Drop height determination method and device, electronic equipment and storage medium

Abstract

The invention relates to a drop height determination method and device, electronic equipment and a storage medium, and relates to the technical field of screen testing, and the method comprises the steps: obtaining drop height data corresponding to a plurality of pieces of first sample equipment; according to the drop height data, performing distribution fitting on the target distribution model to obtain a first distribution result corresponding to the target drop height; determining a plurality of sample height data according to the first distribution result; and according to the multiple pieces of sample height data, performing distribution fitting on the target distribution model to obtain a target distribution result corresponding to the target drop height. Thus, it can be guaranteed that the data finally used for distribution fitting does not have the deleted data under the condition that the number of samples is not additionally increased, the accuracy of distribution fitting of the target distribution model is improved, and then the reliability and accuracy of a target distribution result are guaranteed.

Description

Technical Field

[0001] The present disclosure relates to the technical field of screen testing, and in particular to a method, device, electronic device and storage medium for determining a drop height. Background Art

[0002] For electronic devices such as mobile phones and tablets, drop tests are often carried out before they are put on the market to determine the drop height limit at which the screen will not break when the electronic device is dropped. Usually, each sample mobile phone is dropped from a starting height. If the screen does not break, the test is repeated at a certain height. Based on the recorded experimental height, the corresponding drop height limit of the electronic device is estimated. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the present disclosure provides a method, device, electronic device and storage medium for determining a drop height.

[0004] According to a first aspect of an embodiment of the present disclosure, a method for determining a drop height is provided, the method comprising: acquiring drop height data corresponding to a plurality of first sample devices, the drop height data representing the drop height at which a screen of the first sample device is shattered when it is dropped; performing distribution fitting on a target distribution model based on the drop height data to obtain a first distribution result corresponding to a target drop height, the target drop height representing a drop height limit at which a screen of the sample device will not be shattered when it is dropped; determining a plurality of sample height data based on the first distribution result, the sample height data being within a height distribution interval corresponding to the first distribution result; performing distribution fitting on the target distribution model based on the plurality of sample height data to obtain a target distribution result corresponding to the target drop height.

[0005] Optionally, performing distribution fitting on the target distribution model according to the drop height data to obtain a first distribution result corresponding to the target drop height includes: performing distribution fitting on the target distribution model according to the drop height data to obtain a first distribution parameter of the target distribution model; and determining the first distribution result according to the first distribution parameter.

[0006] Optionally, the drop height data includes censored data and / or non-censored data, the censored data indicating that the drop height at which the screen of the first sample device is shattered when it falls is outside a test height interval corresponding to the test device, and the non-censored data indicating that the drop height at which the screen of the first sample device is shattered when it falls is within the test height interval, and the test device is used to test the drop height at which the screen of the first sample device is shattered when it falls within the test height interval; performing distribution fitting on the target distribution model according to the drop height data to obtain the first distribution parameter of the target distribution model includes: performing distribution fitting on the target distribution model according to the censored data and / or the non-censored data to obtain the first distribution parameter.

[0007] Optionally, performing distribution fitting on the target distribution model according to the missing data and / or the non-missing data to obtain the first distribution parameters includes: determining the maximum likelihood function of the target distribution model according to the missing data and / or the non-missing data; when the function value of the maximum likelihood function is maximum, using the model parameter value corresponding to the target distribution model as the first distribution parameter.

[0008] Optionally, determining a plurality of sample height data according to the first distribution result includes: determining a target distribution interval from a height distribution interval corresponding to the first distribution result; and acquiring the plurality of sample height data according to the target distribution interval.

[0009] Optionally, performing distribution fitting on the target distribution model according to the multiple sample height data to obtain a target distribution result corresponding to the target drop height includes: obtaining a target safe drop number corresponding to multiple second sample devices based on the multiple sample height data; the target safe drop number is the maximum number of drops in which the screen of the second sample device does not shatter when the screen of the second sample device shatters after the second sample device falls from the height corresponding to the sample height data one or more times; performing distribution fitting on the target distribution model according to the target safe drop number to obtain a target distribution result corresponding to the target drop height.

[0010] Optionally, performing distribution fitting on the target distribution model according to the target safe drop number to obtain the target distribution result corresponding to the target drop height includes: determining the target fragmentation probability corresponding to the sample height data according to the target safe drop number corresponding to each sample height data; performing distribution fitting on the target distribution model according to the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain the target distribution result corresponding to the target drop height.

[0011] Optionally, performing distribution fitting on the target distribution model according to the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain a target distribution result corresponding to the target drop height includes: performing distribution fitting on the target distribution model according to the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain a second distribution parameter of the target distribution model; and determining the target distribution result according to the second distribution parameter.

[0012] Optionally, the target distribution model includes a Weibull distribution model.

[0013] According to a second aspect of an embodiment of the present disclosure, a device for determining a drop height is provided, the device comprising: an acquisition module, configured to acquire drop height data corresponding to a plurality of first sample devices, the drop height data representing the drop height at which a screen of the first sample device is shattered when it is dropped; a first fitting module, configured to perform distribution fitting on a target distribution model according to the drop height data, so as to obtain a first distribution result corresponding to a target drop height, the target drop height representing a drop height limit value at which a screen of the sample device will not be shattered when it is dropped; a determination module, configured to determine a plurality of sample height data according to the first distribution result, the sample height data being within a height distribution interval corresponding to the first distribution result; and a second fitting module, configured to perform distribution fitting on the target distribution model according to the plurality of sample height data, so as to obtain a target distribution result corresponding to the target drop height.

[0014] Optionally, the first fitting module is configured to perform distribution fitting on the target distribution model according to the drop height data to obtain first distribution parameters of the target distribution model; and determine the first distribution result according to the first distribution parameters.

