A risk assessment system for lithium battery separator production based on big data

Abstract

The present invention discloses a lithium battery diaphragm production risk assessment system based on big data, which belongs to the technical field of lithium battery diaphragm production risk assessment, and comprises a risk analysis module, a monitoring module and a risk identification module; the risk analysis module is used to analyze the production process of the lithium battery diaphragm, determine the potential risks corresponding to the production process; obtain risk material data corresponding to each potential risk; classify each potential risk, obtain each potential equipment risk and potential management risk; set each corresponding operation monitoring item and management monitoring item; and perform corresponding monitoring processing; the monitoring module is used to perform real-time monitoring, obtain corresponding operation monitoring data and management monitoring data; input the operation monitoring data and management monitoring data into a preset monitoring model for real-time display; the risk identification module is used to perform risk identification in the production process of the lithium battery diaphragm, and obtain corresponding operation risk identification results and management risk identification results.

Description

Technical Field

[0001] The present invention belongs to the technical field of lithium battery diaphragm production risk assessment, and specifically is a lithium battery diaphragm production risk assessment system based on big data. Background Art

[0002] With the rapid rise of the new energy vehicle market, the performance and safety of lithium batteries as its core power source are increasingly valued. As an important component of lithium batteries, the quality of lithium battery separators directly affects the performance and safety of lithium batteries. However, in the production process of lithium battery separators, due to the influence of various factors such as equipment accuracy, raw material quality, and process parameters, product quality may be unstable and even cause safety accidents.

[0003] At present, risk assessment in the production process of lithium battery separators mainly relies on traditional experience judgment and manual inspection, which is not only inefficient but also difficult to ensure the accuracy and real-time performance of the assessment. Based on this, the present invention provides a lithium battery separator production risk assessment system based on big data. Summary of the invention

[0004] In order to solve the problems existing in the above-mentioned solutions, the present invention provides a lithium battery separator production risk assessment system based on big data.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A lithium battery separator production risk assessment system based on big data, including a risk analysis module, a monitoring module and a risk identification module;

[0007] The risk analysis module is used to analyze the production process of lithium battery separators, obtain historical production data corresponding to the production process, determine the potential risks corresponding to the production process based on the historical production data; obtain risk material data corresponding to each potential risk; classify each potential risk according to the risk material data to obtain each potential equipment risk and potential management risk; and set operation monitoring items and management monitoring items corresponding to each potential equipment risk and potential management risk according to the risk material data; and perform corresponding monitoring and processing according to the obtained management monitoring items and operation monitoring items.

[0008] Furthermore, the method of classifying each potential risk according to the risk material data includes:

[0009] Establish an equipment risk judgment model, and the expression of the equipment risk judgment model is:

[0010]

[0011] In the formula: s is the input data, the input data is the risk material data; the output data is the device judgment value 1 or 0;

[0012] Analyze the data of each risk material through the equipment risk judgment model to obtain the equipment judgment value corresponding to each potential risk;

[0013] The potential risk with a device judgment value of 1 is marked as a potential device risk;

[0014] The potential risks with a device judgment value of 0 are marked as potential management risks.

[0015] The monitoring module is used to monitor each operation monitoring item and management monitoring item in real time, obtain corresponding operation monitoring data and management monitoring data; establish a monitoring model, and input the operation monitoring data and management monitoring data into the monitoring model for real-time display.

[0016] The risk identification module is used to perform risk identification in the production process of lithium battery separators, obtain operation monitoring data and management monitoring data corresponding to each operation monitoring item and management monitoring item according to the monitoring model; perform risk identification analysis on the operation monitoring data and management monitoring data respectively to obtain corresponding operation risk identification results and management risk identification results; input the obtained operation risk identification results and management risk identification results into the monitoring model for real-time display, and perform corresponding processing based on the displayed operation risk identification results and management risk identification results.

