Equipment screening method and device, computer equipment and storage medium

Abstract

The invention relates to an equipment screening method and device, computer equipment and a storage medium, and relates to the field of artificial intelligence. The method comprises the steps that the predicted use cycle of at least one piece of to-be-detected equipment is obtained, and the equipment models of the to-be-detected equipment are different; determining a failure probability prediction function corresponding to each to-be-detected device, and according to the failure probability prediction function corresponding to each to-be-detected device, determining a predicted failure probability of each to-be-detected device in the predicted use period; and taking the to-be-detected equipment with the minimum predicted failure probability in the to-be-detected equipment as target equipment. According to the method, the accuracy of failure probability prediction of the to-be-detected equipment can be effectively improved, so that the target equipment obtained by screening according to the predicted failure probability can better meet actual requirements.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a device screening method, apparatus, computer equipment and storage medium. Background Art

[0002] With the continuous development and progress of society, the scale of enterprises is also constantly expanding. Therefore, the amount of equipment purchased and put into use by enterprises each year is gradually increasing. However, some equipment fails shortly after being put into use, which leads to the need for enterprises to increase investment costs and repair and replace the failed equipment.

[0003] Since the probability of failure of different models of the same type of equipment during their service life varies, how to ensure that the probability of failure of the equipment put into use during its service life is minimized has become a problem that companies are very concerned about. Summary of the invention

[0004] Based on this, it is necessary to provide a device screening method, apparatus, computer equipment and storage medium to address the above technical issues.

[0005] In a first aspect, the present application provides a device screening method. The method comprises:

[0006] Obtaining an estimated service life of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0007] Determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected;

[0008] The device to be tested with the smallest predicted failure probability among all the devices to be tested is taken as the target device.

[0009] In one embodiment, determining the failure probability prediction function corresponding to each device to be detected includes:

[0010] For each device to be detected, determine a reference device corresponding to the device to be detected, wherein the device model of the reference device is the same as the device model of the device to be detected;

[0011] Determine the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the censoring quantity corresponding to the failed device; wherein the censoring quantity is used to represent the number of devices that are censored from the reference device when the failed device fails;

[0012] According to the estimated values ​​of the prediction parameters, the failure probability prediction function corresponding to each device to be tested is determined.

[0013] In one embodiment, the failure time and the number of censoring errors are determined as follows:

[0014] Conduct simulation experiments on reference equipment according to the preset ideal number of equipment, minimum number of equipment and preset experimental period;

[0015] If the simulation experiment meets the experimental termination conditions, the failure time of the failed equipment in the reference equipment is determined, and the number of deletions corresponding to the failed equipment is determined.

[0016] In one embodiment, the experiment termination conditions include:

[0017] The actual experimental cycle of the reference device is less than the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the ideal device number; or, the actual experimental cycle of the reference device is greater than or equal to the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the minimum device number.

[0018] In one embodiment, determining the number of deletions corresponding to the failed device includes:

[0019] Determine the failure sequence number corresponding to the failed device;

[0020] Determine the predicted quantity corresponding to the failure sequence number based on the corresponding relationship between the sequence number and the preset quantity;

[0021] The predicted quantity corresponding to the failure sequence number is used as the deleted quantity corresponding to the failed equipment.

[0022] In one embodiment, determining the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the number of deletions corresponding to the failed device includes:

[0023] Based on the candidate estimation method, determining the candidate parameter estimation value corresponding to the candidate estimation method according to the failure time of the failed device and the number of deletions corresponding to the failed device;

[0024] determining the estimated accuracy of candidate parameter estimates;

[0025] The candidate parameter estimate with the highest estimation accuracy is used as the predicted parameter estimate.

[0026] In one embodiment, the candidate estimation method includes at least one of a maximum likelihood estimation method and a Bayesian estimation method.

[0027] In a second aspect, the present application also provides a device for screening equipment. The device comprises:

[0028] A cycle acquisition module, used to acquire an estimated use cycle of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0029] A failure prediction module is used to determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected;

[0030] The device screening module is used to select the device to be detected with the smallest predicted failure probability among all the devices to be detected as the target device.

[0031] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0032] Obtaining an estimated service life of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0033] Determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected;

[0034] The device to be tested with the smallest predicted failure probability among all the devices to be tested is taken as the target device.

[0035] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0036] Obtaining an estimated service life of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0037] Determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected;

[0038] The device to be tested with the smallest predicted failure probability among all the devices to be tested is taken as the target device.

[0039] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0040] Obtaining an estimated service life of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0041] Determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected;

[0042] The device to be tested with the smallest predicted failure probability among all the devices to be tested is taken as the target device.

[0043] The above-mentioned equipment screening method, device, computer equipment and storage medium first determine the failure probability prediction function corresponding to each device to be detected, and then determine the predicted failure probability of each device to be detected within the expected use cycle through the failure probability prediction function corresponding to each device to be detected, and use the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device. According to the above content, the present application can predict the possibility of equipment failure of the device to be detected within the expected use cycle according to the predicted failure probability of the device, and then, by using the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device, it prevents the device from failing soon after it is put into use, thereby reducing the cost investment of the enterprise for repairing and replacing failed equipment, and each failure probability prediction function in the present application is determined according to the estimated value of the prediction parameter of each device to be detected, therefore, the predicted failure probability of each device to be detected within the expected use cycle determined by each failure probability prediction function can accurately reflect the actual situation of each device to be detected, and ensure the accuracy of determining the predicted failure probability of each device to be detected. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 An application environment diagram of a device screening method provided in an embodiment of the present application;

[0045] Figure 2 A flow chart of a device screening method provided in an embodiment of the present application;

[0046] Figure 3 A flowchart of the steps of determining a failure probability prediction function provided in an embodiment of the present application;

[0047] Figure 4 A flowchart of the steps for determining failure time and deletion quantity provided in an embodiment of the present application;

[0048] Figure 5 An example diagram of the first case of the experimental termination condition provided in the embodiment of the present application;

[0049] Figure 6 An example diagram of the second situation of the experimental termination condition provided in the embodiment of the present application;

[0050] Figure 7 An example diagram of the third situation of the experimental termination condition provided in the embodiment of the present application;

[0051] Figure 8 A flowchart of the steps for determining the number of deletions of failed devices provided in an embodiment of the present application;

[0052] Fig. 9 A flowchart of the steps for determining an estimated value of a prediction parameter provided in an embodiment of the present application;

[0053] Fig.10 A flowchart of another device screening method provided in an embodiment of the present application;

[0054] Fig.11 A structural block diagram of a first device screening device provided in an embodiment of the present application;

[0055] Fig.12 A structural block diagram of a second device screening apparatus provided in an embodiment of the present application;

[0056] Fig.13 A structural block diagram of a third device screening device provided in an embodiment of the present application;

[0057] Fig.14 A structural block diagram of a fourth device screening device provided in an embodiment of the present application;

[0058] Fig.15 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0060] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. In the description of the present application, the description of the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms is not necessarily for the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples without contradicting each other.

[0061] Based on the above situation, the device screening method provided in the embodiment of the present application can be applied to Figure 1 In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in FIG. Figure 1 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a data block. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The data block of the computer device is used to store the acquisition data of the device screening method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a device screening method is implemented.