[0015] Optionally, the drop height data includes censored data and / or non-censored data, the censored data indicating that the drop height at which the screen of the first sample device is shattered when it falls is outside a test height interval corresponding to the test device, and the non-censored data indicating that the drop height at which the screen of the first sample device is shattered when it falls is within the test height interval, and the test device is used to test the drop height at which the screen of the first sample device is shattered when it falls within the test height interval; the first fitting module is configured to perform distribution fitting on the target distribution model according to the censored data and / or the non-censored data to obtain the first distribution parameters.

[0016] Optionally, the first fitting module is configured to determine the maximum likelihood function of the target distribution model based on the missing data and / or the non-missing data; when the function value of the maximum likelihood function is maximum, the model parameter value corresponding to the target distribution model is used as the first distribution parameter.

[0017] Optionally, the determination module is configured to determine a target distribution interval from the height distribution interval corresponding to the first distribution result; and obtain the plurality of sample height data according to the target distribution interval.

[0018] Optionally, the second fitting module is configured to obtain a target safe number of drops corresponding to multiple second sample devices based on the multiple sample height data; the target safe number of drops is the maximum number of drops at which the screen of the second sample device does not shatter when the screen of the second sample device shatters after the second sample device falls from the height corresponding to the sample height data one or more times; and distribution fitting is performed on the target distribution model according to the target safe number of drops to obtain a target distribution result corresponding to the target drop height.

[0019] Optionally, the second fitting module is configured to determine the target fragmentation probability corresponding to the sample height data based on the target safe drop number corresponding to each sample height data; and perform distribution fitting on the target distribution model based on the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain the target distribution result corresponding to the target drop height.

[0020] Optionally, the second fitting module is configured to perform distribution fitting on the target distribution model according to the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain second distribution parameters of the target distribution model; and determine the target distribution result according to the second distribution parameters.

[0021] Optionally, the target distribution model includes a Weibull distribution model.

[0022] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; a memory for storing processor executable instructions; wherein the processor is configured to implement the steps of the method for determining the drop height provided in the first aspect of the present disclosure when calling the executable instructions stored on the memory.

[0023] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the method for determining the drop height provided in the first aspect of the present disclosure are implemented.

[0024] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0025] First, the drop height data corresponding to the multiple first sample devices are obtained. Secondly, according to the drop height data, the target distribution model is distributed fitted to obtain the first distribution result corresponding to the target drop height. After that, according to the first distribution result, multiple sample height data are determined. Finally, according to the multiple sample height data, the target distribution model is distributed fitted to obtain the target distribution result corresponding to the target drop height. Through the above method, the target distribution model is first distributed fitted according to the drop height data to obtain the first distribution result corresponding to the target drop height. Since there may be missing data in the drop height data, after obtaining the first distribution result, multiple sample height data located in the height distribution interval corresponding to the first distribution result can be further screened out according to the first distribution result. In this way, it can be ensured that the acquired sample height data are all non-missing data, and then according to the obtained multiple sample height data, the target distribution model is distributed fitted for the second time to obtain the target distribution result corresponding to the target drop height. By obtaining multiple sample height data through the above regression method, it is possible to ensure that there is no missing data in the data finally used for distribution fitting without increasing the number of samples, thereby improving the accuracy of distribution fitting of the target distribution model, and thus ensuring the reliability and accuracy of the target distribution results.

[0026] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0028] Figure 1 The figure is a flow chart of a method for determining a drop height according to an exemplary embodiment.

[0029] Figure 2 The figure is a flow chart of another method for determining a drop height according to an exemplary embodiment.

[0030] Figure 3 The figure is a flow chart of another method for determining a drop height according to an exemplary embodiment.

[0031] Figure 4 The figure is a block diagram of a device for determining a falling height according to an exemplary embodiment.

[0032] Figure 5 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0033] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0034] The terms "first", "second", etc. in the specification and claims of this application and the above drawings are used to distinguish similar objects and do not have to be understood as a specific order or sequence. In addition, in the description with reference to the drawings, the same symbols in different drawings represent the same elements.

[0035] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0036] In the description of the present disclosure, unless otherwise specified, "multiple" means two or more than two, and other quantifiers are similar thereto; "at least one item", "one item or multiple items" or similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one item a can represent any number of a; for another example, one item or multiple items among a, b and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple; "and / or" is a kind of description of the association relationship of associated objects, indicating that there can be three kinds of relationships, for example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " indicates that the associated objects before and after are in an "or" relationship.

[0037] Before introducing the method, device, electronic device and storage medium for determining the drop height provided by the present disclosure, the application scenarios involved in each embodiment of the present disclosure are first introduced. The reliability analysis of a product is generally used to test the performance of a product before it goes on the market or to analyze the actual performance of a product after it goes on the market. Depending on whether there is experimental personnel involved, the research method is generally divided into experimental research (with experimental personnel involved, artificially controlling variables) and observational research (without experimental personnel involved, only obtaining historical observation data). Due to the lack of historical observation data, experimental research is generally adopted for product performance testing before it goes on the market. The general method is to select a certain number of samples for testing and estimate the relevant indicators of the product based on the obtained test data.

[0038] When conducting reliability analysis on a product, a related technology is based on a horizontal error function, a method for estimating the parameters of the Weibull unit life model, but this method relies on the integrity of the data (that is, it must be complete data, and complete data refers to data obtained after the target event is clearly observed in observation or experiment). However, in the test scenario of testing the drop height limit of the electronic device screen (it is generally believed that if the drop height exceeds the drop height limit, the screen will be broken, and if the drop height does not exceed the drop height limit, the screen will be intact and will not be broken), due to the limitations of the test environment or test equipment (for example, the test height corresponding to the test equipment is limited, resulting in a limited range of testable heights, and the drop height limit of some sample equipment may exceed the range of testable heights of the equipment), the collected data may be censored. In this case, this method will not be able to handle censored data. Among them, censored data refers to data obtained after stopping observation because the target event cannot be observed during observation or experiment due to human or other reasons.