[0017] Furthermore, the method for risk identification of operation monitoring data includes:

[0018] Acquire the operation instructions corresponding to the operation monitoring items in real time, and determine the operation standards according to the operation instructions, wherein the operation standards are composed of the unit standards corresponding to each unit item;

[0019] Extracting features from the operation monitoring data according to the operation standard to obtain operation implementation data, wherein the operation implementation data is composed of various operation feature data, and the operation feature data corresponds to the corresponding unit standard;

[0020] Compare the implemented operation data with the operation standard to determine the corresponding operation difference value;

[0021] Generate a corresponding operation monitoring curve according to the monitoring time corresponding to the operation difference value, where the horizontal axis is the monitoring time and the vertical axis is the operation difference value; fit the operation monitoring curve to obtain an operation monitoring function, and mark the operation monitoring function as K(t), where t is the monitoring time;

[0022] Set the conversion formula. The conversion formula is

[0023] Where: ZW(t) is the conversion function, QA is the allowable value;

[0024] The operation monitoring function is transformed by a transformation formula to obtain a transformation function;

[0025] According to the formula Calculate the corresponding operation evaluation value;

[0026] Where: PUR is the operational assessment value;

[0027] Determine the operational risk identification results based on the operational assessment value.

[0028] Further, the method of comparing the implemented operation data with the operation standard includes:

[0029] Subtract each operation characteristic data from the unit standard to obtain the characteristic difference value corresponding to each unit item, and integrate each characteristic difference value into a matching characteristic;

[0030] Establishing a feature database, wherein the feature database is used to store feature value combinations, wherein the feature value combinations include combined feature values ​​and corresponding feature condition data;

[0031] Input the matching features into the feature database for matching, and obtain the combined feature values ​​corresponding to each unit item;

[0032] According to the formula Calculate the corresponding operation difference value;

[0033] Wherein: CR is the operating difference value; j represents the corresponding unit item, j = 1, 2, ..., m, m is a positive integer; βj is the proportional coefficient of the corresponding unit item, 0<βj≤1; HTj is the combined eigenvalue of the corresponding unit item.

[0034] Furthermore, the method for establishing the feature database includes:

[0035] Analyze the risk material data to obtain the unit material data corresponding to each unit item; define the benchmark item and fixed item; split the unit material data according to each fixed item to obtain the variable analysis data; and count the safety accident incidence rates corresponding to different characteristic differences based on the variable analysis data;

[0036] According to the formula TR = e δ -1 calculates the corresponding first eigenvalue; where TR is the first eigenvalue; e is a natural constant; δ is the safety accident rate;

[0037] Marking the fixed item data corresponding to the first eigenvalue as unit condition data;

[0038] Merging the condition data of each unit according to each first eigenvalue to obtain each eigenvalue combination;

[0039] A feature database is established based on the combination of each feature value.

[0040] Furthermore, the method for merging the unit condition data according to the first characteristic values ​​includes:

[0041] Step SA1: merge the unit condition data with the same first eigenvalue to obtain an initial set; sort the first eigenvalues ​​in descending order to obtain a first sequence;

[0042] Step SA2: marking the first eigenvalue ranked first in the first sequence as a reference eigenvalue; marking the first eigenvalue that is not a reference eigenvalue in the first sequence as an auxiliary eigenvalue;

[0043] Step SA3: Calculate the difference between the reference eigenvalue and each subsidiary eigenvalue, mark it as the first difference, merge each subsidiary eigenvalue whose first difference is less than the threshold value X1 with the reference eigenvalue to obtain the corresponding merged eigenvalue; remove the reference eigenvalue and the subsidiary eigenvalue corresponding to the merged eigenvalue from the first sequence to obtain a new first sequence; merge each initial set and unit condition data corresponding to the merged eigenvalue to obtain a merged set;

[0044] Step SA4: looping step SA2 to step SA3 until the first characteristic values ​​in the first sequence are merged, setting each matching condition according to the merged set, and integrating each matching condition into characteristic condition data.