[0062] The present application discloses a device screening method, apparatus, computer equipment and storage medium thereof, which may specifically include the following contents: firstly, determining the failure probability prediction function corresponding to each device to be detected, and then determining the predicted failure probability of each device to be detected within the expected use period through the failure probability prediction function corresponding to each device to be detected, and taking the device to be detected with the smallest predicted failure probability among all the devices to be detected as the target device.

[0063] In an exemplary embodiment, Figure 2 As shown, Figure 2 A flow chart of a device screening method provided in an embodiment of the present application, providing a device screening method, which is applied to Figure 1 The computer device in the example is used to illustrate, including the following steps 201 to 203. Among them:

[0064] Step 201: Obtain an estimated usage period of at least one device to be detected.

[0065] Among them, the device types of each device to be detected are the same, but the device models of each device to be detected are different; for example, there are three devices to be detected, and the device types of the three devices to be detected can all be automatic resource access devices, but the device models of the three devices to be detected are model A, model B and model C respectively.

[0066] Among them, the expected use period refers to the predetermined time period during which the equipment to be tested is expected to be used; further, the expected use period can be determined based on the experience of the staff and the actual needs of the equipment to be tested, and the value of the expected use period is not limited here.

[0067] It should be noted that each device to be tested can be a device of the same type with different models produced by different manufacturers, or can be a device of the same type with different models produced by the same manufacturer in different batches. For example, the device to be tested can be an automatic resource access device. If there are four devices to be tested, the four devices to be tested are automatic resource access device 1, automatic resource access device 2, automatic resource access device 3 and automatic resource access device 4, wherein automatic resource access device 1 is produced by manufacturer a and its model is 001; automatic resource access device 2 is produced by manufacturer b and its model is 002, or both automatic resource access device 3 and automatic resource access device 4 are produced by manufacturer c, wherein the model of automatic resource access device 3 is 003 and the model of automatic resource access device 4 is 004.

[0068] As an example, the expected usage cycle of the device to be detected can be specified according to actual conditions. For example, the device to be detected is an automatic resource access device. When it is necessary to determine the expected usage cycle of the automatic resource access device, if the automatic resource access device needs to be used in a certain area for ten years, then ten years will be used as the expected usage cycle of the automatic deposit and withdrawal device.

[0069] As another example, the expected usage cycle corresponding to the device to be detected can be determined based on the historical usage cycle of a reference device corresponding to the device to be detected, wherein the device type of the reference device is the same as the device type of the device to be detected, and the device model of the reference device is the same as the device model of the device to be detected. Specifically, the historical usage cycle of at least one reference device is determined, and the historical usage cycles of each reference device are averaged, and the result is used as the expected usage cycle corresponding to the device to be detected.

[0070] As another example, the expected use period of the device to be detected can be determined based on the device information of the device to be detected, wherein the device information of the device to be detected may include but is not limited to: the manufacturing material of the device to be detected (for example, alloy, iron, copper, etc.), the manufacturing process of the device to be detected (for example, cutting, drilling, milling, etc.), the environment of the device to be detected (for example, temperature, humidity, dust, etc.), etc. Specifically, the device information of the device to be detected is input into the cycle prediction model to obtain the output result of the cycle prediction model, which is the expected use period of the device to be detected.

[0071] Among them, the training process of the cycle prediction model may specifically include the following contents: obtaining sample equipment and equipment information of the sample equipment, marking the sample usage cycle of the equipment information based on the historical experience of the staff, obtaining the equipment information marked with the sample usage cycle, and inputting the attribute information of the sample equipment marked with the sample usage cycle into the cycle prediction model to realize the training of the cycle prediction model and obtain the trained cycle prediction model.

[0072] Step 202, determining the failure probability prediction function corresponding to each device to be detected, and determining the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected.

[0073] Wherein, each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be tested;

[0074] It should be noted that when it is necessary to determine the failure probability prediction function corresponding to each device to be detected, the following may be specifically included: for each device to be detected, determine the reference device corresponding to the device to be detected; further, based on the failure time of the failed devices in the reference devices, and the number of deletions corresponding to the failed devices, determine the estimated values ​​of the prediction parameters of the device to be detected; and then, based on the estimated values ​​of the prediction parameters of the device to be detected, determine the failure probability prediction function corresponding to the device to be detected.

[0075] To further explain, when it is necessary to determine the predicted failure probability of each device to be detected within the expected use cycle based on the failure probability prediction function corresponding to each device to be detected, the following may be specifically included: the expected use cycle can be substituted into the failure probability prediction function corresponding to each device to be detected to obtain the calculation result of the failure probability prediction function corresponding to each device to be detected based on the expected use cycle, and the calculation result is the predicted failure probability of each device to be detected within the expected use cycle.

[0076] Step 203: The device to be detected with the smallest predicted failure probability among all the devices to be detected is taken as the target device.

[0077] It should be noted that since the size of the predicted failure probability corresponding to the equipment to be tested reflects the possibility of equipment failure in the expected use cycle of the equipment to be tested, the smaller the predicted failure probability, the smaller the possibility of equipment failure in the expected use cycle of the equipment to be tested. Conversely, the larger the predicted failure probability, the greater the possibility of equipment failure in the expected use cycle of the equipment to be tested. Therefore, in order to ensure that the probability of equipment failure of the target equipment in the expected use cycle is minimized when the enterprise puts the target equipment into use again, the equipment to be tested with the smallest predicted failure probability among all the equipment to be tested will be used as the target equipment.

[0078] The equipment failure phenomenon may include, but is not limited to: the equipment fails but does not affect the operation of the equipment, the equipment fails and affects the operation of the equipment, etc.

[0079] In one embodiment of the present application, when it is necessary to determine the target device, the predicted failure probabilities corresponding to the devices to be detected can be sorted from small to large. After sorting, the device to be detected corresponding to the predicted failure probability ranked first is the device to be detected with the smallest predicted failure probability. Therefore, the device to be detected corresponding to the predicted failure probability ranked first is selected as the target device.

[0080] In one embodiment of the present application, if there are three devices to be detected, namely: device a1 to be detected, device a2 to be detected and device a3 to be detected, wherein the predicted failure probability corresponding to device a1 to be detected is 0.56, the predicted failure probability corresponding to device a2 to be detected is 0.62, and the predicted failure probability corresponding to device a3 to be detected is 0.31. Since 0.31<0.56<0.62, the device a3 to be detected corresponding to the predicted failure probability of 0.31 is the target device.

[0081] The above-mentioned equipment screening method first determines the failure probability prediction function corresponding to each device to be detected, and then determines the predicted failure probability of each device to be detected within the expected use cycle through the failure probability prediction function corresponding to each device to be detected, and takes the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device. According to the above content, the present application can predict the possibility of equipment failure of the device to be detected within the expected use cycle according to the predicted failure probability of the device, and then, by taking the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device, it prevents the device from failing soon after it is put into use, thereby reducing the cost investment of the enterprise for repairing and replacing the failed equipment, and each failure probability prediction function in the present application is determined according to the estimated value of the prediction parameter of each device to be detected, therefore, the predicted failure probability of each device to be detected within the expected use cycle determined by each failure probability prediction function can accurately reflect the actual situation of each device to be detected, and ensure the accuracy of determining the predicted failure probability of each device to be detected. In one embodiment, since the probability of failure of the same type of equipment of different models in the use cycle is different. In order to ensure that the probability of failure of the equipment put into use is minimized during its service life, the computer equipment of the present application can be Figure 3 The method shown in the figure determines the failure probability prediction function corresponding to each device to be detected, including the following steps 301 to 303. Among them:

[0082] Step 301: for each device to be detected, determine a reference device corresponding to the device to be detected.