[0039] In another related technology, each sample phone is dropped from a starting height. If the screen does not break, the test is repeated at a higher height. The distribution model is directly fitted based on the recorded experimental height to obtain the model parameters. Although this method can handle missing data, when the number of samples is small, the estimated results may have large deviations. If there is missing data, the deviation of the results will be further increased.

[0040] In order to solve the above technical problems, the present invention provides a method, device, electronic device and storage medium for determining a drop height. First, a target distribution model is subjected to a first distribution fitting according to the drop height data to obtain a first distribution result corresponding to the target drop height. Since there may be missing data in the drop height data, after obtaining the first distribution result, multiple sample height data located in the height distribution interval corresponding to the first distribution result can be further screened out according to the first distribution result. In this way, it can be ensured that the acquired sample height data are all non-missing data, and then, according to the obtained multiple sample height data, a second distribution fitting is performed on the target distribution model to obtain a target distribution result corresponding to the target drop height. By obtaining multiple sample height data by the above regression method, it can be ensured that there is no missing data in the data finally used for distribution fitting without increasing the number of samples, thereby improving the accuracy of distribution fitting of the target distribution model, and thus ensuring the reliability and accuracy of the target distribution result.

[0041] Before describing the specific implementation of the present invention in detail, the design principle of the present invention is briefly described. In this embodiment, the drop height limit of the electronic device screen meets the following conditions:

[0042] (1) When the same electronic device is dropped repeatedly, the drop height limit values ​​corresponding to each drop are independent and follow the same distribution.

[0043] (2) The drop height limits corresponding to a single drop of different electronic devices are independent and follow the same distribution.

[0044] (3) Weibull distribution assumption: For any electronic device, the drop height limit value H obeys the Weibull distribution.

[0045] Based on the above situation, the following model can be defined for the drop height limit value H:

[0046] H=M+Σ

[0047] The random variable M is used to describe the randomness caused by the differences between different electronic devices, such as the quality of screen glass workmanship and quality control. The random variable Σ is used to describe the randomness caused by the differences between each drop of the same electronic device, such as the drop angle and force position.

[0048] Compared with estimating the two random variables separately, in order to improve computational efficiency, the two can be combined and the Weibull model can be used to estimate the drop height limit H. Optionally, the following two methods can be used:

[0049] (1) Let H=M': When the randomness of the drop height limit value comes entirely from the individual differences of electronic devices, it is determined that for each electronic device, the corresponding drop height limit value is fixed.

[0050] (2) Let H = Σ': When the randomness of the drop height limit value comes entirely from the difference between drops, it is determined that for two different electronic devices of the same model, their drop height limit values ​​are independent and follow the same distribution.

[0051] In this embodiment, the corresponding first distribution result can be obtained by the above method (1), and then based on the first distribution result, the corresponding target distribution result can be further obtained by the above method (2) to obtain the distribution of the drop height limit value.

[0052] Generally speaking, when an electronic device falls from a certain height, it is generally believed that there is a drop height limit value H. If the drop height exceeds H, the screen of the electronic device will be broken. If the drop height does not exceed H, the screen of the electronic device will be intact and will not be broken. The drop height limit value H can be used to evaluate the reliability of a mobile phone screen, and the drop height limit value has a certain randomness, which mainly comes from the differences between different individual devices and the differences between each drop. Therefore, in this embodiment, determining the drop height limit value is actually equivalent to determining the distribution of the random variable H.

[0053] The specific implementation modes of the present invention are described in detail below with reference to the accompanying drawings.

[0054] Figure 1 is a flow chart of a method for determining a drop height according to an exemplary embodiment. Figure 1 As shown, the method may include the following steps.

[0055] In step S101, drop height data corresponding to a plurality of first sample devices is obtained.

[0056] The drop height data represents the drop height at which the screen of the corresponding first sample device is broken when it is dropped. The first sample device may be part of all sample devices (electronic devices used to determine the target distribution result, the sample device has a screen and can be used to perform a drop experiment), and the specific number of the first sample devices may be selected according to actual needs.

[0057] For example, taking the first sample device including m units as an example, for each first sample device, a drop test can be performed starting from an initial height h0. The initial height can be set to a smaller value manually, but due to the limitation of the test device, the initial height also needs to be within the test height range allowed by the test device. If the screen of the first sample device is not broken after falling, then increase the height by Δh (preset value) and drop again, and so on, until the screen of the first sample device is broken or the test height reaches the maximum value h of the test height range allowed by the test device. max . Record the observed drop height values ​​when the screen of each first sample device breaks, such as h1, h2…h m , and use the observed drop height value as the drop height data.

[0058] It should be noted that in some cases, the target drop height may be less than the initial height h0, that is, the screen of the first sample device is broken when it is dropped for the first time. In this case, the drop height data of the first sample device can be recorded as the initial height h0. In other cases, the target drop height may be greater than the maximum value h0 of the test height range allowed by the test device. max , that is, the maximum value h in the test height interval max When dropped, the screen of the first sample device did not break. At this time, the drop height data of the first sample device can be recorded as the maximum value h of the test height interval. max The target drop height represents the drop height limit value at which the screen of the sample device will not break when it falls. Through the above steps, the drop height data corresponding to each first sample device can be obtained.

[0059] In step S102, distribution fitting is performed on a target distribution model according to the drop height data to obtain a first distribution result corresponding to the target drop height.

[0060] Wherein, the target distribution model may include a Weibull distribution model. In this step, the target distribution model may be distribution fitted according to the drop height data corresponding to each first sample device, and then the first distribution parameter of the target distribution model may be obtained. Furthermore, the first distribution result may be obtained according to the first distribution parameter. Wherein, the first distribution result is used to characterize the distribution of the target drop height obtained according to the drop height data. The first distribution result may be, for example, a distribution probability graph obtained according to the first distribution parameter, for example, in which the horizontal axis represents the value range of the drop height corresponding to the random variable H (that is, the value range of the target drop height), and the vertical axis represents the function value of the distribution function corresponding to each drop height.

