A method and system for detecting the quality of finished protectors based on data analysis

Abstract

The present invention relates to the field of data processing, and particularly to a method and system for detecting the quality of finished protectors based on data analysis. The method includes: extracting a set number of protectors for detection and obtaining their short-circuit response times, and calculating the normal indexes of each short-circuit response time; setting multiple short-circuit current levels, sorting the normal indexes of the protectors at each current level according to the current level, and constructing a normal index sequence of the protectors; obtaining the mean value of the normal indexes of all historical qualified protectors at all set current levels, and constructing a reference normal index sequence; calculating the abnormal index of the protector based on the correlation between the normal index sequence of the protector and the reference normal index sequence; when the abnormal index of the protector is greater than a preset abnormal threshold, determining that the current sampling detection is abnormal. The present invention can effectively identify protectors with abnormal performance and avoid misjudgment caused by accidental errors or individual data fluctuations.

Description

Technical Field

[0001] The present invention relates to the field of data processing. More specifically, the present invention relates to a method and system for detecting the quality of finished protectors based on data analysis. Background Art

[0002] In an electrical system, a protector is an important safety device. Its core function is to detect abnormal conditions in the circuit (such as short circuit or overload), and quickly cut off the circuit when these abnormal conditions occur, so as to prevent the expansion of faults. For example, if a short circuit occurs in the circuit, the protector will immediately cut off the power supply to avoid safety accidents such as equipment damage, fire or electric shock to personnel.

[0003] The short - circuit response time refers to the time interval from the occurrence of a short circuit to the protector's action to cut off the circuit. The shorter this time is, the better the performance of the protector. A fast short - circuit response time can effectively reduce the damage of short - circuit faults to equipment, reduce the risk of safety accidents, and thus ensure the stable operation of the entire electrical system.

[0004] In the actual production environment, the electrical system will inevitably be affected by various current fluctuations, especially factors such as unstable grid voltage. These interference sources may cause the protector to fail to accurately cut off the circuit within the expected time, or produce faults that misjudge the circuit state. Such performance deviations will not only affect the normal operation of electrical equipment, but also may endanger the safety and stability of the entire power system. Currently, in the field of protector quality detection, traditional methods mainly adopt sampling detection. Specifically, this detection method randomly selects some samples from a batch of products for quality evaluation. However, since this detection method only detects some products, there is statistical probability uncertainty, that is, there is a risk that products with quality defects are not selected, resulting in an inability to accurately evaluate the quality of the protector, thus affecting equipment safety and system stability. Summary of the Invention

[0005] To solve the above - mentioned technical problem that the existing detection method cannot accurately evaluate the quality of the protector, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a method for detecting the quality of finished protectors based on data analysis includes:

[0007] Extracting a set number of protectors for detection and obtaining their short - circuit response times, and calculating the normal index of each short - circuit response time, where the normal index is obtained by comparing the short - circuit response time of a single protector with the overall average response time of all protectors;

[0008] Set multiple short - circuit current levels, sort the normal indexes of the protector at each current level according to the current level, and construct a normal index sequence of the protector; obtain the average value of the normal indexes of all historical qualified protectors at all set current levels, and construct a reference normal index sequence; calculate the abnormal index of the protector based on the correlation between the normal index sequence of the protector and the reference normal index sequence.

[0009] When the abnormal index of the protector is greater than the preset abnormal threshold, it is determined that the current sampling detection is abnormal.

[0010] By setting multiple short - circuit current levels, the present invention further refines the evaluation of the performance of the protector. Specifically, the normal indexes of the protector at each current level are sorted to construct a normal index sequence of the protector. At the same time, by obtaining the average value of the normal indexes of all historical qualified protectors at the same current level, a reference normal index sequence is constructed. Comparing the normal index sequence of the protector with the reference normal index sequence of historical qualified protectors can effectively identify protectors with abnormal performance and avoid misjudgment caused by accidental errors or individual data fluctuations; based on the correlation between the normal index sequence of the protector and the reference normal index sequence, the deviation degree between the performance of the current protector and the historical qualified level can be quantified, effectively identifying protectors with abnormal performance, providing a strong guarantee for the safe operation of the power system.