[0083] It should be noted that the device model of the reference device is the same as the device model of the device to be detected, and the device type of the reference device is the same as the device type of the device to be detected. Therefore, when it is necessary to determine the reference device corresponding to the device to be detected, a device with the same device type and the same device model as the device to be detected can be used as the reference device.

[0084] It should be noted that the reference device and the device to be tested may be produced by the same manufacturer or by different manufacturers.

[0085] To further explain, if the reference device and the device to be tested are produced by the same manufacturer, a device produced by the manufacturer of the device to be tested that has the same device type and the same device model as the device to be tested will be used as the reference device; for example, it is determined that the device to be tested is produced by manufacturer x and the device model is 125, so the reference device can be any device produced by manufacturer x and with the device model 125.

[0086] To further explain, if the reference device and the device to be tested are produced by different manufacturers, a device with the same device type and device model as the device to be tested and produced by other manufacturers different from the manufacturer of the device to be tested will be used as the reference device; for example, it is determined that the device to be tested is produced by manufacturer x and the device model is 126, so the reference device can be any device produced by manufacturer y and the device model is 126.

[0087] Step 302: Determine the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the number of deletions corresponding to the failed device.

[0088] The censored quantity is used to indicate the number of devices that are censored from the reference device when the failed device fails.

[0089] In one embodiment of the present application, when it is necessary to determine the estimated value of the prediction parameter, the following may be specifically included: a simulation experiment is performed on the reference device according to a preset ideal number of devices, a minimum number of devices and a preset experimental period; if the simulation experiment satisfies the experimental termination condition, the failure time of the failed device in the reference device and the number of deletions corresponding to the failed device can be determined according to the experimental termination condition; further, according to the failure time of the failed device in the reference device and the number of deletions corresponding to the failed device, a candidate estimation method is used to perform parameter estimation on the parameters of the failure probability prediction function to obtain candidate parameter estimation values ​​corresponding to the candidate estimation method; and then, the candidate parameter estimation values ​​are evaluated for accuracy, and the prediction parameter estimation values ​​are determined from the candidate parameter estimation values.

[0090] Among them, the experiment termination condition can be: the actual experiment cycle of the reference device is less than the preset experiment cycle, and the failure number of failed devices in the reference device is greater than or equal to the ideal device number; or, the actual experiment cycle of the reference device is greater than or equal to the preset experiment cycle, and the failure number of failed devices in the reference device is greater than or equal to the minimum device number.

[0091] Step 303: Determine the failure probability prediction function corresponding to each device to be tested according to the estimated value of the prediction parameter.

[0092] The failure probability prediction function may be a cumulative density function of a Rayleigh distribution. Specifically, the failure probability prediction function may be calculated as shown in formula (1). The calculation formula (1) is as follows:

[0093]

[0094] Here, x refers to the expected service life of the equipment to be tested, and σ refers to the estimated value of the prediction parameter.

[0095] Further explanation: according to the failure probability prediction function, under the premise of a fixed expected usage cycle, the larger the estimated value of the prediction parameter, the lower the predicted failure probability of the device to be tested; conversely, the smaller the estimated value of the prediction parameter, the higher the predicted failure probability of the device to be tested; therefore, in order to ensure the accuracy of the predicted failure probability of the device to be tested determined according to the failure probability prediction function, it is necessary to process the failure time of the failed devices in the reference equipment and the number of deletions corresponding to the failed devices to obtain the estimated value of the prediction parameter.

[0096] The above-mentioned device screening method, by determining the failure probability prediction function corresponding to each device to be detected, realizes determining the predicted failure probability of each device to be detected within the expected use period according to the failure probability prediction function corresponding to each device to be detected, thereby ensuring that the target device can be successfully determined from the devices to be detected.

[0097] In one embodiment, the computer device of the present application can be Figure 4 The method shown in the figure determines the failure time and the number of deletions, including the following steps 401 and 402. Among them:

[0098] Step 401 , performing a simulation experiment on a reference device according to a preset ideal device quantity, a minimum device quantity and a preset experiment period.

[0099] The ideal number of devices refers to the maximum number of failed devices in the reference devices; the minimum number of devices refers to the minimum number of failed devices in the reference devices, and the ideal number of devices is greater than the minimum number of devices. Furthermore, the ideal number of devices, the minimum number of devices, and the preset experimental period are all pre-set based on the historical experience of the staff.

[0100] In one embodiment of the present application, if the device to be detected is an automatic resource access device of model A, since the device type of the reference device is the same as the device type of the device to be detected, the reference device is also an automatic resource access device of model A. Then, when it is necessary to perform a simulation experiment on the reference device, the following contents may be specifically included: determining that the preset ideal number of devices is m, the minimum number of devices is k, and the preset experimental period is T; based on the preset ideal number of devices being m, the minimum number of devices being k, and the preset experimental period being T, in the experimental scenario of the reference device, the operation and use of the reference device are simulated to achieve a simulation experiment on the reference device.

[0101] Step 402: If the simulation experiment meets the experiment termination condition, the failure time of the failed device in the reference device is determined, and the deletion quantity corresponding to the failed device is determined.

[0102] Among them, the experiment termination condition can be: the actual experiment cycle of the reference device is less than the preset experiment cycle, and the failure number of failed devices in the reference device is greater than or equal to the ideal device number; or, the actual experiment cycle of the reference device is greater than or equal to the preset experiment cycle, and the failure number of failed devices in the reference device is greater than or equal to the minimum device number.

[0103] In one embodiment of the present application, Figure 5 As shown, the first case of the specified experiment termination condition is: during the simulation experiment on the reference device, if the Kth reference device fails after the preset experimental period T, and the predetermined minimum number of devices is K, the simulation experiment ends when the Kth reference device fails. Figure 5 Where X1 refers to the experimental time when the first reference device fails; X2 refers to the experimental time when the second reference device fails; k refers to the experimental time when the Kth reference device fails; X m refers to the experimental time when the Mth reference device fails; T refers to the preset experimental period; R1 refers to the number of devices that are deleted from the reference device when the first reference device fails; R2 refers to the number of devices that are deleted from the reference device when the second reference device fails; R1 * Refers to the number of devices that are censored from the reference devices when the Kth reference device fails.

[0104] in, Where n is the number of reference devices and k is the minimum number of devices.