[0061] It can be understood that in the process of determining the first distribution result, it is equivalent to determining that the drop height limit value corresponding to each first sample device is fixed, that is, the randomness of the drop height limit value (i.e., the target drop height) comes from the individual differences between different first sample devices. Based on this setting, for each first sample device, only one experiment is required at the same drop height. If the screen is broken, the corresponding drop height data is obtained. If the screen is intact, continue to experiment with the next drop height until the drop height data is obtained. Then, based on the obtained drop height data, the first distribution result is determined. It can be seen that the first distribution result is obtained based on the case where the randomness of determining the drop height limit value comes from the individual differences of the devices.

[0062] In step S103, a plurality of sample height data are determined according to the first distribution result.

[0063] Among them, the sample height data is located in the height distribution interval corresponding to the first distribution result, and the height distribution interval is also the value range corresponding to the random variable H in the first distribution result.

[0064] According to the above description, the first distribution result is obtained based on the drop height data corresponding to each first sample device. In actual scenarios, there may be missing data in the drop height data, and due to the small number of samples, the first distribution result will have a certain degree of deviation, which cannot accurately reflect the distribution of the target drop height. Therefore, in this embodiment, based on the first distribution result, the target distribution interval can be determined from the height distribution interval corresponding to the first distribution result. Then, according to the target distribution interval, the multiple sample height data are obtained. Among them, the number of sample height data obtained can be set according to actual needs. Furthermore, for each sample height data, a certain number of second sample devices are allocated for subsequent experiments. Among them, the second sample device can also include sample devices in the first sample device that did not break after the drop test.

[0065] In some embodiments, the target distribution interval can be determined according to the height distribution interval corresponding to the first distribution result and the test height interval corresponding to the test device. For example, if the height distribution interval corresponding to the first distribution result is S1 and the test height interval corresponding to the test device is S2, then the target distribution interval can be the intersection of S1 and S2.

[0066] Furthermore, when obtaining sample height data from the target distribution interval, it can be selected in an arithmetic progression, or it can be obtained by manual or random collection. When collecting manually or randomly, the difference between the selected maximum and minimum values ​​should be avoided. If the difference between the maximum and minimum values ​​of the sample height data is too small, the variance of the regression slope will be too large, and a negative slope may occur, which is contrary to the actual scenario. It should also be avoided as much as possible that the selected sample height data falls near the lowest or highest value of the first distribution result to ensure that the height corresponding to the obtained sample height data is able to cause the second sample device to break.

[0067] For example, in order to avoid selecting the maximum and minimum values ​​in the height distribution interval, the interval corresponding to the 20% to 80% quantiles can be taken from the height distribution interval corresponding to the first distribution result as the interval corresponding to S1. Assuming that the test height interval corresponding to the test equipment is S2, including the interval corresponding to the 5% to 50% quantiles, 4 sample height data need to be obtained. First, the target distribution interval can be obtained as the intersection of S1 and S2, that is, including 20% ​​to 50%. Then, 4 quantiles can be evenly collected from 20% to 50%, and then the sample height data can be obtained. That is, the sample height data includes the height values ​​corresponding to the 20%, 30%, 40% and 50% quantiles respectively.

[0068] In step S104, distribution fitting is performed on the target distribution model according to the plurality of sample height data to obtain a target distribution result corresponding to the target drop height.

[0069] For example, the target distribution model can be fitted according to a plurality of sample height data to obtain a second distribution parameter of the target distribution model. Then, the target distribution result can be determined according to the second distribution parameter to obtain the distribution of the target drop height. The target distribution result is used to characterize the distribution of the target drop height obtained according to the sample height data. The target distribution result can be, for example, a distribution probability graph obtained according to the second distribution parameter.

[0070] It is understandable that in the process of determining the target distribution result, the randomness equivalent to determining the drop height limit value (i.e., the target drop height) comes from the difference caused by the drop. Based on this setting, for each sample height data, multiple second sample devices can be allocated to determine the drop results between different drop times under the drop height corresponding to each sample height data, the drop result is whether the screen of the second sample device is broken, and the screen is broken after several drops. Then, according to the drop result, the target distribution result is determined. Since the sample height data is obtained based on the first distribution result, the target distribution result is obtained based on the sample height data. In other words, the target distribution result finally obtained combines the above two settings, that is, the randomness of the target drop height caused by the individual differences of the sample devices is taken into account, and the randomness of the target drop height caused by the differences between the drops is taken into account. In this way, the distribution of the target drop height is determined by the regression method, the integrity (i.e., there will be no missing data) and accuracy of the sample height data are guaranteed, the dependence on the corresponding sample number is weakened, and the reliability and accuracy of the distribution estimation of the target drop height can be effectively improved without increasing the number of samples.

[0071] Using the above method, the target distribution model is firstly subjected to the first distribution fitting according to the drop height data to obtain the first distribution result corresponding to the target drop height. Since there may be missing data in the drop height data, after obtaining the first distribution result, multiple sample height data located in the height distribution interval corresponding to the first distribution result can be further screened out according to the first distribution result. In this way, it can be ensured that the acquired sample height data are all non-missing data, and then the target distribution model is subjected to the second distribution fitting according to the obtained multiple sample height data to obtain the target distribution result corresponding to the drop height of the target height. By obtaining multiple sample height data by the above regression method, it can be ensured that there is no missing data in the data finally used for distribution fitting without increasing the number of samples, thereby improving the accuracy of distribution fitting of the target distribution model, and thus ensuring the reliability and accuracy of the target distribution result.

[0072] The above step S102 is described in detail below. Figure 2 As shown, in step S102, performing distribution fitting on the target distribution model according to the drop height data to obtain a first distribution result corresponding to the target drop height may include the following steps:

[0073] In step S1021, distribution fitting is performed on the target distribution model according to the drop height data to obtain a first distribution parameter of the target distribution model.