[0011] Preferably, the step of extracting a set number of protectors and obtaining their short - circuit response times further includes:

[0012] Obtain the production serial numbers corresponding to the extracted protectors.

[0013] Preferably, the normal index satisfies the relational expression:

[0014] ; where is the normal index of the short - circuit response time of the th protector, is the short - circuit response time of the first protector, is the short - circuit response time of the th protector, is the short - circuit response time of the th protector, represents taking the average value.

[0015] By calculating the ratio of the short - circuit response time of each protector to the average response time of all protectors, it is easier to compare the performance of different protectors. This standardization enables the intuitive comparison of the speed of different protectors relative to the whole even if the response time values of different protectors vary greatly through the normal index.

[0016] Preferably, the normal index further includes:

[0017] Calculate the mutation adjustment factor of the normal index corresponding to the protector, and take the product of the mutation adjustment factor and the normal index as the adjusted normal index of the protector.

[0018] Preferably, the adjustment factor satisfies the relational expression:

[0019] ; where is the mutation adjustment factor of the normal index corresponding to the th protector, is the short - circuit response time of the th protector, is the short - circuit response time of the th protector, is the short - circuit response time of the th protector, is the average number of normally produced protectors in the batch where the abnormal response time protector is located under the same current condition, is the production serial number corresponding to the th protector, is the production serial number corresponding to the th protector, is the production serial number corresponding to the th protector.

[0020] In the adjustment factor, and reflect the relative change in the short - circuit response time between adjacent protectors, and reflect the relative position of the protector in the production batch. Further calculate the product of the time change and the batch position change, and then obtain an adjustment factor, which can amplify the normal index difference of those protectors whose response time changes significantly in a short time or are located in batches that may have quality problems.

[0021] Preferably, the abnormal index satisfies the relational expression:

[0022] ; where is the abnormal index of the th protector, is the normal index sequence of the th protector, is the reference normal index sequence of the th protector, represents the Pearson correlation coefficient.

[0023] By calculating the Pearson correlation coefficient between the actual normal index sequence of the protector and the reference normal index sequence, to evaluate whether the working state of the protector is normal. If has a large absolute value (close to 1), it indicates that there is a significant linear difference between the normal index sequence of the protector and its reference normal index sequence, which may mean that there is some abnormality or fault in the protector; conversely, if is close to 0, it indicates that there is no obvious linear difference between the normal index sequence of the protector and the reference normal index sequence, and the protector may be in a normal state.

[0024] Preferably, determining that the current sampling detection is abnormal further includes:

[0025] Calculating the quality evaluation of the protector. When the quality evaluation of the protector is less than the preset quality threshold, determine that the protector is an abnormal protector; the quality evaluation satisfies the relational expression as:

[0026] ; In the formula, is the quality evaluation of the th protector, is the normal index sequence of the th protector, is the normal index sequence of the th qualified protector before the th protector, is the number of qualified protectors before the th protector, represents the cosine similarity.

[0027] By calculating the cosine similarity between the normal index sequence of the th protector and the normal index sequences of the previous N qualified protectors, and taking the average of these similarities, an index reflecting the similarity degree between the th protector and the group of qualified protectors can be obtained, that is , if has a high value, it indicates that the normal index sequence of the th protector is similar to the normal index sequences of multiple qualified protectors, so it is more likely to be normal; conversely, if has a low value, it indicates that there is a large difference between it and the qualified protectors, and there may be problems.

[0028] Preferably, the process of obtaining the normal index is:

[0029] Calculate the mean and standard deviation of the short - circuit response times of all the extracted protectors, and use the ratio of the difference between the short - circuit response time of each protector and the mean to the standard deviation as the normal index of the short - circuit response time of the corresponding protector.