[0105] In one embodiment of the present application, Figure 6 As shown, the second case of the specified experiment termination condition is: in the process of conducting a simulation experiment on the reference equipment, if the Kth reference equipment fails before the preset experiment period T, and the Mth reference equipment fails after the preset experiment period T, and the predetermined minimum number of equipment is K, and the ideal number of equipment is M, therefore, the simulation experiment ends when the experiment time reaches the preset experiment period T. Figure 6 Where X1 refers to the experimental time when the first reference device fails; X2 refers to the experimental time when the second reference device fails; k refers to the experimental time when the Kth reference device fails; X m Refers to the experimental time when the Mth reference device fails; X D Refers to the experimental time when the Dth reference device fails; X D+1refers to the experimental time when the D+1th reference device fails; T refers to the preset experimental period; R1 refers to the number of devices that are deleted from the reference device when the first reference device fails; R2 refers to the number of devices that are deleted from the reference device when the second reference device fails; R D Refers to the number of devices that are censored from the reference device when the Dth reference device fails; R1 * Refers to the number of devices that are censored from the reference device when the Kth reference device fails; R2 * Refers to the number of devices that are censored from the reference device when the D+1th reference device fails.

[0106] in, Where n is the number of reference devices, and D is the serial number of the reference device that failed before time T.

[0107] In one embodiment of the present application, Figure 7 As shown, the third case of the specified experiment termination condition is: in the process of conducting a simulation experiment on the reference device, if the Mth reference device fails before the preset experiment period T, and the predetermined number of ideal devices is M, the simulation experiment ends when the Mth reference device fails. Figure 7 Where X1 refers to the experimental time when the first reference device fails; X2 refers to the experimental time when the second reference device fails; k refers to the experimental time when the Kth reference device fails; X m refers to the experimental time when the Mth reference device fails; T refers to the preset experimental period; R1 refers to the number of devices that are deleted from the reference device when the first reference device fails; R2 refers to the number of devices that are deleted from the reference device when the second reference device fails; R k Refers to the number of devices that are deleted from the reference device when the Kth reference device fails; R3 * Refers to the number of devices that are censored from the reference device when the Mth reference device fails.

[0108] in, Where n is the number of reference devices and m is the number of ideal devices.

[0109] It should be noted that, since the experiment termination conditions include multiple situations, the deletion numbers corresponding to failed devices under different experiment termination conditions are different. Therefore, when it is necessary to determine the deletion number corresponding to failed devices, the following contents may also be included: determining the failure sequence number corresponding to the failed device, and filtering according to the failure sequence number from the correspondence between the sequence number and the preset number according to the pre-set correspondence between the sequence number and the preset number, filtering out the preset number corresponding to the failure sequence number, and further, using the predicted number corresponding to the failure sequence number as the deletion number corresponding to the failed device.

[0110] The above-mentioned equipment screening method, by conducting simulation experiments on reference equipment, realizes the determination of the failure time of failed equipment in the reference equipment and the number of deletions corresponding to the failed equipment; it provides a data basis for determining the estimated values ​​of the prediction parameters, and ensures the accuracy of the failure probability prediction of the equipment to be tested based on the failure probability prediction function containing the estimated values ​​of the prediction parameters.

[0111] In an exemplary embodiment, when it is necessary to determine the number of deletions corresponding to the failed device, the following method can be used: Figure 8 The method shown includes the following steps 801 to 803. Among them:

[0112] Step 801, determine the failure sequence number corresponding to the failed device.

[0113] The failure sequence numbering refers to determining the number of each failed device according to the order of the failure time corresponding to each failed device.

[0114] For example, if there are three failed devices, namely: failed device 1, failed device 2 and failed device 3, among which the failure time of failed device 1 is September 20, 2023, the failure time of failed device 2 is July 15, 2023, and the failure time of failed device 3 is October 9, 2023. Since the order of failure time is: July 15, 2023, September 20, 2023, October 9, 2023, it can be known that the failure sequence number of failed device 2 is 1, the failure sequence number of failed device 1 is 2, and the failure sequence number of failed device 3 is 3.

[0115] Step 802: Determine the predicted quantity corresponding to the failure sequence number based on the correspondence between the sequence number and the preset quantity.

[0116] It should be noted that the correspondence between the sequence number and the preset quantity refers to the correspondence determined based on the historical experience of the staff and the actual situation, and the preset quantities corresponding to different sequence numbers are recorded in the correspondence; therefore, when it is necessary to determine the predicted quantity corresponding to the failure sequence number, the sequence number corresponding to the failure sequence number can be determined in the correspondence between the sequence number and the preset quantity, and the preset quantity corresponding to the sequence number is the predicted quantity corresponding to the failure sequence number.

[0117] Further explanation: the correspondence between the sequence number and the preset number can also be used to represent the functional relationship between the sequence number and the preset number, and the functional relationship is used to determine the number of devices to be deleted from the reference devices when the failed device corresponding to the failure sequence number fails.

[0118] In an embodiment of the present application, the corresponding relationship between the sequence number and the preset number can be: R1=nm, R i =0, i=2, 3, ..., m, where R i is the preset number corresponding to the sequence number i, n is the number of reference devices, and m is the ideal number of devices;

[0119] In another embodiment of the present application, the corresponding relationship between the sequence number and the preset number may also be: R i =0if i is even, which means that if the sequence number i is an odd number, the preset number corresponding to the sequence number i If the sequence number i is an even number, the preset number R corresponding to the sequence number i i =0.

[0120] In another embodiment of the present application, the corresponding relationship between the sequence number and the preset number can also be: m =nm,R i =0, i=1, 2, ..., m-1, where R i is the preset number corresponding to the sequence number i, n is the number of reference devices, and m is the ideal number of devices.

[0121] As an example, if the correspondence between the sequence number and the preset number is correspondence 1, the number of reference devices n=20, the number of ideal devices m=12, if the failure sequence number is 1, then according to the correspondence 1, it can be known that R1=0, that is, the predicted number corresponding to the failure sequence number 1 is 8; if the failure sequence number is 3, then according to the correspondence 1, it can be known that R3=0, that is, the predicted number corresponding to the failure sequence number 3 is 0.

[0122] As an example, if the correspondence between the sequence number and the preset number is correspondence 2, the number of reference devices n = 30, the number of ideal devices m = 20, and if the failure sequence number is 5, then according to correspondence 2, it can be known that R5 = 1, that is, the predicted number corresponding to the failure sequence number 5 is 1; if the failure sequence number is 6, then according to correspondence 2, it can be known that R6 = 0, that is, the predicted number corresponding to the failure sequence number 6 is 0.

[0123] As an example, if the correspondence between the sequence number and the preset number is correspondence 3, the number of reference devices n = 30, the number of ideal devices m = 21, and if the failure sequence number is 10, then according to correspondence 3, it can be known that R 10 =0, that is, the predicted quantity corresponding to failure sequence number 10 is 0; if the failure sequence number is 21, then according to the corresponding relationship 3, we know that R 21 =9, that is, the predicted quantity corresponding to failure sequence number 21 is 9.

[0124] As an example, if the correspondence between the sequence number and the preset number is correspondence 4, the number of reference devices n = 28, if the failure sequence number is 3, then according to correspondence 4, it can be known that R3 = 3, that is, the predicted number corresponding to the failure sequence number 3 is 3; if the failure sequence number is 9, then according to correspondence 4, it can be known that R9 = 9, that is, the predicted number corresponding to the failure sequence number 9 is 9.

[0125] Step 803: Use the predicted quantity corresponding to the failure sequence number as the deleted quantity corresponding to the failed device.

[0126] The deletion quantity is used to indicate the number of devices deleted from the reference device. For example, if the deletion quantity is 2, it indicates that 2 devices are deleted from the reference device.