[0074] Among them, the drop height data includes missing data and / or non-missing data, the missing data indicates that the drop height at which the screen of the first sample device is shattered when it falls is outside the test height interval corresponding to the test device, and the non-missing data indicates that the drop height at which the screen of the first sample device is shattered when it falls is within the test height interval, and the test device is used to test the drop height at which the screen of the first sample device is shattered when it falls within the test height interval.

[0075] It is understandable that in the actual test experiment, since the test height range of the test equipment is limited, and the drop height limit value of the first sample device may exceed the test height range of the test equipment, missing data will be generated. Specifically, there may be the following two situations: one is that the drop height limit value of the first sample device is less than the initial height (that is, the screen is shattered when it is dropped for the first time), and the other is that the drop height limit value of the first sample device is greater than the maximum value of the test height range allowed by the test equipment (that is, the screen of the first sample device is intact and not shattered in the end). According to the above two situations, the missing data sets C1 and C2 can be defined:

[0076] If i∈C1, the maximum drop height value of the first sample device of the i-th device is less than hi (that is, the screen is broken when it is dropped for the first time, and hi is recorded as the initial height h0);

[0077] If i∈C2, the maximum drop height of the first sample device of the i-th device is greater than hi (that is, the screen is intact and not broken in the end, hi is the maximum value h of the test height range allowed by the test device). max ).

[0078] In this way, all the drop height data can be obtained, and the drop height data includes censored data and / or non-censored data. Afterwards, the target distribution model can be fitted according to the censored data and / or the non-censored data to obtain the first distribution parameter.

[0079] In one embodiment, the maximum likelihood function of the target distribution model may be determined based on the censored data and / or the non-censored data, and when the function value of the maximum likelihood function is the largest, the model parameters corresponding to the target distribution model are used as the first distribution parameters. Wherein, when the target distribution model includes a Weibull distribution model, the distribution parameters (i.e., model parameters) corresponding to the Weibull distribution model may include k and λ, where k is a shape parameter and λ is a scale parameter.

[0080] For example, the maximum likelihood function of the target distribution model can be determined by the following formula:

[0081]

[0082] Among them, l(k,λ|h0,h1…h m ) is the maximum likelihood function expression, k is the shape parameter (its value determines the basic shape of the curve), λ is the scale parameter (its value determines the size of the curve), h i is the drop height data, i is a different first sample device, C1 is a drop height data set in which the limit drop height value of the first sample device is less than the initial height, and C2 is a drop height data set in which the limit drop height value of the first sample device is greater than the maximum value of the test height range allowed by the test device.

[0083] By maximizing the function value of the maximum likelihood function in the above formula, the model parameters of the corresponding target distribution model can be obtained: and That is, in l(k,λ|h0,h1…h m ) obtains the maximum value, and obtains the corresponding and and will As k, As λ. In this way, the first distribution parameter of the target distribution model can be obtained.

[0084] In step S1022, the first distribution result is determined according to the first distribution parameter.

[0085] After obtaining the first distribution parameter, a distribution probability diagram of the target distribution model can be obtained, that is, a first distribution result corresponding to the target drop height is obtained.

[0086] The above step S104 is described in detail below. Figure 3 As shown, in step S104, performing distribution fitting on the target distribution model according to the plurality of sample height data to obtain a target distribution result corresponding to the target drop height may include the following steps:

[0087] In step S1041, based on the plurality of sample height data, target safe falling times corresponding to a plurality of second sample devices are obtained.

[0088] The target number of safe drops is the maximum number of drops under which the screen of the second sample device does not break when the screen of the second sample device breaks after the second sample device falls from the height corresponding to the sample height data one or more times.

[0089] After the sample height data is determined, a certain number of second sample devices are allocated for each sample height data. Then, for each sample height data, a further drop experiment is performed. From the previous analysis, it can be seen that the randomness equivalent to setting the drop height limit value (i.e., the target drop height) in this step comes from the differences caused by the drops. Therefore, in this drop experiment, for each second sample device, one or more drops are performed according to the drop height corresponding to the allocated sample height data until the screen of the second sample device is shattered, and the target number of safe drops after the shattering is determined. The target number of safe drops represents the maximum number of drops at which the second sample device falls but the screen does not shatter. For example, if the second sample device shatters after the fifth drop, then the target number of safe drops can be determined to be 4 times.

[0090] In step S1042, distribution fitting is performed on the target distribution model according to the target safe drop times to obtain a target distribution result corresponding to the target drop height.

[0091] In one embodiment, the target fragmentation probability corresponding to each sample height data can be determined according to the target safe drop times corresponding to each sample height data. Then, the target distribution model can be fitted according to the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain the target distribution result corresponding to the target drop height.

[0092] It can be understood that in the drop test, the screen of the second sample device is either broken or intact, so it can be obtained that the target safe drop number follows the geometric distribution X j ~GEOM(p(h j )).

[0093] For example, if the drop height h corresponding to the sample height data is j n j The target safe drop times observed in the second sample equipment experiment are Then the maximum likelihood method can be used to estimate the parameters of the geometric distribution in, yes The average value of is p(h j ) is an estimated value.

[0094] In this embodiment, p(h j ) is actually physical: the drop height h corresponding to the sample height data j The probability of a single drop and fragmentation is also the probability of target fragmentation. Therefore, through mathematical derivation, we can get p(h j )=Pr(H <hj )=F(h j ).

[0095] For example, performing distribution fitting on the target distribution model based on the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain a target distribution result corresponding to the target drop height may include: performing distribution fitting on the target distribution model based on the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain a second distribution parameter of the target distribution model; and determining the target distribution result based on the second distribution parameter.

[0096] It can be seen from the principle section above that for any second sample device, the drop height limit value H obeys the Weibull distribution. Therefore, the following formula can be obtained based on the Weibull distribution function:

[0097]

[0098] Among them, h j is the sample height data, F(h j ) is the Weibull distribution function of the sample height data, k is the shape parameter, and λ is the scale parameter.

