Mobile phone battery puncture detection system and method based on Internet of Things

Abstract

The invention discloses a mobile phone battery puncture detection system and method based on the Internet of Things, and relates to the technical field of the Internet of Things, and the method comprises the steps: obtaining a plurality of historical puncture detection records of a mobile phone battery, obtaining a characteristic value and a characteristic degree of the mobile phone battery, and extracting a characteristic detection record; obtaining a characteristic ratio of the mobile phone battery according to the characteristic degree, and establishing a battery detection model according to the characteristic value and the characteristic ratio; dividing the batteries into a plurality of classification categories, setting an initial feature ratio range of each classification category, and determining a final feature ratio range of each category according to the feature ratio of each feature detection record; all batteries in the current batch are obtained, all the batteries are classified into corresponding categories, the target sampling number corresponding to all the categories is determined, and then all the target sampling batteries are determined. According to the invention, the actual state of the battery is analyzed, and sampling data is selected intelligently and reasonably, so that the sampled battery sample is more suitable for puncture detection.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a mobile phone battery puncture detection system and method based on the Internet of Things. Background Art

[0002] Mobile phone battery puncture test is an important test method for testing battery safety. It simulates mechanical damage that the battery may suffer during use, such as being pierced by sharp objects and other extreme situations, to evaluate whether the battery is sufficiently safe and whether it can avoid dangerous situations such as leakage, explosion, and fire, thereby providing a basis for battery quality control, technical improvement, and safe use. The puncture test mainly pierces the battery cell with a steel needle to cause a short circuit between the positive and negative electrodes, simulating the situation of an internal short circuit in the battery. However, in the existing mobile phone battery puncture test, samples are usually collected by manual sampling or simple random sampling methods. The above methods will have the problems of high subjectivity of sampling data, improper selection, and low efficiency. The actual state of the battery and various characteristic parameters are not fully considered, and the actual impact on the puncture test results is not fully considered, resulting in the lack of representativeness of the sampled data. Therefore, the sampling data should be selected intelligently and reasonably according to the actual state of the battery and various characteristic parameters, so that the sampled battery is more suitable for puncture testing, so as to fully understand the actual condition of the battery. Summary of the invention

[0003] The purpose of the present invention is to provide a mobile phone battery puncture detection system and method based on the Internet of Things to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solutions: A method for detecting mobile phone battery puncture based on the Internet of Things comprises the following steps: Step S100: obtaining puncture test records of several mobile phone batteries of the same batch in history, obtaining characteristic values ​​of the mobile phone batteries according to the energy density and battery power of the mobile phone batteries during the test, and obtaining characteristic degrees of the mobile phone batteries in combination with pressure value changes and temperature value changes during the test, and then extracting characteristic test records therein according to the characteristic degrees; In this solution, the models of the mobile phone batteries in the same batch are the same, and for the reliability of the results of this solution, the models of all the batteries in the current batch in step S400 are the same as the models of the mobile phone batteries in step S100; Step S200: obtaining the characteristic ratio of the mobile phone battery recorded in each characteristic detection according to the characteristic degree, and then establishing a battery detection model between the characteristic ratio and the characteristic value; Step S300: setting the number of classification categories into which the feature detection records are divided, and setting the initial feature ratio range of each category, and determining the final feature ratio range of each category according to the feature ratio corresponding to each feature detection record; Step S400: Obtain the characteristic values ​​of all batteries in the current batch, and classify each battery into a corresponding category according to the battery detection model and the characteristic ratio range, determine the current total sampling quantity, and determine the target sampling quantity corresponding to each category according to the category to which each battery in the current batch belongs, and then determine all target sampled batteries in the current batch.