[0045] Furthermore, in step SA3, the calculation method of the combined eigenvalue includes:

[0046] The subsidiary eigenvalues ​​and the reference eigenvalues ​​are marked as Pi, i = 1, 2, ..., n, and n is a positive integer; wherein p1 represents the reference eigenvalue, and p2 and onwards are subsidiary eigenvalues; the conditional quantities corresponding to the subsidiary eigenvalues ​​and the reference eigenvalues ​​are identified; and the conditional quantities are marked as LTi;

[0047] According to the formula Calculate the corresponding merged eigenvalue;

[0048] Where: HT is the combined eigenvalue.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] Through the mutual cooperation among the risk analysis module, monitoring module and risk identification module, intelligent risk assessment of lithium battery diaphragm production is realized based on big data, abnormalities are dynamically identified according to operational differences, and it is determined whether the operation is abnormal, and then the corresponding operational risk identification results are obtained. This solves the risk of ignoring the accumulation of operational deviations in the existing risk identification system and has a high early warning effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0053] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. 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.

[0054] like Figure 1 As shown, a lithium battery separator production risk assessment system based on big data includes a risk analysis module, a monitoring module and a risk identification module;

[0055] The risk analysis module is used to analyze the production process of lithium battery separators, obtain historical production data that serves as a reference for the production process based on big data, conduct statistics on the historical production data, determine various production risks that may occur in the production process, and mark each production risk as a potential risk; organize the historical production data corresponding to each potential risk to obtain risk material data corresponding to each potential risk; establish an equipment risk judgment model, which is used to analyze the risk material data to determine whether the potential risk is caused by the production equipment. If so, it is considered to meet the equipment risk judgment requirements, and other problems such as manual operation problems, material problems, and external factors are considered other situations; the expression of the equipment risk judgment model is In the formula: s is the input data, the input data is the risk material data; the output data is the device judgment value 1 or 0;

[0056] Analyze the data of each risk material through the equipment risk judgment model to obtain the equipment judgment value corresponding to each potential risk; classify each potential risk according to the obtained equipment judgment value to obtain potential equipment risk and potential management risk;

[0057] For potential management risks, determine the corresponding management monitoring items based on risk material data;

[0058] For potential equipment risks, determine which equipment operation abnormalities may lead to the potential equipment risk based on the risk material data, and mark the corresponding operations as operation monitoring items;

[0059] Corresponding monitoring processing is performed according to each obtained management monitoring item and operation monitoring item to realize monitoring of each management monitoring item and operation monitoring item.

[0060] The monitoring module is used to monitor each operation monitoring item and management monitoring item in real time to obtain corresponding operation monitoring data and management monitoring data; establish a monitoring model based on user display requirements, and the monitoring model is used for data display, which is established using technology; input the obtained operation monitoring data and management monitoring data into the monitoring model for real-time display.

[0061] The risk identification module is used to perform risk identification in the production process of lithium battery separators, and obtains the operation monitoring data and management monitoring data corresponding to each operation monitoring item and management monitoring item according to the monitoring model; performs risk identification analysis on the operation monitoring data and the management monitoring data respectively to obtain corresponding operation risk identification results and management risk identification results; performs risk identification on the management monitoring data by utilizing existing technology, which is not part of the invention of the present invention and is therefore not described in detail; inputs the obtained operation risk identification results and management risk identification results into the monitoring model for real-time display, and performs corresponding processing based on the displayed operation risk identification results and management risk identification results; that is, when the corresponding safety risk is identified, it is processed according to the preset risk processing method, and if there is no risk, no processing is performed.