[0127] In one embodiment of the present application, if the failure sequence number corresponding to the failed device is 3, and the predicted number corresponding to the failure sequence number 3 is determined to be 8, since the predicted number corresponding to the failure sequence number is the deleted number corresponding to the failed device, the deleted number corresponding to the failed device is 8.

[0128] The above-mentioned equipment screening method determines the deletion quantity corresponding to the failed equipment through the correspondence between the sequential number and the preset quantity, which can ensure that the deletion quantity of the failed equipment meets the quantity requirement for failure prediction of the equipment to be tested, thereby achieving the effect of improving the accuracy of failure prediction of the equipment to be tested.

[0129] In an exemplary embodiment, when determining the prediction parameter estimate, the following method can be used: Fig. 9 The method shown includes the following steps 901 to 903. Among them:

[0130] Step 901 , based on the candidate estimation method, according to the failure time of the failed device and the number of deletions corresponding to the failed device, determine the candidate parameter estimation value corresponding to the candidate estimation method.

[0131] It should be noted that when it is necessary to determine the candidate parameter estimation values ​​corresponding to the candidate estimation method, it is necessary to use the candidate estimation method to perform parameter estimation on the parameters of the failure probability prediction function based on the failure time of the failed equipment and the number of deletions corresponding to the failed equipment, and obtain the candidate parameter estimation values ​​corresponding to the candidate estimation method.

[0132] It is further specified that the candidate estimation method includes at least one of a maximum likelihood estimation method and a Bayesian estimation method.

[0133] In one embodiment of the present application, if the candidate estimation method is the maximum likelihood estimation method, and the failure probability prediction function is the cumulative density function of the Rayleigh distribution, then when it is necessary to determine the candidate parameter estimation value corresponding to the candidate estimation method, it may specifically include the following contents: determine the likelihood function of the Rayleigh distribution, perform a logarithmic operation on the likelihood function of the Rayleigh distribution to obtain a first operation result, perform a zeroing operation on the first operation result, that is, set the first operation result to zero, solve the equation, and obtain the candidate parameter estimation value corresponding to the candidate estimation method.

[0134] As an example, in the first case of the experimental termination condition, that is, in the process of simulating the reference device, if the Kth reference device fails after the preset experimental period T, and the predetermined minimum number of devices is K, the likelihood function of the Rayleigh distribution can be determined as shown in formula (2):

[0135]

[0136] In the second case of the experimental termination condition, that is, in the process of simulating the reference device, if the Kth reference device fails before the preset experimental period T, and the Mth reference device fails after the preset experimental period T, and the predetermined minimum number of devices is K and the ideal number of devices is M, the likelihood function of the Rayleigh distribution can be determined as shown in formula (3):

[0137]

[0138] In the third case of the experimental termination condition, that is, in the process of simulating the reference device, if the Mth reference device fails before the preset experimental period T, and the predetermined number of ideal devices is M, the likelihood function of the Rayleigh distribution can be determined as shown in formula (4):

[0139]

[0140] Furthermore, by unifying the likelihood functions in the three cases, we can obtain the likelihood function L(σ) of the Rayleigh distribution as shown in formula (5):

[0141]

[0142] Where K represents the coefficient, J represents the number of failed devices, i represents the i-th failed device, and x i represents the failure time of failed device i, R i represents the number of deletions of failed equipment i, according to Figure 6 From the content in the text, we can know that the value of W(σ) can be: 0 and Where σ represents the candidate parameter estimate of the cumulative density function, T represents the experimental time of the simulation experiment on the reference device, n represents the number of reference devices in the simulation experiment, D represents the number of failed devices that failed before the experimental time T, and the value of D is greater than the minimum number of devices and less than the ideal number of devices.

[0143] Furthermore, the likelihood function of the Rayleigh distribution is logarithmically operated to obtain the first operation result l(σ) as shown in formula (6):

[0144]

[0145] Where K represents the coefficient, J represents the number of failed devices, i represents the i-th failed device, and x i represents the failure time of failed device i, R i represents the number of deletions of failed equipment i, according to Figure 6 From the content in the text, we can know that the value of W(σ) can be: 0 and Where σ represents the candidate parameter estimate of the cumulative density function, T represents the experimental time of the simulation experiment on the reference device, n represents the number of reference devices in the simulation experiment, D represents the number of failed devices that failed before the experimental time T, and the value of D is greater than the minimum number of devices and less than the ideal number of devices.

[0146] Furthermore, the first operation result is assigned a zero value, that is, the first operation result is set to zero, and the equation is solved to obtain the candidate parameter estimation value corresponding to the candidate estimation method. Since the value of W(σ) can be: 0 and Therefore, there are two candidate parameter estimates:

[0147] The candidate parameter estimates in the first and third cases of the experimental termination conditions are shown in formula (7):

[0148]

[0149] Where J represents the number of failed devices, i represents the i-th failed device, and x i represents the failure time of failed device i, R i represents the number of deletions of failed equipment i.

[0150] The candidate parameter estimates in the second case of the experimental termination condition are shown in formula (8):

[0151]

[0152] Where J represents the number of failed devices, i represents the i-th failed device, and x i represents the failure time of failed device i, R i represents the number of deletions of failed equipment i, and T represents the experimental time of the simulation experiment on the reference equipment.

[0153] In one embodiment of the present application, if the candidate estimation method is the Bayesian estimation method, and the failure probability prediction function is the cumulative density function of the Rayleigh distribution, then when it is necessary to determine the candidate parameter estimation value corresponding to the candidate estimation method, it may specifically include the following contents: determining the prior distribution, joint distribution and posterior distribution of the cumulative density function parameters of the Rayleigh distribution; determining the candidate parameter estimation value corresponding to the candidate estimation method based on the square error loss function, the generalized entropy loss function and the equilibrium loss function.

[0154] Among them, the balanced loss function includes two cases, one is the balanced loss function based on the square error loss function, and the other is the balanced loss function based on the generalized entropy loss function.

[0155] As an example, the prior distribution for determining the cumulative density function parameters of the Rayleigh distribution is shown in formula (9):

[0156]

[0157] Among them, σ represents the candidate parameter estimate of the cumulative density function, and σ>0, α and β both represent hyperparameters, and α>0, β>0, which can be set to α=4.5, β=4.5 according to the historical experience of the staff, and Γ(.) represents the square root inverse gamma distribution.

[0158] Furthermore, the joint distribution L(data, σ) of the cumulative density function parameters of the Rayleigh distribution is determined as shown in formula (10):

[0159]

[0160] Where σ represents the candidate parameter estimate of the cumulative density function, and σ>0, α and β both represent hyperparameters, and α>0, β>0, J represents the number of failed devices, i represents the i-th failed device, and x i represents the failure time of failed device i, R i represents the number of deletions of failed equipment i, Where T represents the experimental time of the simulation experiment on the reference device, β represents the hyperparameter, n represents the number of reference devices in the simulation experiment, D represents the number of failed devices that failed before the experimental time T, and the value of D is greater than the minimum number of devices and less than the ideal number of devices.