[0005] Furthermore, step S100 includes: Step S110: generating a unique puncture test record according to the puncture test on the single battery, obtaining a plurality of puncture test records, extracting the energy density and battery power of the battery during the test, the change of the pressure value applied to the battery during the test, and the change of the temperature value of the battery collected from the puncture test records; Get the normalized energy density D of a battery b corresponding to a puncture detection record b and battery level Q b , get the characteristic value of a battery b , where e is the natural index, W1 and W2 are the first weight and the second weight respectively; and the pressure value S is used as the horizontal coordinate and the temperature value T is used as the vertical coordinate to establish a relationship line graph of a battery b; Step S120: extract the first coordinate point P1 (S1, T1) and the second coordinate point P2 (S2, T2) in the relationship line graph, where S1 is the minimum pressure value in the relationship line graph, T1 is the temperature value corresponding to the pressure value S1, S2 is the maximum pressure value in the relationship line graph, and T2 is the temperature value corresponding to the pressure value S2. If the temperature value corresponding to each coordinate point between coordinate points P1 and P2 is higher than the temperature value of the previous coordinate point, a battery b is marked and the battery slope of the battery b is obtained: W b =(T2-T1) / (S2-S1), and then the characteristic degree C of a battery b is obtained b =K C *V b *W b , K C is the slope coefficient; Then the characteristic degree of each marked battery is obtained, and the maximum characteristic degree C max The corresponding battery is taken as st, and the battery slope of battery st is taken as W st , calculate the battery slope and W of each marked battery st The puncture detection records corresponding to the marked batteries whose variance is less than the variance threshold are taken as feature detection records.

[0006] Further, step S200 includes: according to the maximum value C of the characteristic degree max , divide the characteristic degree of each marked battery by C max, get the characteristic ratio corresponding to each marked battery, the numerical range of the characteristic ratio is [0,1], and use the characteristic value as the horizontal coordinate and the characteristic ratio as the vertical coordinate to establish a two-dimensional coordinate system, obtain the characteristic ratio and characteristic value of the battery corresponding to each characteristic detection record, and according to the least squares method, get the battery detection model R=k*V+h, where R is the characteristic ratio, V is the characteristic value, k is the slope, and h is the intercept.

[0007] It should be noted that in this scheme, the characteristic degree of the mobile phone battery after being subjected to puncture pressure is represented by the speed of temperature rise and the characteristic value of the battery, and the characteristic degree is equivalent to the danger level of the mobile phone battery. In actual use, when the mobile phone battery is punctured, the faster the temperature rises, the more dangerous the battery is and the greater the degree of danger. The reasons that affect the danger level of the mobile phone battery are related to the characteristics of the mobile phone battery itself. In this scheme, the relevant characteristics are specifically energy density and battery capacity. Batteries with high energy density and battery capacity can store more energy per unit mass or unit volume. When the battery is punctured, the more heat released by such high-energy batteries means that the greater the degree of danger. The larger the characteristic value, the greater the characteristic degree, that is, the larger the characteristic ratio. Therefore, this scheme uses a battery detection model to determine the relationship between the characteristic value and the characteristic ratio. The battery detection model provides strong support for the analysis and calculation of the current mobile phone battery in the following steps.

[0008] Each detection record in the characteristic detection record is obtained by using the battery s corresponding to the maximum characteristic degree as the judgment standard. This is because the battery s here is equivalent to a battery with a higher degree of danger, and the battery slope corresponding to this type of battery in this scheme is usually not too small. Batteries with too small battery slopes may have abnormal sensor failures during the detection process. Since such batteries are not representative, the detection records corresponding to such batteries are not reliable enough for establishing a battery detection model, so battery s is needed as the judgment standard.