[0062] Among them, the methods for risk identification of operation monitoring data include:

[0063] Obtain the operation instructions corresponding to the operation monitoring items in real time, determine the corresponding operation content according to the operation instructions, and set the operation standards according to the operation content, such as how much material should be added according to the instructions and when to add it, etc. According to the standards for different contents, the operation standards are regarded as composed of various unit standards, such as the unit standard corresponding to the delivery time and the unit standard corresponding to the amount of added materials; the data items corresponding to the unit standards are marked as unit items;

[0064] Extracting features from the operation monitoring data according to the operation standards, obtaining operation implementation data corresponding to the operation feature data corresponding to each unit standard, that is, the operation implementation data is composed of each operation feature data, and the operation feature data is the implementation operation data corresponding to the unit standard;

[0065] Compare the implemented operation data with the operation standard to determine the corresponding operation difference value.

[0066] Generate a corresponding operation monitoring curve according to the monitoring time corresponding to the operation difference value, where the horizontal axis is the monitoring time and the vertical axis is the operation difference value; fit the operation monitoring curve to obtain an operation monitoring function, and mark the operation monitoring function as K(t), where t is the monitoring time;

[0067] Set the allowable value, that is, if the operation difference value is within the allowable value, it can be ignored and regarded as normal. It can be calculated based on the safety accident rate and set according to user needs;

[0068] Set the conversion formula. The conversion formula is Where: ZW(t) is the conversion function, QA is the allowable value;

[0069] The operation monitoring function is transformed by a transformation formula to obtain a transformation function; the horizontal axis of the transformation function is the monitoring time, and the vertical axis is the transformation value;

[0070] According to the formula Calculate the corresponding operation evaluation value;

[0071] Where: PUR is the operation evaluation value; 0-t is the time corresponding to this production;

[0072] The operational risk identification result is determined based on the operational assessment value; that is, when the operational assessment value is greater than the preset value, it is considered to have operational risk, otherwise, there is no risk.

[0073] Methods for comparing implementation data to operational standards include:

[0074] Subtract each operation feature data from the unit standard to obtain the feature difference corresponding to each unit item, such as input time difference, input position distance difference, input amount difference, speed difference, etc., that is, perform quantitative subtraction. If there is non-numerical data, it can be assigned a value using existing quantitative methods and calculated based on the assigned value; integrate each feature difference into a matching feature;

[0075] Analyze the risk material data according to the control variable method, obtain the risk material data corresponding to each unit item, mark it as unit material data, that is, the data of other unit items in the unit material data are fixed, and only the data corresponding to the unit item is a variable, and then perform control variable analysis; mark the unit item as a variable in the unit material data as a benchmark item, and mark other unit items as fixed items; split the unit material data according to each fixed item, that is, split the unit material data with the same fixed item data into one category, and obtain each variable analysis data; count the safety accident incidence rate corresponding to different feature differences according to the variable analysis data;

[0076] According to the formula TR = e δ -1 calculates the corresponding first eigenvalue; where TR is the first eigenvalue; e is a natural constant; δ is the safety accident rate;

[0077] Marking the fixed item data corresponding to the first eigenvalue as unit condition data;

[0078] Merging each unit condition data according to each first eigenvalue to obtain each eigenvalue combination, wherein the eigenvalue combination includes a merged eigenvalue and corresponding eigencondition data;

[0079] A feature database is established, and each feature value combination is input into the feature database for storage.

[0080] Input the obtained matching features into the feature database for matching, and obtain the combined feature values ​​corresponding to each unit item;

[0081] According to the formula Calculate the corresponding operation difference value;

[0082] Wherein: CR is the operating difference value; j represents the corresponding unit item, j = 1, 2, ..., m, m is a positive integer; βj is the proportional coefficient of the corresponding unit item, 0<βj≤1; HTj is the combined eigenvalue of the corresponding unit item.