[0161] Furthermore, the posterior distribution π(σ|data) of the candidate parameter estimate of the Rayleigh distribution is determined as shown in formula (11):

[0162]

[0163] Where σ represents the candidate parameter estimate of the cumulative density function, and σ>0, α and β both represent hyperparameters, and α>0, β>0, J represents the number of failed devices, i represents the i-th failed device, and x i represents the failure time of failed device i, R i represents the number of deletions of failed equipment i, Where T represents the experimental time of the simulation experiment on the reference device, β represents the hyperparameter, n represents the number of reference devices in the simulation experiment, D represents the number of failed devices that failed before the experimental time T, and the value of D is greater than the minimum number of devices and less than the ideal number of devices.

[0164] As an example, the candidate parameter estimation value corresponding to the candidate estimation method is determined according to the square error loss function, where the square error loss function is shown in formula (12):

[0165]

[0166] Among them, φ represents the true value, Represents the predicted value.

[0167] Furthermore, the candidate parameter estimation values ​​corresponding to the candidate estimation method are obtained as shown in formula (13):

[0168]

[0169] in, Represents the candidate parameter estimates obtained by Bayesian estimation of the squared error loss function.

[0170] As another example, the candidate parameter estimation value corresponding to the candidate estimation method is determined according to the generalized entropy loss function, where the generalized entropy loss function is shown in formula (14):

[0171]

[0172] Among them, φ represents the true value, Represents the predicted value.

[0173] Furthermore, the candidate parameter estimation values ​​corresponding to the candidate estimation method are obtained as shown in formula (15):

[0174]

[0175] in, represents the candidate parameter estimates obtained by Bayesian estimation of the generalized entropy loss function.

[0176] As another example, the candidate parameter estimation values ​​corresponding to the candidate estimation method are determined according to the balanced loss function, where the balanced loss function is: If the balanced loss function is a balanced loss function based on the square error loss function, then the balanced loss function The expression of is shown in formula (16):

[0177]

[0178] Among them, φ represents the true value, represents the predicted value, ω represents the coefficient, and φ0 represents the initial value.

[0179] Furthermore, the candidate parameter estimation values ​​corresponding to the candidate estimation method are obtained as shown in formula (17):

[0180]

[0181] in, represents the candidate parameter estimate obtained by Bayesian estimation of the equalization loss function based on the square error loss function, ω represents the coefficient, represents the candidate parameter estimate corresponding to the maximum likelihood estimate, and E(σ|data) represents the candidate parameter estimate obtained by Bayesian estimation of the squared error loss function.

[0182] If the equilibrium loss function is an equilibrium loss function based on the generalized entropy loss function, then the equilibrium loss function The expression of is shown in formula (18):

[0183]

[0184] Among them, φ represents the true value, represents the predicted value, ω represents the coefficient, φ0 represents the initial value, and q represents the hyperparameter.

[0185] Furthermore, the candidate parameter estimation values ​​corresponding to the candidate estimation method are obtained as shown in formula (19):

[0186]

[0187] in, represents the candidate parameter estimate obtained by Bayesian estimation of the equilibrium loss function based on the generalized entropy loss function, ω represents, represents the candidate parameter estimate corresponding to the maximum likelihood estimate, represents the candidate parameter estimates obtained by Bayesian estimation of the generalized entropy loss function.

[0188] Step 902, determining the estimation accuracy of the candidate parameter estimates.

[0189] It should be noted that since the candidate parameter estimation values ​​obtained by different candidate estimation methods have different accuracies in predicting the failure probability of the equipment to be tested, in order to ensure the accuracy of the candidate parameter estimation values ​​in predicting the failure probability of the equipment to be tested, it is necessary to determine the estimation accuracy of the candidate parameter estimation values.

[0190] In one embodiment of the present application, the candidate parameter estimates can be evaluated according to an accuracy evaluation model. When it is necessary to determine the estimated accuracy of the candidate parameter estimates, the following may be specifically included: inputting the candidate parameter estimates into the accuracy evaluation model, performing accuracy evaluation on the candidate parameter estimates through the accuracy evaluation model, and obtaining the accuracy result output by the accuracy evaluation model, which is the estimated accuracy of the candidate parameter estimates.

[0191] Among them, the training process of the accuracy assessment model may specifically include the following contents: obtaining sample parameters and sample accuracy, marking the accuracy of the sample parameters based on the historical experience of the staff, obtaining sample parameters marked with sample accuracy, inputting the sample parameters marked with sample accuracy into the accuracy assessment model, realizing the training of the accuracy assessment model, and obtaining the trained accuracy assessment model.

[0192] To further illustrate, the accuracy assessment model may include but is not limited to: a consistency assessment model and an unbiased assessment model.

[0193] In another embodiment of the present application, when it is necessary to determine the estimated accuracy of the candidate parameter estimates, the candidate parameter estimates can be substituted into a failure probability prediction function, and a failure probability prediction is performed on a reference device based on the failure probability prediction function containing the candidate parameter estimates to obtain the predicted failure probability of the reference device. Based on the actual failure probability of the reference device, the accuracy of the predicted failure probability of the reference device is determined, and this accuracy is the estimated accuracy of the candidate parameter estimates.

[0194] Step 903: Use the candidate parameter estimate with the highest estimation accuracy as the prediction parameter estimate.

[0195] It should be noted that, since the candidate parameter estimation value with the highest estimation accuracy can more accurately predict the failure probability of the device to be tested, the candidate parameter estimation value with the highest estimation accuracy is used as the prediction parameter estimation value.

[0196] In one embodiment of the present application, when it is necessary to determine a prediction parameter estimate, the candidate parameter estimate can be sorted from large to small according to its estimation accuracy. After sorting, the candidate parameter estimate ranked first is the candidate parameter estimate with the highest estimation accuracy. Therefore, the candidate parameter estimate ranked first is used as the prediction parameter estimate.

[0197] In one embodiment of the present application, if it is determined that the candidate parameter estimate obtained by Bayesian estimation based on the balanced loss function of the generalized entropy loss function has the highest accuracy, the candidate parameter estimate is used as the prediction parameter estimate.

[0198] The above-mentioned equipment screening method screens the candidate parameter estimation values ​​by the estimation accuracy of the candidate parameter estimation values, so that the estimation accuracy of the screened candidate parameter estimation values ​​is the highest, ensuring that the predicted parameter estimation values ​​obtained based on the screened candidate parameter estimation values ​​can more accurately predict the failure probability of the equipment to be tested, thereby improving the accuracy of the failure probability prediction.

[0199] In an exemplary embodiment, when it is necessary to obtain the predicted failure probability of the device to be tested, the following process may be specifically included: Fig.10 As shown:

[0200] Step 1001: Obtain an estimated usage period of at least one device to be detected.

[0201] Step 1002: for each device to be detected, determine a reference device corresponding to the device to be detected.

[0202] Step 1003, performing a simulation experiment on the reference device according to a preset ideal device quantity, a minimum device quantity and a preset experiment period.

[0203] Step 1004, determining the failure time of the failed device in the reference device according to the experiment termination condition of the simulation experiment.

[0204] Step 1005: determine the failure sequence number corresponding to the failed device.

[0205] Step 1006: Based on the correspondence between the sequence number and the preset quantity, determine the predicted quantity corresponding to the failure sequence number, and use the predicted quantity corresponding to the failure sequence number as the censored quantity corresponding to the failed device.