[0009] Furthermore, step S300 includes: Step S310: Obtain the feature ratios corresponding to all feature detection records, set the number of categories for classifying feature detection records to M, and set the initial feature ratio range of the mth category to R m ∈((m-1) / M, m / M], 1≤m≤M; the characteristic ratio is in R m The number of feature detection records in A m , get the category weight X=A of the mth category m / m, and then get the category weight of each category, add them up to get X1, divide each category weight by X1 to get the category ratio corresponding to each category; Step S320: Subtract the minimum value from the maximum value in the characteristic ratio range R m to obtain the range difference of the m-th category. If the category ratio of a certain category divided by the corresponding range difference results in D0 being less than D1, where D1 is the first result value, then take the category ratio of that category as D1; if D0 > D2, where D2 is the second result value, take the category ratio of that category as D2, with 0 < D1 < D2. Thus, obtain the sum D sum of the category ratios of all categories, and divide each category ratio by D sum to obtain the iterative ratio of each category. Take each iterative ratio as the new range difference of each category, and based on the range difference, obtain the final corresponding characteristic ratio range for each category.

[0010] This step is to summarize the rules through the historical characteristic detection records of each feature, and based on the obtained rules, determine the target sampling batteries in the following steps. The number of categories in the classification here is set, and its specific value directly affects the determination of the sampling samples in the following steps. When the number of categories is set to be small, that is, when there are few categories, there may be problems such as insufficient representativeness and lack of refinement of the sampling data. However, when the number of categories is set to be large, there will be problems of complex calculation of the sampling data. Therefore, the detection categories in this solution should be determined according to the actual situation.

[0011] Furthermore, step S400 includes: Step S410: Obtain the characteristic values of all batteries in the current batch, substitute them into the battery detection model to obtain the characteristic ratio corresponding to each battery. Then, take the batteries with characteristic ratios within the characteristic ratio range of the same category as one category; set the total sampling number as PT, the initial adjustment value of each category is PT, the initial number of cycles is pt = 1, and the initial sampling number of each category is 0. Obtain the target value L q = k q * N q / W q for the q-th category, where k q is the adjustment value of the q-th category, N q is the number of batteries in the q-th category, and W q is the middle value of the characteristic ratio range corresponding to the q-th category; Step S420: According to the obtained target value of each category, increase the sampling number of the category with the largest target value by 1, then subtract 1 from the adjustment value corresponding to the category with the largest target value, increase the number of cycles pt by 1, and obtain the target values of each category again. Repeat the cycle until the number of cycles pt = PT and stop. Thus, obtain the sampling number corresponding to each category, randomly select the corresponding number of sampling batteries from the corresponding categories as the target sampling batteries, and perform battery puncture detection on the target sampling batteries.

[0012] The target value of the category here is the value used by each category to characterize the degree of detectability. Of course, when the number of batteries in a certain category is large and the corresponding intermediate value of the category is small, it means that the detectability value of the battery is larger and should be tested more. When it is determined that the batteries in a certain category should be tested, the corresponding adjustment value should be reduced. In this way, the actual number of batteries that need to be tested in each category can be reasonably determined.

[0013] A mobile phone battery puncture detection system based on the Internet of Things, including a feature detection record acquisition module, a battery detection model establishment module, a feature ratio range calculation module and a target sampling battery determination module; Feature detection record acquisition module: used to obtain the puncture detection records of several mobile phone batteries of the same batch in history, obtain the characteristic value of the mobile phone battery according to the energy density and battery power of the mobile phone battery during the detection, and obtain the characteristic degree of the mobile phone battery in combination with the pressure value change and temperature value change during the detection process, and then extract the feature detection records according to the characteristic degree; Battery detection model building module: used to obtain the characteristic ratio of the mobile phone battery recorded by each characteristic detection according to the characteristic degree, and then establish a battery detection model between the characteristic ratio and the characteristic value; Feature ratio range calculation module: used to set the number of classification categories into which feature detection records are divided, and to set the initial feature ratio range of each category, and to determine the final feature ratio range of each category according to the feature ratio corresponding to each feature detection record; Target sampling battery determination module: used to obtain the characteristic values ​​of all batteries in the current batch, and classify each battery into the corresponding category according to the battery detection model and characteristic ratio range, determine the current total sampling quantity, and determine the target sampling quantity corresponding to each category according to the category of each battery in the current batch, and then determine all target sampling batteries in the current batch.