[0083] The method for merging the condition data of each unit according to each first eigenvalue includes:

[0084] Step SA1: merge the unit condition data with the same first eigenvalue to obtain an initial set, that is, a combination of the unit condition data; sort the first eigenvalues ​​in descending order to obtain a first sequence;

[0085] Step SA2: marking the first eigenvalue ranked first in the first sequence as a reference eigenvalue; marking the first eigenvalue that is not a reference eigenvalue in the first sequence as an auxiliary eigenvalue;

[0086] Step SA3: Calculate the difference between the reference eigenvalue and each subsidiary eigenvalue, mark it as the first difference, merge each subsidiary eigenvalue whose first difference is less than the threshold value X1 with the reference eigenvalue to obtain the corresponding merged eigenvalue; remove the reference eigenvalue and the subsidiary eigenvalue corresponding to the merged eigenvalue from the first sequence to obtain a new first sequence; merge each initial set and unit condition data corresponding to the merged eigenvalue to obtain a merged set;

[0087] Step SA4: loop step SA2 to step SA3 until the first feature values ​​in the first sequence are merged, calculate the feature difference values ​​according to the merged set, that is, the feature difference values ​​corresponding to the unit condition data, integrate them into matching conditions, and integrate the matching conditions into feature condition data.

[0088] The calculation method of the combined eigenvalue includes:

[0089] The subsidiary eigenvalues ​​and the reference eigenvalues ​​are marked as Pi, i = 1, 2, ..., n, and n is a positive integer; wherein p1 represents the reference eigenvalue, and p2 and onwards are subsidiary eigenvalues; the number of unit conditional data corresponding to the subsidiary eigenvalues ​​and the reference eigenvalues ​​is identified, that is, the number of data in the initial set corresponding to the initial set, and the number is 1 if there is no initial set, and the corresponding number is marked as the conditional quantity value; the conditional quantity value is marked as LTi;

[0090] According to the formula Calculate the corresponding merged eigenvalue; where: HT is the merged eigenvalue.

[0091] The above formulas are all calculated by removing dimensions and taking numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0092] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A lithium battery separator production risk assessment system based on big data, characterized in that: It includes risk analysis module, monitoring module and risk identification module; The risk analysis module is used to analyze the production process of the lithium battery separator, obtain the historical production data corresponding to the production process, determine the potential risks corresponding to the production process based on the historical production data; and obtain the risk material data corresponding to each potential risk; Classify each potential risk according to the risk material data to obtain each potential equipment risk and potential management risk; and setting operation monitoring items and management monitoring items corresponding to each potential equipment risk and potential management risk respectively according to the risk material data; and performing corresponding monitoring processing according to each obtained management monitoring item and operation monitoring item; The monitoring module is used to monitor each operation monitoring item and management monitoring item in real time to obtain corresponding operation monitoring data and management monitoring data; Establishing a monitoring model, inputting the operation monitoring data and the management monitoring data into the monitoring model for real-time display; The risk identification module is used to identify risks in the production process of lithium battery separators, and obtains operation monitoring data and management monitoring data corresponding to each operation monitoring item and management monitoring item according to the monitoring model; Conduct risk identification and analysis on the operation monitoring data and management monitoring data respectively to obtain corresponding operation risk identification results and management risk identification results; Input the obtained operational risk identification results and management risk identification results into the monitoring model for real-time display, and perform corresponding processing based on the displayed operational risk identification results and management risk identification results; Methods for classifying potential risks based on risk material data include: Establish an equipment risk judgment model, and the expression of the equipment risk judgment model is: In the formula: s is the input data, the input data is the risk material data; the output data is the device judgment value 1 or 0; Analyze the data of each risk material through the equipment risk judgment model to obtain the equipment judgment value corresponding to each potential risk; The potential risk with a device judgment value of 1 is marked as a potential device risk; The potential risk with a device judgment value of 0 is marked as a potential management risk; Methods for risk identification of operational monitoring data include: Acquire the operation instructions corresponding to the operation monitoring items in real time, and determine the operation standards according to the operation instructions, wherein the operation standards are composed of the unit standards corresponding to each unit item; Extracting features from the operation monitoring data according to the operation standard to obtain operation implementation data, wherein the operation implementation data is composed of various operation feature data, and the operation feature data corresponds to the corresponding unit standard; Compare the implemented operation data with the operation standard to determine the corresponding operation difference value; Generate a corresponding operation monitoring curve according to the monitoring time corresponding to the operation difference value, where the horizontal axis is the monitoring time and the vertical axis is the operation difference value; fit the operation monitoring curve to obtain an operation monitoring function, and mark the operation monitoring function as K(t), where t is the monitoring time; Set the conversion formula. The conversion formula is Where: ZW(t) is the conversion function, QA is the allowable value; The operation monitoring function is transformed by a transformation formula to obtain a transformation function; According to the formula Calculate the corresponding operation evaluation value; Where: PUR is the operational assessment value; Determine the operational risk identification results based on the operational assessment value.