[0206] Step 1007: Based on the candidate estimation method, according to the failure time of the failed device and the number of deletions corresponding to the failed device, determine the candidate parameter estimation value corresponding to the candidate estimation method.

[0207] Step 1008, determining the estimation accuracy of the candidate parameter estimation values, and taking the candidate parameter estimation value with the highest estimation accuracy as the prediction parameter estimation value.

[0208] Step 1009: Determine the failure probability prediction function corresponding to each device to be tested according to the estimated value of the prediction parameter.

[0209] Step 1010: Determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected.

[0210] Step 1011 , the device to be detected with the smallest predicted failure probability among all the devices to be detected is taken as the target device.

[0211] The above-mentioned equipment screening method first determines the failure probability prediction function corresponding to each device to be detected, and then determines the predicted failure probability of each device to be detected within the expected use cycle through the failure probability prediction function corresponding to each device to be detected, and takes the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device. According to the above content, the present application can predict the possibility of equipment failure of the device to be detected within the expected use cycle according to the predicted failure probability of the device, and then, by taking the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device, it prevents the device from failing soon after it is put into use, thereby reducing the cost investment of the enterprise for repairing and replacing failed equipment, and each failure probability prediction function in the present application is determined according to the estimated value of the prediction parameter of each device to be detected, therefore, the predicted failure probability of each device to be detected within the expected use cycle determined through each failure probability prediction function can accurately reflect the actual situation of each device to be detected, and ensure the accuracy of determining the predicted failure probability of each device to be detected.

[0212] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0213] Based on the same inventive concept, the embodiment of the present application also provides a device screening device for implementing the device screening method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more device screening device embodiments provided below can refer to the limitations on the device screening method above, and will not be repeated here.

[0214] In one embodiment, Fig.11 As shown, a device screening device is provided, comprising: a cycle acquisition module 10, a failure prediction module 20 and a device screening module 30, wherein:

[0215] The cycle acquisition module 10 is used to acquire the estimated use cycle of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0216] The failure prediction module 20 is used to determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected;

[0217] The device screening module 30 is used to select the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device.

[0218] The above-mentioned equipment screening device first determines the failure probability prediction function corresponding to each device to be detected, and then determines the predicted failure probability of each device to be detected within the expected use cycle through the failure probability prediction function corresponding to each device to be detected, and takes the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device. According to the above content, the present application can predict the possibility of equipment failure of the device to be detected within the expected use cycle according to the predicted failure probability of the device, and then, by taking the device to be detected with the smallest predicted failure probability among the devices to be detected as the target device, it prevents the device from failing soon after it is put into use, thereby reducing the cost investment of the enterprise for repairing and replacing failed equipment, and each failure probability prediction function in the present application is determined according to the estimated value of the prediction parameter of each device to be detected, therefore, the predicted failure probability of each device to be detected within the expected use cycle determined through each failure probability prediction function can accurately reflect the actual situation of each device to be detected, and ensure the accuracy of determining the predicted failure probability of each device to be detected.

[0219] In one embodiment, Fig.12 As shown, a device screening apparatus is provided, in which a failure prediction module 20 includes: a first determination unit 21, a second determination unit 22 and a third determination unit 23, wherein:

[0220] The first determining unit 21 is configured to determine, for each device to be detected, a reference device corresponding to the device to be detected.

[0221] The second determining unit 22 is used to determine the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the number of deletions corresponding to the failed device.

[0222] The third determination unit 23 is used to determine the failure probability prediction function corresponding to each device to be detected according to the estimated value of the prediction parameter.

[0223] In one embodiment, Fig.13 As shown, a device screening apparatus is provided, in which the second determination unit 22 includes: a simulation experiment subunit 221 and a first determination subunit 222, wherein:

[0224] The simulation experiment subunit 221 is used to perform a simulation experiment on the reference device according to a preset ideal device quantity, a minimum device quantity and a preset experiment period.

[0225] The first determining subunit 222 is used to determine the failure time of the failed device in the reference device and the deletion quantity corresponding to the failed device if the simulation experiment meets the experiment termination condition.

[0226] The first determination subunit 222 is specifically used for the experiment termination conditions including that the actual experiment cycle of the reference device is less than the preset experiment cycle, and the failure number of failed devices in the reference device is greater than or equal to the ideal device number; or, the actual experiment cycle of the reference device is greater than or equal to the preset experiment cycle, and the failure number of failed devices in the reference device is greater than or equal to the minimum device number; determining the failure sequence number corresponding to the failed device; determining the predicted number corresponding to the failure sequence number based on the corresponding relationship between the sequence number and the preset number; and using the predicted number corresponding to the failure sequence number as the deletion number corresponding to the failed device.

[0227] In one embodiment, Fig.14 As shown, a device screening device is provided, in which the second determination unit 22 further includes: a second determination subunit 223, a third determination subunit 224 and a fourth sub-determination unit 225, wherein:

[0228] The second determining subunit 223 is used to determine, based on the candidate estimation method, the candidate parameter estimation value corresponding to the candidate estimation method according to the failure time of the failed device and the number of deletions corresponding to the failed device.

[0229] The second determining subunit 223 is specifically used for candidate estimation methods including at least one of a maximum likelihood estimation method and a Bayesian estimation method.

[0230] The third determining subunit 224 is used to determine the estimation accuracy of the candidate parameter estimation value.

[0231] The fourth determining subunit 225 is configured to use the candidate parameter estimation value with the highest estimation accuracy as the prediction parameter estimation value.

[0232] Each module in the above-mentioned device screening apparatus can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0233] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Fig.15As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a device screening method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0234] Those skilled in the art will understand that Fig.15 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0235] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0236] Obtaining an estimated service life of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0237] Determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected;

[0238] The device to be tested with the smallest predicted failure probability among all the devices to be tested is taken as the target device.

[0239] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0240] For each device to be detected, determine a reference device corresponding to the device to be detected, wherein the device model of the reference device is the same as the device model of the device to be detected;

[0241] Determine the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the censoring quantity corresponding to the failed device; wherein the censoring quantity is used to represent the number of devices that are censored from the reference device when the failed device fails;

[0242] According to the estimated values ​​of the prediction parameters, the failure probability prediction function corresponding to each device to be tested is determined.

[0243] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0244] Conduct simulation experiments on reference equipment according to the preset ideal number of equipment, minimum number of equipment and preset experimental period;

[0245] If the simulation experiment meets the experimental termination conditions, the failure time of the failed equipment in the reference equipment is determined, and the number of deletions corresponding to the failed equipment is determined.

[0246] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0247] The actual experimental cycle of the reference device is less than the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the ideal device number; or, the actual experimental cycle of the reference device is greater than or equal to the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the minimum device number.

[0248] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0249] Determine the failure sequence number corresponding to the failed device;

[0250] Determine the predicted quantity corresponding to the failure sequence number based on the corresponding relationship between the sequence number and the preset quantity;

[0251] The predicted quantity corresponding to the failure sequence number is used as the deleted quantity corresponding to the failed equipment.

[0252] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0253] Based on the candidate estimation method, determining the candidate parameter estimation value corresponding to the candidate estimation method according to the failure time of the failed device and the number of deletions corresponding to the failed device;

[0254] determining the estimated accuracy of candidate parameter estimates;

[0255] The candidate parameter estimate with the highest estimation accuracy is used as the predicted parameter estimate.