[0014] Further, the feature detection record acquisition module includes a relationship line graph establishment unit and a feature detection record acquisition unit; Relationship line graph establishment unit: used to obtain the puncture test records of several mobile phone batteries of the same batch in history, obtain the characteristic values ​​of the mobile phone batteries according to the energy density and battery power of the mobile phone batteries during the test, and establish a relationship line graph according to the pressure value change and temperature value change during the test process; Feature detection record acquisition unit: used to obtain each marked battery and the feature degree of each marked battery according to the relationship line graph, and obtain the feature detection record according to the feature degree of the battery.

[0015] Further, the feature ratio range calculation module includes a classification category setting unit and a feature ratio range calculation unit; Classification category setting unit: used to obtain the feature ratios corresponding to all feature detection records, set the number of categories for classifying feature detection records, set the initial feature ratio range of each category, and obtain the category ratios of each category; Feature ratio range calculation unit: used to obtain the range difference of each category, and then determine the iterative ratio of each category based on the obtained category ratio, and use each iterative ratio as the new range difference of each category, and obtain the final feature ratio range corresponding to each category based on the range difference.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a mobile phone battery puncture detection system and method based on the Internet of Things, including: obtaining the puncture detection records of several mobile phone batteries in history, obtaining the characteristic values ​​and characteristic degrees of the mobile phone batteries, and then extracting the characteristic detection records therein; obtaining the characteristic ratio of the mobile phone battery according to the characteristic degree, and establishing a battery detection model according to the characteristic value and characteristic ratio; dividing the battery into several classification categories, setting the initial characteristic ratio range of each classification category, and determining the final characteristic ratio range of each category according to the characteristic ratio of each characteristic detection record; obtaining all batteries in the current batch, classifying each battery into the corresponding category, and determining the target sampling quantity corresponding to each category, and then determining all target sampling batteries. The present invention analyzes the actual state of the battery and intelligently and reasonably selects the sampling data, so that the sampled battery samples are more suitable for puncture detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of a flow chart of a method for detecting mobile phone battery puncture based on the Internet of Things of the present invention; Figure 2 The present invention is a structural diagram of a mobile phone battery puncture detection system based on the Internet of Things. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Example: Figure 1 As shown, the present invention provides a mobile phone battery puncture detection system and method technical solution based on the Internet of Things, comprising the following steps: Step S100: Obtain the puncture test records of several mobile phone batteries of the same batch in history, obtain the characteristic values ​​of the mobile phone batteries according to the energy density and battery power of the mobile phone batteries during the test, and obtain the characteristic degree of the mobile phone batteries in combination with the pressure value change and the temperature value change during the test, and then extract the characteristic test records according to the characteristic degree.

[0020] Step S110: generating a unique puncture test record according to the puncture test on the single battery, obtaining a plurality of puncture test records, extracting the energy density and battery power of the battery during the test, the change of the pressure value applied to the battery during the test, and the change of the temperature value of the battery collected from the puncture test records; Get the normalized energy density D of a battery b corresponding to a puncture detection record b and battery level Q b , get the characteristic value of a battery b , where e is the natural index, W1 and W2 are the first weight and the second weight respectively; and a relationship line graph of a battery b is established with the pressure value S as the horizontal coordinate and the temperature value T as the vertical coordinate.

[0021] Step S120: extract the first coordinate point P1 (S1, T1) and the second coordinate point P2 (S2, T2) in the relationship line graph, where S1 is the minimum pressure value in the relationship line graph, T1 is the temperature value corresponding to the pressure value S1, S2 is the maximum pressure value in the relationship line graph, and T2 is the temperature value corresponding to the pressure value S2. If the temperature value corresponding to each coordinate point between coordinate points P1 and P2 is higher than the temperature value of the previous coordinate point, a battery b is marked and the battery slope of the battery b is obtained: W b =(T2-T1) / (S2-S1), and then the characteristic degree C of a battery b is obtained b =K C *V b *W b , K C is the slope coefficient; Then the characteristic degree of each marked battery is obtained, and the maximum characteristic degree C max The corresponding battery is taken as st, and the battery slope of battery st is taken as W st , calculate the battery slope and W of each marked battery st The puncture detection records corresponding to the marked batteries whose variance is less than the variance threshold are taken as feature detection records.