[0099] According to mathematical derivation, the above equation can be converted into the following equation:

[0100]

[0101] Furthermore, in order to facilitate calculation, we can define x=ln(h j ), The above formula can be converted to:

[0102] y=kx-kln(λ)

[0103] Through linear regression, k and λ in the above formula can be solved, and then the second distribution parameter of the target distribution model can be obtained. After obtaining the second distribution parameter, the distribution probability diagram of the target distribution model can be obtained, that is, the target distribution result corresponding to the target drop height can be obtained.

[0104] In addition, after obtaining the target distribution results corresponding to the target drop height, in order to more intuitively analyze the reliability of the sample device screen, descriptive indicator parameters can also be determined based on the target distribution results, such as the estimated value of the mean or median of the target drop height.

[0105] In order to verify the accuracy of this embodiment, a simulation experiment is illustrated by taking the target distribution model including the Weibull distribution model as an example, in which the drop height limit value satisfies all the conditions proposed in the above embodiment.

[0106] First, the parameters of the Weibull distribution model of the target drop height can be randomly generated, for example: shape parameter k = 4.664, scale parameter λ = 2.052. Assume that the test height range of the test equipment is 0.6m to 2m, and a total of 40 sample equipment are included, and any 10 of them are used as the first sample equipment.

[0107] The first step is to set the initial height h0 = 0.6m, and generate the drop height data of each first sample device according to the true distribution of the Weibull distribution model corresponding to the randomly generated target drop height, as shown in the following table:

[0108]

[0109]

[0110] The table above shows the drop height data corresponding to 10 first sample devices. The first column in the table above indicates the device number of the first sample device, the second column indicates the drop height data corresponding to each first sample device, the third column indicates whether each drop height data is left-censored data, and the fourth column in the table above indicates whether each drop height data is right-censored data. The left-censored data indicates that the drop height limit value (that is, the drop height data) of the first sample device is less than the initial height (that is, the screen is shattered after the first drop), and the right-censored data indicates that the drop height limit value of the first sample device is greater than the maximum value of the test height range allowed by the test device (that is, the screens of the first sample devices are all intact and not shattered).

[0111] For example, the left censoring state of the first sample device 4 in the above table is FALSE, and the right censoring state is TRUE, then it can be seen that the drop height data corresponding to the first sample device 4 is censored data, and specifically right censored data. The left censoring state of the first sample device 5 in the above table is FALSE, and the right censoring state is FALSE, then it can be seen that the drop height data corresponding to the first sample device 5 is non-censored data. The types of drop height data corresponding to the remaining first sample devices (that is, censored data or non-censored data) can be obtained in the same way, and will not be repeated here.

[0112] After obtaining the drop height data corresponding to each first sample device, the model parameters of the Weibull distribution model corresponding to the maximum likelihood function of the Weibull distribution model when the function value is the largest can be determined according to the drop height data, that is, That is, the first distribution parameter of the Weibull distribution model is obtained, and then the first distribution result of the Weibull distribution model can be obtained.

[0113] The second step is, after obtaining the first distribution result of the Weibull distribution model, based on the height distribution interval corresponding to the first distribution result and combined with the test height interval of the test equipment, the target distribution interval is set to include the 20% quantile to the 53% quantile. Assuming that 4 sample height data need to be obtained, 4 quantiles can be evenly selected from the 20% quantile to the 53% quantile to determine the sample height data, that is, the 20% quantile is 1.72197196 meters, the 31% quantile is 1.83283599 meters, the 42% quantile is 1.92137201 meters and the 53% quantile is 2 meters. For each sample height data, a certain number of second sample devices are allocated (including devices in the sample devices other than the first sample devices, and devices of the first sample devices that did not break after the first step of the drop test, a total of 35).

[0114] According to the true distribution of the Weibull distribution model corresponding to the randomly generated target drop height, the target fragmentation probability corresponding to each sample height data is calculated, and then substituted into the geometric distribution to obtain the target safe drop times in the following table:

[0115]

[0116]

[0117] Among them, the first column in the above table represents the device number corresponding to the second sample device, the second column represents the sample height data corresponding to each second sample device, and the third column represents the target safe falling number corresponding to each second sample device.

[0118] For example, the sample height data corresponding to the second sample device 20 in the above table is 1.92137201m, and the corresponding target safe drop number is 2 times, that is, the second sample device falls from a drop height of 1.92137201m for the third time, and the screen is shattered. The target safe drop numbers corresponding to the other second sample devices can be obtained in the same way, and will not be repeated here.

[0119] After obtaining the target number of safe drops corresponding to the second sample device, Figures 1 to 3 The method and formula in the embodiment obtain the corresponding target fragmentation probability and the values ​​of x and y, as shown in the following table:

[0120] Sample height data (unit: m) Target fragmentation probability x y 1.72197196 0.375 0.54347012 -0.7550149 1.83283599 0.42857143 0.60586449 -0.5805048 1.92137201 0.45 0.65303952 -0.5144371 2 0.61538462 0.69314718 -0.0455085

[0121] Substitute the data in the above table into y=kx-kln(λ) mentioned above, and let According to the linear regression calculation method, we can get Transform to obtain the estimated values ​​of parameters k and λ corresponding to the Weibull distribution model That is, the second distribution parameter of the Weibull distribution model is obtained, and then the target distribution result corresponding to the target drop height can be obtained.