[0030] Preferably, the quality evaluation further satisfies the relational expression:

[0031] ; where is the quality evaluation of the th protector, is the normal index sequence of the th protector, is the normal index sequence of the th qualified protector before the th protector, is the number of qualified protectors before the th protector, represents the cosine similarity, is the weight of the th qualified protector for the th protector.

[0032] Different qualified protectors may have different impacts on the quality evaluation of the th protector due to factors such as time, model, and usage environment. By introducing weights and cosine similarity, the contribution of different protectors to the evaluation result can be adjusted more precisely, making the evaluation result more accurate and persuasive.

[0033] In a second aspect, a quality detection system for finished protectors based on data analysis includes: a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, any of the above-mentioned quality detection methods for finished protectors based on data analysis is implemented.

[0034] The beneficial effects of the present invention are:

[0035] By calculating quantitative indicators such as the normal index, abnormal index, and quality evaluation of the short-circuit response time, the present invention transforms the quality detection from traditional qualitative analysis to quantitative analysis. This method not only reduces the interference of human factors but also improves the accuracy and consistency of the detection results; at the same time, the present invention combines data from multiple dimensions such as short-circuit response time, production serial number, and current level to comprehensively evaluate the quality of the protector, thus avoiding misjudgment that may be caused by a single indicator. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flowchart of the method of steps S1 - S3 in a quality detection method for finished protectors based on data analysis according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments.

[0038] The application scenario of the present invention is: to conduct quality inspection on the protectors participating in the sampling test.

[0039] Refer to Figure 1 , a method for detecting the quality of finished protectors based on data analysis includes steps S1 - S3, specifically as follows:

[0040] S1: Extract a set number of protectors for detection and obtain their short - circuit response times, and calculate the normal index of each short - circuit response time, where the normal index is obtained by comparing the short - circuit response time of a single protector with the overall average response time of all protectors.

[0041] In one embodiment, first, according to the requirements of experiments or tests, determine the number of protectors to be sampled and tested. Exemplarily, randomly select 10 protectors from 100 protectors for short - circuit protection tests, simulate short - circuit situations, and use test equipment such as multimeters to measure and record the response time of each protector in the short - circuit situation (the time it takes for the protector to take protective measures (such as disconnecting the circuit) from the start of detecting the fault after detecting a short - circuit fault).

[0042] Among them, it is also necessary to record the production serial numbers corresponding to the above 100 protectors.

[0043] Then, calculate the overall average response time of all protectors' short - circuit response times. For the short - circuit response time of each protector, calculate the ratio of its short - circuit response time to the overall average response time as the normal index of the short - circuit response time of the corresponding protector.

[0044] Exemplarily, the relationship satisfied by the above normal index is:

[0045]

[0046] In the formula, is the normal index of the short - circuit response time of the th protector, is the short - circuit response time of the first protector, is the short - circuit response time of the th protector, is the short - circuit response time of the th protector, represents taking the average value. Among them, = 10.

[0047] The above The higher the normal index of the short - circuit response time of a protector, the closer the short - circuit response time is to the overall average level, that is, the more stable the performance of the th protector; conversely, it indicates that its performance deviates greatly and the reasons need to be further analyzed.

[0048] In another embodiment, calculate the mean and standard deviation of the short - circuit response times of all the extracted protectors. Take the ratio of the difference between the short - circuit response time of each protector and the mean to the standard deviation as the normal index of the short - circuit response time of the corresponding protector, that is, the relational expression is satisfied as:

[0049]

[0050] In the formula, is the normal index of the short - circuit response time of the th protector, is the short - circuit response time of the 1st protector, is the th protector's short - circuit response time, is the th protector's short - circuit response time, is the standard deviation of the short - circuit response times of all protectors, represents taking the average value. Among them, = 10.

[0051] It should be added that over time, production equipment may experience wear, and the performance of raw materials may also change accordingly, resulting in changes in the short - circuit response time of protectors. Therefore, when evaluating the performance of protectors, the variable of production time also needs to be considered.