[0022] Step S200: According to the characteristic degree, the characteristic ratio of the mobile phone battery recorded in each characteristic detection is obtained, and then a battery detection model between the characteristic ratio and the characteristic value is established.

[0023] According to the maximum value C of the feature degree max , divide the feature degree of each marked battery by C max , obtain the feature ratio corresponding to each marked battery. The numerical range of the feature ratio is [0, 1]. Taking the characteristic value as the abscissa and the feature ratio as the ordinate, establish a two-dimensional coordinate system, obtain the feature ratio and characteristic value of the battery corresponding to each feature detection record. According to the least squares method, the battery detection model is R = k*V + h, where R is the feature ratio, V is the characteristic value, k is the slope, and h is the intercept.

[0024] Step S300: Set the number of categories into which the feature detection records are divided, and set the initial feature ratio range for each category. Determine the final feature ratio range for each category according to the feature ratio corresponding to each feature detection record.

[0025] Step S310: Obtain the feature ratios corresponding to all feature detection records. Set the number of categories for classifying the feature detection records as M, and set the initial feature ratio range for the m-th category as R m ∈((m - 1) / M, m / M], 1 ≤ m ≤ M; Take the number of feature detection records with feature ratios within R m as A m , obtain the category weight X of the m-th category = A m / m, and then obtain the category weights of each category. Add them up to get X1, divide the category weights of each category by X1 to obtain the category ratios corresponding to each category; Step S320: Subtract the minimum value from the maximum value in the feature ratio range R m as the range difference of the m-th category. If the result D0 of dividing the category ratio of a certain category by the corresponding range difference is less than D1, where D1 is the first result value, then take the category ratio of a certain category as D1; if D0 > D2, where D2 is the second result value, take the category ratio of a certain category as D2, 0 < D1 < D2, and then obtain the sum D sum of the category ratios of all categories, and divide the category ratios of each category by D sum , obtain the iterative ratios of each category, take the iterative ratios as the new range differences of each category, and obtain the final corresponding feature ratio ranges of each category according to the range differences.

[0026] To better illustrate this solution, the following example is given: Set the number of categories M=5. Since the feature ratio is [0,1], it is equivalent to dividing the feature ratio into 5 equal parts, and the ranges of each feature ratio are 0 to 1 / 5, 1 / 5 to 2 / 5, 2 / 5 to 3 / 5, 3 / 5 to 4 / 5, and 4 / 5 to 1. Assuming that the number of feature detection records in each category is 10, 8, 9, 8, and 5, the category weights are 10, 4, 3, 2, and 1, and the sum of the category weights is X1=20. The category ratios corresponding to the categories are: 0.5, 0.2, 0.15, 0.1, and 0.05. In this embodiment, the first value D1 is 0.7, and the second value D2 is 1.3. Since the range difference is 0.2, the results D0 corresponding to each category are 2.5, 1, 0.75, 0.5, and 0.25, respectively. According to the evaluation criteria of this step, the category ratios of each category are adjusted to: 1.3, 1, 0.75, 0.7, and 0.7, respectively, and then the sum of the category ratios D is obtained. sum =4.45, and the iterative ratios of each category are 0.29, 0.22, 0.17, 0.16, and 0.16, respectively. Since the sum of all the iterative ratios is 1, each iterative ratio can be used as the new range difference of each category, and then according to the range difference, the final feature ratio range of each category can be re-obtained, which are 0 to 0.29, 0.29 to 0.51, 0.51 to 0.68, 0.68 to 0.84, and 0.84 to 1, respectively.