[0122] According to the parameters of the Weibull distribution model for randomly generating the target drop height mentioned above, namely the shape parameter k = 4.664 and the scale parameter λ = 2.052, the true mean of the target drop height can be obtained to be 1.876. According to the first step, the first distribution parameter corresponding to the first distribution result is It can be determined that the corresponding mean is 1.953, and the error is 4.11% compared with the true mean. According to the second distribution parameter corresponding to the target distribution result obtained in the second step It can be determined that the corresponding mean is 1.897, and the error is 1.12% compared with the true mean. It can be seen that in this embodiment, after the data calculation of the first and second steps, the error between the mean corresponding to the target distribution result finally obtained and the true mean is small, and the accuracy of the mean corresponding to the target distribution result is significantly greater than the accuracy of the mean corresponding to the first distribution result obtained only by the first step.

[0123] Furthermore, according to the parameters of the Weibull distribution model for randomly generating the target drop height mentioned above, namely, shape parameter k = 4.664, scale parameter λ = 2.052, it can be obtained that the true median of the target drop height is 1.897. It can be determined that the corresponding median is 1.981, and the error is 4.46% compared with the true mean. According to the second distribution parameter corresponding to the target distribution result obtained in the second step It can be determined that the corresponding median is 1.914, and the error is 0.899% compared with the true mean. It can be seen that in this embodiment, after the data calculation in the first and second steps, the error between the median corresponding to the target distribution result and the true median is small, and the accuracy of the median corresponding to the target distribution result is significantly greater than the accuracy of the median corresponding to the first distribution result obtained only by the first step.

[0124] Using the above method, the target distribution model is firstly subjected to the first distribution fitting according to the drop height data to obtain the first distribution result corresponding to the target drop height. Since there may be missing data in the drop height data, after obtaining the first distribution result, multiple sample height data located in the height distribution interval corresponding to the first distribution result can be further screened out according to the first distribution result. In this way, it can be ensured that the acquired sample height data are all non-missing data, and then the target distribution model is subjected to the second distribution fitting according to the obtained multiple sample height data to obtain the target distribution result corresponding to the drop height of the target height. By obtaining multiple sample height data by the above regression method, it can be ensured that there is no missing data in the data finally used for distribution fitting without increasing the number of samples, thereby improving the accuracy of distribution fitting of the target distribution model, and thus ensuring the reliability and accuracy of the target distribution result.

[0125] Figure 4 is a block diagram of a device for determining a drop height according to an exemplary embodiment. Figure 4 As shown, the device 200 includes:

[0126] An acquisition module 201 is configured to acquire drop height data corresponding to a plurality of first sample devices, where the drop height data represents a drop height at which a screen of the first sample device is broken when the device falls;

[0127] A first fitting module 202 is configured to perform distribution fitting on a target distribution model according to the drop height data to obtain a first distribution result corresponding to a target drop height, where the target drop height represents a drop height limit value at which a screen of the sample device will not break when the sample device drops;

[0128] A determination module 203 is configured to determine a plurality of sample height data according to the first distribution result, wherein the sample height data is located within a height distribution interval corresponding to the first distribution result;

[0129] The second fitting module 204 is configured to perform distribution fitting on the target distribution model according to the plurality of sample height data to obtain a target distribution result corresponding to the target drop height.

[0130] Optionally, the first fitting module 202 is configured to perform distribution fitting on the target distribution model according to the drop height data to obtain first distribution parameters of the target distribution model; and determine the first distribution result according to the first distribution parameters.

[0131] Optionally, the drop height data includes censored data and / or non-censored data, the censored data indicating that the drop height at which the screen of the first sample device is shattered when it falls is outside a test height interval corresponding to the test device, and the non-censored data indicating that the drop height at which the screen of the first sample device is shattered when it falls is within the test height interval, and the test device is used to test the drop height at which the screen of the first sample device is shattered when it falls within the test height interval; the first fitting module 202 is configured to perform distribution fitting on the target distribution model according to the censored data and / or the non-censored data to obtain the first distribution parameters.

[0132] Optionally, the first fitting module 202 is configured to determine the maximum likelihood function of the target distribution model based on the missing data and / or the non-missing data; when the function value of the maximum likelihood function is maximum, the model parameter value corresponding to the target distribution model is used as the first distribution parameter.

[0133] Optionally, the determination module 203 is configured to determine a target distribution interval from the height distribution interval corresponding to the first distribution result; and obtain the plurality of sample height data according to the target distribution interval.

[0134] Optionally, the second fitting module 204 is configured to obtain a target safe number of drops corresponding to multiple second sample devices based on the multiple sample height data; the target safe number of drops is the maximum number of drops at which the screen of the second sample device does not shatter when the screen of the second sample device shatters after the second sample device falls one or more times from the height corresponding to the sample height data; and distribution fitting is performed on the target distribution model based on the target safe number of drops to obtain a target distribution result corresponding to the target drop height.

[0135] Optionally, the second fitting module 204 is configured to determine the target fragmentation probability corresponding to the sample height data based on the number of target safe falls corresponding to each sample height data; and perform distribution fitting on the target distribution model based on the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain the target distribution result corresponding to the target drop height.

[0136] Optionally, the second fitting module 204 is configured to perform distribution fitting on the target distribution model according to the multiple sample height data and the target fragmentation probability corresponding to each sample height data to obtain a second distribution parameter of the target distribution model; and determine the target distribution result according to the second distribution parameter.

[0137] Optionally, the target distribution model includes a Weibull distribution model.

[0138] Using the above device, the target distribution model is firstly subjected to a first distribution fitting according to the drop height data to obtain a first distribution result corresponding to the target drop height. Since there may be missing data in the drop height data, after obtaining the first distribution result, multiple sample height data located in the height distribution interval corresponding to the first distribution result can be further screened out according to the first distribution result. In this way, it can be ensured that the acquired sample height data are all non-missing data, and then the target distribution model is subjected to a second distribution fitting according to the obtained multiple sample height data to obtain the target distribution result corresponding to the drop height of the target height. By obtaining multiple sample height data by the above regression method, it can be ensured that there is no missing data in the data finally used for distribution fitting without increasing the number of samples, thereby improving the accuracy of the distribution fitting of the target distribution model, and thus ensuring the reliability and accuracy of the target distribution result.

