time series failure data set of each single device is subjected to fast Fourier transform to extract the frequency domain features of the failure data set. At the same time, the time series dimension features of the failure data set are extracted based on the time series analysis model. The frequency domain features and the time series dimension features are integrated into the failure mode features of the single device. The failure status recognition model of each single device in the equipment network is trained based on a deep neural network, and the trained model is deployed to the data processing center to perform status recognition on the collected real-time equipment data.

4. The method for improving product reliability based on equipment failure data analysis according to claim 3 is characterized in that: In S3, the core decision model of the central control device in the device network is obtained, and the decision weight of each device in the device network in the decision model is calculated based on the hierarchical analysis method, which specifically includes: Check whether the core decision model of the central control device has preset the decision weight of each single device in the device network; If so, the decision weights of individual devices are synchronously calibrated according to the preset weights of the core decision model; If not, the communication signals between each single device and the central control device in the device network and the response signals of the central control device within the set time window are collected, and a hierarchical model is established to calculate the decision weight of the single device. The specific steps are as follows: Set the top level of the hierarchical model to the decision weight of a single device; The second layer of the hierarchical structure model is set as the weight judgment standard, which includes the communication frequency between the single device and the central control device, the control conversion rate after communication, and the average response speed; Set the bottom layer of the hierarchical model to be a single device entity of device networking; The importance scores of the individual judgment criteria in the weight judgment criteria are set to form a judgment matrix, and the ratio of all elements in the judgment matrix to the total of the column is calculated to obtain a normalized matrix; Calculate the average value of each row of the normalized matrix to obtain each weight judgment standard coefficient, bring the communication frequency between each single device and the central control device, the control conversion rate after communication, and the average response speed into the weight judgment standard coefficient, and calculate the weighted score of each single device; The decision weight of each device in the device network in the decision model is calibrated based on the calculation result of the weighted score.

5. The method for improving product reliability based on equipment failure data analysis according to claim 4, characterized in that: In S4, when it is identified that a single device in the device network is in a failed state, the data transmission channel between the failed device and the central control device is disconnected, and the data processing center sends the simulation data to the central control device, which specifically includes: When the failure state identification model of a single device in the data processing center determines that the current device is in a failure state, the bus isolation mechanism is activated to disconnect the real-time data transmission channel between the failed device and the central control device; The data processing center analyzes the historical data trends of failed equipment and generates simulated equipment data in real time based on the time series prediction algorithm and sends it to the central control device.

6. A method for improving product reliability based on equipment failure data analysis according to claim 5, characterized in that: In S5, based on the decision weight ratio of the failed device in the core decision model of the central control device, the decision weight of the normal devices in the device network is redistributed, specifically including: Obtain the decision weight calibrated in the decision model for each device in the device network, and store the calibrated decision weight record as a decision weight allocation snapshot in the decision weight snapshot table of the data processing center; All decision weights of failed devices are recovered, and the decision weights of failed devices are distributed to other normal devices in an equal proportion load balancing manner according to the ratio of the decision weight of each normal device in the device network to the cumulative value of the decision weights of all normal devices.

7. The method for improving product reliability based on equipment failure data analysis according to claim 6, characterized in that: In S6, when it is identified that the state of the failed device has turned to normal, the device data simulation is suspended, and the decision-making power in the core decision model of the central control device is reset, and the data transmission channel between the failed device and the central processing device is restored, which specifically includes: When the failure state identification model of a single device in the data processing center determines that the failed device has returned to a normal state, the data center suspends data simulation of the failed device; According to the device identification of the failed device, retrieve the weight distribution snapshot of the device from the decision weight snapshot table of the data processing center, and reset the decision weight in the device network based on the weight distribution snapshot; After the decision weight is reset, the snapshot record in the decision weight snapshot table is deleted, and the data transmission channel connection between the failed device and the central processing device is restored.

8. A product reliability improvement system based on equipment failure data analysis, used to implement a product reliability improvement method based on equipment failure data analysis as described in any one of claims 1 to 7, characterized in that: It includes equipment parameter acquisition module, failure state identification module, simulation data transmission module and decision weight distribution module: Equipment parameter acquisition module: selects corresponding data acquisition technology according to the communication interface and interaction mode of the equipment in the collaborative operation environment, collects equipment data in real time and uploads it to the data processing center; Failure status identification module: Based on the historical failure data characteristics of all single devices in the equipment network during the historical period when the product function shifted, a failure status identification model is established to identify the status of the collected real-time equipment data; Simulation data transmission module: when the device is judged to be in failure state, it generates simulation device data in real time to replace the failure data and sends it to the central control device; Decision weight allocation module: Calibrate the decision weight of a single device through decision model preset or hierarchical analysis. When the device becomes invalid, record the decision weight snapshot of the current device networking status, and recover all decision weights of the invalid device and distribute them to other normal devices in a balanced manner.

Images