[0027] Step S400: Obtain the characteristic values ​​of all batteries in the current batch, and classify each battery into a corresponding category according to the battery detection model and the characteristic ratio range, determine the current total sampling quantity, and determine the target sampling quantity corresponding to each category according to the category to which each battery in the current batch belongs, and then determine all target sampled batteries in the current batch.

[0028] Step S410: Obtain the characteristic values ​​of all batteries in the current batch, substitute them into the battery detection model, obtain the characteristic ratio corresponding to each battery, and then classify the batteries with characteristic ratios within the characteristic ratio range of the same category as one category; set the total number of samples to PT, the initial adjustment value size of each category to PT, the initial number of cycles to pt=1, the initial number of samples for each category to 0, and obtain the target value of the qth category to L q =k q *N q / W q , k q is the adjustment value of the qth category, N q is the number of batteries in the qth category, W q is the middle value of the feature ratio range corresponding to the qth category.

[0029] Wq It is the middle value of the feature ratio range corresponding to the category. In this embodiment, the middle values ​​are 0.145, 0.4, 0.595, 0.76, and 0.92 respectively. Then, the target value of each category is obtained, and the sampling number of the category corresponding to the maximum target value is added by 1 to obtain the sampling number in all categories. Random sampling is performed in each category. In this way, the actual number of tests that need to be performed in each category can be reasonably determined.

[0030] Step S420: According to the target value of each category obtained, the sampling quantity of the category with the largest target value is increased by 1, and then the adjustment value corresponding to the category with the largest target value is reduced by 1, the number of cycles pt is increased by 1, and the target value of each category is obtained again, and the cycle is repeated in sequence until the number of cycles pt=PT, and then the sampling quantity corresponding to each category is obtained, and the corresponding number of sampling batteries are randomly obtained from the corresponding categories as target sampling batteries, and battery puncture detection is performed on the target sampling batteries.

[0031] The present invention also provides a mobile phone battery puncture detection system based on the Internet of Things, including a feature detection record acquisition module, a battery detection model establishment module, a feature ratio range calculation module and a target sampling battery determination module; Feature detection record acquisition module: used to obtain the puncture detection records of several mobile phone batteries of the same batch in history, obtain the characteristic value of the mobile phone battery according to the energy density and battery power of the mobile phone battery during the detection, and obtain the characteristic degree of the mobile phone battery in combination with the pressure value change and temperature value change during the detection process, and then extract the feature detection records according to the characteristic degree; Battery detection model building module: used to obtain the characteristic ratio of the mobile phone battery recorded by each characteristic detection according to the characteristic degree, and then establish a battery detection model between the characteristic ratio and the characteristic value; Feature ratio range calculation module: used to set the number of classification categories into which feature detection records are divided, and to set the initial feature ratio range of each category, and to determine the final feature ratio range of each category according to the feature ratio corresponding to each feature detection record; Target sampling battery determination module: used to obtain the characteristic values ​​of all batteries in the current batch, and classify each battery into the corresponding category according to the battery detection model and characteristic ratio range, determine the current total sampling quantity, and determine the target sampling quantity corresponding to each category according to the category of each battery in the current batch, and then determine all target sampling batteries in the current batch.

[0032] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A mobile phone battery puncture detection method based on the Internet of Things, characterized in that: The following steps are involved: Step S100: obtaining puncture test records of several mobile phone batteries of the same batch in history, obtaining characteristic values ​​of the mobile phone batteries according to the energy density and battery power of the mobile phone batteries during the test, and obtaining characteristic degrees of the mobile phone batteries in combination with pressure value changes and temperature value changes during the test, and then extracting characteristic test records therein according to the characteristic degrees; Step S200: obtaining the characteristic ratio of the mobile phone battery recorded in each characteristic detection according to the characteristic degree, and then establishing a battery detection model between the characteristic ratio and the characteristic value; Step S300: setting the number of classification categories into which the feature detection records are divided, and setting the initial feature ratio range of each category, and determining the final feature ratio range of each category according to the feature ratio corresponding to each feature detection record; Step S400: Obtain the characteristic values ​​of all batteries in the current batch, and classify each battery into a corresponding category according to the battery detection model and the characteristic ratio range, determine the current total sampling quantity, and determine the target sampling quantity corresponding to each category according to the category to which each battery in the current batch belongs, and then determine all target sampled batteries in the current batch.