[0139] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0140] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon, and when the program instructions are executed by a processor, the steps of the method for determining the drop height provided by the present disclosure are implemented.

[0141] In another exemplary embodiment, a computer program product is further provided. The computer program product includes a computer program executable by a programmable device. The computer program has a code portion for executing the above-mentioned method for determining the drop height when executed by the programmable device.

[0142] Figure 5 is a block diagram of an electronic device 300 according to an exemplary embodiment. For example, the electronic device 300 may be provided as a server. Figure 5 The electronic device 300 includes a processing component 322, which further includes one or more processors, and a memory resource represented by a memory 332 for storing instructions executable by the processing component 322, such as an application. The application stored in the memory 332 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 322 is configured to execute instructions to perform the above-mentioned method for determining the drop height.

[0143] The electronic device 300 may also include a power supply component 326 configured to perform power management of the electronic device 300, a wired or wireless network interface 350 configured to connect the electronic device 300 to a network, and an input / output interface 358. The electronic device 300 may operate based on an operating system stored in the memory 332, such as Windows Server 2008. TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM or similar.

[0144] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the present disclosure. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0145] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for determining a drop height, characterized in that: The method comprises: Acquire drop height data corresponding to a plurality of first sample devices, wherein the drop height data represents a drop height at which a screen of the first sample device is broken when the device is dropped; According to the drop height data, a target distribution model is distributedly fitted to obtain a first distribution result corresponding to the target drop height, wherein the target drop height represents a drop height limit value at which a screen of the sample device will not be broken when the sample device is dropped; Determine a plurality of sample height data according to the first distribution result, wherein the sample height data is located within a height distribution interval corresponding to the first distribution result; According to the plurality of sample height data, distribution fitting is performed on the target distribution model to obtain a target distribution result corresponding to the target drop height.

2. The method according to claim 1, characterized in that: The step of performing distribution fitting on the target distribution model according to the drop height data to obtain a first distribution result corresponding to the target drop height includes: According to the drop height data, distribution fitting is performed on the target distribution model to obtain a first distribution parameter of the target distribution model; The first distribution result is determined according to the first distribution parameter.

3. The method according to claim 2, characterized in that The drop height data includes censored data and / or non-censored data, the censored data indicating that the drop height at which the screen of the first sample device is shattered when it is dropped is outside the test height interval corresponding to the test device, and the non-censored data indicating that the drop height at which the screen of the first sample device is shattered when it is dropped is within the test height interval, and the test device is used to test the drop height at which the screen of the first sample device is shattered when it is dropped within the test height interval; The performing distribution fitting on the target distribution model according to the drop height data to obtain the first distribution parameter of the target distribution model comprises: Distribution fitting is performed on the target distribution model according to the censored data and / or the non-censored data to obtain the first distribution parameters.

4. The method according to claim 3, characterized in that The performing distribution fitting on the target distribution model according to the censored data and / or the non-censored data to obtain the first distribution parameter comprises: Determining a maximum likelihood function of the target distribution model based on the censored data and / or the non-censored data; When the function value of the maximum likelihood function is maximum, the model parameter value corresponding to the target distribution model is used as the first distribution parameter.

5. The method according to claim 1, characterized in that Determining a plurality of sample height data according to the first distribution result comprises: Determine a target distribution interval from the height distribution interval corresponding to the first distribution result; According to the target distribution interval, the plurality of sample height data are acquired.

6. The method according to claim 1, characterized in that The step of performing distribution fitting on the target distribution model according to the plurality of sample height data to obtain a target distribution result corresponding to the target drop height includes: Based on the multiple sample height data, a target number of safe drops corresponding to multiple second sample devices is obtained; the target number of safe drops is the maximum number of drops obtained without causing the screen of the second sample device to break when the screen of the second sample device breaks after the second sample device falls from the height corresponding to the sample height data one or more times; According to the target safe falling times, distribution fitting is performed on the target distribution model to obtain a target distribution result corresponding to the target falling height.

7. The method according to claim 6, characterized in that The step of performing distribution fitting on the target distribution model according to the target safe falling times to obtain a target distribution result corresponding to the target falling height includes: Determine the target fragmentation probability corresponding to the sample height data according to the target safe drop number corresponding to each sample height data; According to the multiple sample height data and the target fragmentation probability corresponding to each sample height data, distribution fitting is performed on the target distribution model to obtain a target distribution result corresponding to the target drop height.

8. The method according to claim 7, characterized in that The performing distribution fitting on the target distribution model according to the plurality of sample height data and the target fragmentation probability corresponding to each of the sample height data to obtain a target distribution result corresponding to the target drop height includes: According to the plurality of sample height data and the target fragmentation probability corresponding to each of the sample height data, distribution fitting is performed on the target distribution model to obtain a second distribution parameter of the target distribution model; The target distribution result is determined according to the second distribution parameter.

9. The method according to any one of claims 1 to 8, characterized in that The target distribution model includes a Weibull distribution model.

10. A device for determining a drop height, characterized in that: The device comprises: an acquisition module, configured to acquire drop height data corresponding to a plurality of first sample devices, wherein the drop height data represents a drop height at which a screen of the first sample device is broken when the device falls; A first fitting module is configured to perform distribution fitting on a target distribution model according to the drop height data to obtain a first distribution result corresponding to a target drop height, wherein the target drop height represents a drop height limit value at which a screen of the sample device will not be broken when the sample device is dropped; A determination module is configured to determine a plurality of sample height data according to the first distribution result, wherein the sample height data is located within a height distribution interval corresponding to the first distribution result; The second fitting module is configured to perform distribution fitting on the target distribution model according to the multiple sample height data to obtain a target distribution result corresponding to the target drop height.

11. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to implement the steps of the method described in any one of claims 1 to 9 when calling the executable instructions stored in the memory.

12. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the steps of the method described in any one of claims 1 to 9 are implemented.

Images