2. According to the big data-based lithium battery separator production risk assessment system of claim 1, it is characterized in that: Methods for comparing implementation data to operational standards include: Subtract each operation characteristic data from the unit standard to obtain the characteristic difference value corresponding to each unit item, and integrate each characteristic difference value into a matching characteristic; Establishing a feature database, wherein the feature database is used to store feature value combinations, wherein the feature value combinations include combined feature values ​​and corresponding feature condition data; Input the matching features into the feature database for matching, and obtain the combined feature values ​​corresponding to each unit item; According to the formula Calculate the corresponding operation difference value; Wherein: CR is the operating difference value; j represents the corresponding unit item, j = 1, 2, ..., m, m is a positive integer; βj is the proportional coefficient of the corresponding unit item, 0<βj≤1; HTj is the combined eigenvalue of the corresponding unit item.

3. A lithium battery separator production risk assessment system based on big data according to claim 2, characterized in that: The method for establishing a feature database includes: Analyze the risk material data to obtain the unit material data corresponding to each unit item; define the benchmark item and fixed item; split the unit material data according to each fixed item to obtain the variable analysis data; and count the safety accident incidence rates corresponding to different characteristic differences based on the variable analysis data; According to the formula TR = e δ -1 calculates the corresponding first eigenvalue; where TR is the first eigenvalue; e is a natural constant; δ is the safety accident rate; Marking the fixed item data corresponding to the first eigenvalue as unit condition data; Merging the condition data of each unit according to each first eigenvalue to obtain each eigenvalue combination; A feature database is established based on the combination of each feature value.

4. A lithium battery separator production risk assessment system based on big data according to claim 3, characterized in that: The method for merging the condition data of each unit according to each first eigenvalue includes: Step SA1: merge the unit condition data with the same first eigenvalue to obtain an initial set; sort the first eigenvalues ​​in descending order to obtain a first sequence; Step SA2: marking the first eigenvalue ranked first in the first sequence as a reference eigenvalue; marking the first eigenvalue that is not a reference eigenvalue in the first sequence as an auxiliary eigenvalue; Step SA3: Calculate the difference between the reference eigenvalue and each subsidiary eigenvalue, mark it as the first difference, merge each subsidiary eigenvalue whose first difference is less than the threshold value X1 with the reference eigenvalue to obtain the corresponding merged eigenvalue; remove the reference eigenvalue and the subsidiary eigenvalue corresponding to the merged eigenvalue from the first sequence to obtain a new first sequence; merge each initial set and unit condition data corresponding to the merged eigenvalue to obtain a merged set; Step SA4: looping step SA2 to step SA3 until the first characteristic values ​​in the first sequence are merged, setting each matching condition according to the merged set, and integrating each matching condition into characteristic condition data.

5. A lithium battery separator production risk assessment system based on big data according to claim 4, characterized in that: In step SA3, the calculation method of the combined eigenvalue includes: The subsidiary eigenvalues ​​and the reference eigenvalues ​​are marked as Pi, i = 1, 2, ..., n, and n is a positive integer; wherein p1 represents the reference eigenvalue, and p2 and onwards are subsidiary eigenvalues; the conditional quantities corresponding to the subsidiary eigenvalues ​​and the reference eigenvalues ​​are identified; and the conditional quantities are marked as LTi; According to the formula Calculate the corresponding merged eigenvalue; Where: HT is the combined eigenvalue.

Images