2. According to the method for detecting mobile phone battery puncture based on the Internet of Things in claim 1, it is characterized in that: Step S100 includes: Step S110: generating a unique puncture test record according to the puncture test on the single battery, obtaining a plurality of puncture test records, extracting the energy density and battery power of the battery during the test, the change of the pressure value applied to the battery during the test, and the change of the temperature value of the battery collected from the puncture test records; Get the normalized energy density D of a battery b corresponding to a puncture detection record b and battery level Q b , get the characteristic value of a battery b , where e is the natural index, W1 and W2 are the first weight and the second weight respectively; and the pressure value S is used as the horizontal coordinate and the temperature value T is used as the vertical coordinate to establish a relationship line graph of a battery b; Step S120: extract the first coordinate point P1 (S1, T1) and the second coordinate point P2 (S2, T2) in the relationship line graph, where S1 is the minimum pressure value in the relationship line graph, T1 is the temperature value corresponding to the pressure value S1, S2 is the maximum pressure value in the relationship line graph, and T2 is the temperature value corresponding to the pressure value S2. If the temperature value corresponding to each coordinate point between coordinate points P1 and P2 is higher than the temperature value of the previous coordinate point, a battery b is marked and the battery slope of the battery b is obtained: W b =(T2-T1) / (S2-S1), and then the characteristic degree C of a battery b is obtained b =K C *V b *W b , K C is the slope coefficient; Then the characteristic degree of each marked battery is obtained, and the maximum characteristic degree C max The corresponding battery is taken as st, and the battery slope of battery st is taken as W st , calculate the battery slope and W of each marked battery st The puncture detection records corresponding to the marked batteries whose variance is less than the variance threshold are taken as feature detection records.

3. The method for detecting mobile phone battery puncture based on the Internet of Things according to claim 2, characterized in that: Step S200 includes: according to the maximum value C of the characteristic degree max , divide the characteristic degree of each marked battery by C max , get the characteristic ratio corresponding to each marked battery, the numerical range of the characteristic ratio is [0,1], and use the characteristic value as the horizontal coordinate and the characteristic ratio as the vertical coordinate to establish a two-dimensional coordinate system, obtain the characteristic ratio and characteristic value of the battery corresponding to each characteristic detection record, and according to the least squares method, get the battery detection model R=k*V+h, where R is the characteristic ratio, V is the characteristic value, k is the slope, and h is the intercept.

4. The method for detecting mobile phone battery puncture based on the Internet of Things according to claim 1, characterized in that: Step S300 includes: Step S310: Obtain the feature ratios corresponding to all feature detection records, set the number of categories for classifying feature detection records to M, and set the initial feature ratio range of the mth category to R m ∈((m-1) / M, m / M], 1≤m≤M; the characteristic ratio is in R m The number of feature detection records in A m , get the category weight X=A of the mth category m / m, and then get the category weight of each category, add them up to get X1, divide each category weight by X1 to get the category ratio corresponding to each category; Step S320: Subtract the minimum value from the maximum value in the characteristic ratio range R m as the range difference of the m-th category. If the category ratio of a certain category divided by the corresponding range difference results in D0 less than D1, where D1 is the first result value, then take the category ratio of the certain category as D1; if D0 > D2, where D2 is the second result value, take the category ratio of the certain category as D2, 0 < D1 < D2, and then obtain the sum D sum of the category ratios of all categories, and divide each category ratio by D sum to obtain the iterative ratio of each category. Take each iterative ratio as the new range difference of each category, and obtain the final corresponding characteristic ratio range of each category according to the range difference.