[0256] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0257] The candidate estimation methods include at least one of a maximum likelihood estimation method and a Bayesian estimation method.

[0258] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0259] Obtaining an estimated service life of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0260] Determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected;

[0261] The device to be tested with the smallest predicted failure probability among all the devices to be tested is taken as the target device.

[0262] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0263] For each device to be detected, determine a reference device corresponding to the device to be detected, wherein the device model of the reference device is the same as the device model of the device to be detected;

[0264] Determine the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the censoring quantity corresponding to the failed device; wherein the censoring quantity is used to represent the number of devices that are censored from the reference device when the failed device fails;

[0265] According to the estimated values ​​of the prediction parameters, the failure probability prediction function corresponding to each device to be tested is determined.

[0266] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0267] Conduct simulation experiments on reference equipment according to the preset ideal number of equipment, minimum number of equipment and preset experimental period;

[0268] If the simulation experiment meets the experimental termination conditions, the failure time of the failed equipment in the reference equipment is determined, and the number of deletions corresponding to the failed equipment is determined.

[0269] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0270] The actual experimental cycle of the reference device is less than the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the ideal device number; or, the actual experimental cycle of the reference device is greater than or equal to the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the minimum device number.

[0271] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0272] Determine the failure sequence number corresponding to the failed device;

[0273] Determine the predicted quantity corresponding to the failure sequence number based on the corresponding relationship between the sequence number and the preset quantity;

[0274] The predicted quantity corresponding to the failure sequence number is used as the deleted quantity corresponding to the failed equipment.

[0275] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0276] Based on the candidate estimation method, determining the candidate parameter estimation value corresponding to the candidate estimation method according to the failure time of the failed device and the number of deletions corresponding to the failed device;

[0277] determining the estimated accuracy of candidate parameter estimates;

[0278] The candidate parameter estimate with the highest estimation accuracy is used as the predicted parameter estimate.

[0279] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0280] The candidate estimation methods include at least one of a maximum likelihood estimation method and a Bayesian estimation method.

[0281] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0282] Obtaining an estimated service life of at least one device to be detected, wherein the device models of the devices to be detected are different;

[0283] Determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected service life according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected;

[0284] The device to be tested with the smallest predicted failure probability among all the devices to be tested is taken as the target device.

[0285] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0286] For each device to be detected, determine a reference device corresponding to the device to be detected, wherein the device model of the reference device is the same as the device model of the device to be detected;

[0287] Determine the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the censoring quantity corresponding to the failed device; wherein the censoring quantity is used to represent the number of devices that are censored from the reference device when the failed device fails;

[0288] According to the estimated values ​​of the prediction parameters, the failure probability prediction function corresponding to each device to be tested is determined.

[0289] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0290] Conduct simulation experiments on reference equipment according to the preset ideal number of equipment, minimum number of equipment and preset experimental period;

[0291] If the simulation experiment meets the experimental termination conditions, the failure time of the failed equipment in the reference equipment is determined, and the number of deletions corresponding to the failed equipment is determined.

[0292] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0293] The actual experimental cycle of the reference device is less than the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the ideal device number; or, the actual experimental cycle of the reference device is greater than or equal to the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the minimum device number.

[0294] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0295] Determine the failure sequence number corresponding to the failed device;

[0296] Determine the predicted quantity corresponding to the failure sequence number based on the corresponding relationship between the sequence number and the preset quantity;

[0297] The predicted quantity corresponding to the failure sequence number is used as the deleted quantity corresponding to the failed equipment.

[0298] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0299] Based on the candidate estimation method, determining the candidate parameter estimation value corresponding to the candidate estimation method according to the failure time of the failed device and the number of deletions corresponding to the failed device;

[0300] determining the estimated accuracy of candidate parameter estimates;

[0301] The candidate parameter estimate with the highest estimation accuracy is used as the predicted parameter estimate.

[0302] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0303] The candidate estimation methods include at least one of a maximum likelihood estimation method and a Bayesian estimation method.

[0304] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0305] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, data block or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The data blocks involved in the embodiments provided in this application may include at least one of relational data blocks and non-relational data blocks. Non-relational data blocks may include distributed data blocks based on blockchain, etc., but are not limited to this. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, data processing logic devices based on quantum computing, etc., but are not limited to this.

[0306] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0307] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A device screening method, characterized in that: The method comprises: Obtaining an estimated service life of at least one device to be detected, wherein the device models of the devices to be detected are different; Determine a failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected use period according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected; The device to be tested with the smallest predicted failure probability among all the devices to be tested is taken as the target device.

2. The method according to claim 1, characterized in that The step of determining the failure probability prediction function corresponding to each device to be detected includes: For each device to be detected, determine a reference device corresponding to the device to be detected, wherein the device model of the reference device is the same as the device model of the device to be detected; Determine the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the censoring number corresponding to the failed device; wherein the censoring number is used to represent the number of devices that are censored from the reference device when the failed device fails; According to the estimated values ​​of the prediction parameters, a failure probability prediction function corresponding to each device to be detected is determined.

3. The method according to claim 2, characterized in that The failure time and the number of deletions are determined as follows: Performing simulation experiments on the reference equipment according to a preset ideal number of equipment, a minimum number of equipment, and a preset experimental period; If the simulation experiment meets the experiment termination condition, the failure time of the failed device in the reference device is determined, and the deletion quantity corresponding to the failed device is determined.

4. The method according to claim 3, characterized in that The experiment termination conditions include: The actual experimental cycle of the reference device is less than the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the ideal device number; or, the actual experimental cycle of the reference device is greater than or equal to the preset experimental cycle, and the failure number of failed devices in the reference device is greater than or equal to the minimum device number.

5. The method according to claim 3, characterized in that: The determining the number of deletions corresponding to the failed device includes: Determine the failure sequence number corresponding to the failed device; Determine the predicted number corresponding to the failure sequence number based on the corresponding relationship between the sequence number and the preset number; The predicted number corresponding to the failure sequence number is used as the deleted number corresponding to the failed device.

6. The method according to claim 2, characterized in that The step of determining the estimated value of the prediction parameter according to the failure time of the failed device in the reference device and the number of deletions corresponding to the failed device includes: Based on the candidate estimation method, determining the candidate parameter estimation value corresponding to the candidate estimation method according to the failure time of the failed device and the number of deletions corresponding to the failed device; determining the estimated accuracy of candidate parameter estimates; The candidate parameter estimate with the highest estimation accuracy is used as the prediction parameter estimate.

7. The method according to claim 6, characterized in that The candidate estimation methods include at least one of a maximum likelihood estimation method and a Bayesian estimation method.

8. An equipment screening device, characterized in that: The device comprises: A cycle acquisition module, used to acquire an estimated use cycle of at least one device to be detected, wherein the device models of the devices to be detected are different devices; A failure prediction module is used to determine the failure probability prediction function corresponding to each device to be detected, and determine the predicted failure probability of each device to be detected within the expected use period according to the failure probability prediction function corresponding to each device to be detected; wherein each failure probability prediction function is determined according to the estimated value of the prediction parameter of each device to be detected; The device screening module is used to select the device to be detected with the smallest predicted failure probability among all the devices to be detected as the target device.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images