5. The method for detecting mobile phone battery puncture based on the Internet of Things according to claim 4, characterized in that: Step S400 includes: Step S410: Obtain the characteristic values ​​of all batteries in the current batch, substitute them into the battery detection model, obtain the characteristic ratio corresponding to each battery, and then classify the batteries with characteristic ratios within the characteristic ratio range of the same category as one category; set the total number of samples to PT, the initial adjustment value size of each category to PT, the initial number of cycles to pt=1, the initial number of samples for each category to 0, and obtain the target value of the qth category to L q =k q *N q / W q , k q is the adjustment value of the qth category, N q is the number of batteries in the qth category, W q is the middle value of the feature ratio range corresponding to the qth category; Step S420: According to the target value of each category obtained, the sampling quantity of the category with the largest target value is increased by 1, and then the adjustment value corresponding to the category with the largest target value is reduced by 1, the number of cycles pt is increased by 1, and the target value of each category is obtained again, and the cycle is repeated in sequence until the number of cycles pt=PT, and then the sampling quantity corresponding to each category is obtained, and the corresponding number of sampling batteries are randomly obtained from the corresponding categories as target sampling batteries, and battery puncture detection is performed on the target sampling batteries.

6. A mobile phone battery puncture detection system, used to execute a mobile phone battery puncture detection method based on the Internet of Things as described in any one of claims 1 to 5, characterized in that: The system includes a feature detection record acquisition module, a battery detection model establishment module, a feature ratio range calculation module and a target sampling battery determination module; Feature detection record acquisition module: used to obtain the puncture detection records of several mobile phone batteries of the same batch in history, obtain the characteristic value of the mobile phone battery according to the energy density and battery power of the mobile phone battery during the detection, and obtain the characteristic degree of the mobile phone battery in combination with the pressure value change and temperature value change during the detection process, and then extract the feature detection records according to the characteristic degree; Battery detection model building module: used to obtain the characteristic ratio of the mobile phone battery recorded by each characteristic detection according to the characteristic degree, and then establish a battery detection model between the characteristic ratio and the characteristic value; Feature ratio range calculation module: used to set the number of classification categories into which feature detection records are divided, and to set the initial feature ratio range of each category, and to determine the final feature ratio range of each category according to the feature ratio corresponding to each feature detection record; Target sampling battery determination module: used to obtain the characteristic values ​​of all batteries in the current batch, and classify each battery into a corresponding category according to the battery detection model and characteristic ratio range, determine the current total sampling quantity, and determine the target sampling quantity corresponding to each category according to the category to which each battery in the current batch belongs, and then determine all target sampling batteries in the current batch.

7. A mobile phone battery puncture detection system according to claim 6, characterized in that: The feature detection record acquisition module includes a relationship line graph establishment unit and a feature detection record acquisition unit; Relationship line graph establishment unit: used to obtain the puncture test records of several mobile phone batteries of the same batch in history, obtain the characteristic values ​​of the mobile phone batteries according to the energy density and battery power of the mobile phone batteries during the test, and establish a relationship line graph according to the pressure value change and temperature value change during the test process; Feature detection record acquisition unit: used to obtain each marked battery and the feature degree of each marked battery according to the relationship line graph, and obtain the feature detection record according to the feature degree of the battery.

8. A mobile phone battery puncture detection system according to claim 6, characterized in that: The feature ratio range calculation module includes a classification category setting unit and a feature ratio range calculation unit; Classification category setting unit: used to obtain the feature ratios corresponding to all feature detection records, set the number of categories for classifying feature detection records, set the initial feature ratio range of each category, and obtain the category ratios of each category; Feature ratio range calculation unit: used to obtain the range difference of each category, and then determine the iterative ratio of each category based on the obtained category ratio, and use each iterative ratio as the new range difference of each category, and obtain the final feature ratio range corresponding to each category based on the range difference.

Images