[0052] Specifically, if the difference in production serial numbers corresponding to two adjacent extracted protectors is very small, but the short - circuit response times have changed significantly, this may indicate some abnormality or inconsistency; on the contrary, if the difference in production serial numbers corresponding to two adjacent extracted protectors is very large, the change in the short - circuit response time may be more due to normal factors (such as equipment aging, material batch differences, etc.), and these changes are within expectations and should not be regarded as abnormal.

[0053] By considering the difference in production serial numbers and using it as a mutation adjustment factor, the normal index of the short - circuit response time of protectors can be evaluated more accurately.

[0054] In one embodiment, in historical data, find the mean value (obtained by checking the production serial numbers of protectors) of the number of protectors produced normally between all protectors that show abnormal short - circuit response times (i.e., do not meet the preset threshold) under the same current condition, and denote it as , A benchmark is provided to evaluate whether the performance of the protectors in the current batch meets the average level of normal production.

[0055] Further, taking the th protector among the extracted protectors as an example, the above mutation adjustment factor is calculated, that is, the relational expression is satisfied as:

[0056]

[0057] In the formula, is the mutation adjustment factor of the normal index corresponding to the th protector, is the short-circuit response time of the th protector, is the short-circuit response time of the th protector, is the short-circuit response time of the th protector, is the average number of protectors in normal production in the batch where the abnormal response time protector is located under the same current condition, is the production serial number corresponding to the th protector, is the production serial number corresponding to the th protector, is the production serial number corresponding to the th protector.

[0058] Among them, and respectively calculate the ratio of the change in production quantity between adjacent protectors to , and its function is to standardize the change in production quantity and avoid misjudgment of the change in normal short-circuit time response caused by too long production time interval; and respectively calculate the relative change in short-circuit response time between adjacent protectors.

[0059] The larger the above value, the more it means that the short-circuit response time of the th protector has an abnormal mutation, which may be caused by some problems in the production process or defects in the protector itself.

[0060] Further, the calculated above is introduced into the normal index corresponding to the th protector to obtain the adjusted normal index, that is, the relational expression is satisfied as:

[0061]

[0062] In the formula, is the normal index after adjustment for the th protector.

[0063] By introducing the mutation adjustment factor into the original normal index , it helps to more accurately evaluate the performance of the protector, especially when considering the dynamic changes in the production process and potential quality issues.

[0064] According to the series of operations on the normal index corresponding to the th protector above, the normal indices corresponding to all protectors in the sampling test can be obtained in the same way.

[0065] S2: Set multiple short - circuit current levels, sort the normal indices of the protector at each current level according to the current level to construct a normal index sequence of the protector; obtain the mean value of the normal indices of all historical qualified protectors at all set current levels to construct a reference normal index sequence; calculate the detection abnormal index of the protector based on the correlation between the normal index sequence of the protector and the reference normal index sequence.

[0066] Under the condition of consistent external environment, the greater the current passing through the protector, the shorter its short - circuit response time. That is to say, the current and the response time are negatively correlated. When there is interference in the surrounding electromagnetic environment, this normal relationship will be broken and the short - circuit response time of the protector will change abnormally. Therefore, in order to judge whether there is interference, the correlation between the current and the short - circuit response time of the detected protector and the normal protector can be analyzed to determine whether the detection result is reliable.

[0067] In one embodiment, when detecting the short - circuit response time of the protector, set multiple short - circuit current levels, and use the operation of calculating the normal index of the short - circuit response time of the protector in the above S1 to obtain the normal index corresponding to each protector at each short - circuit current level, and sort the normal indices of the protector at each short - circuit current level according to the current level to construct a normal index sequence of the protector.

[0068] Then, obtain the mean value of the normal indices of all historical qualified protectors at all set current levels to construct a reference normal index sequence.

[0069] Specifically, obtain the average value of the normal indices of all qualified protectors before the th protector at the same short - circuit current level to reflect the average performance state of all qualified protectors at the same short - circuit current level. Further, sort the average values of the normal indices of all qualified protectors before the th protector at all short - circuit current levels according to the current level to construct a reference normal index sequence.

[0070] Further, based on the correlation between the normal index sequence of the th protector and its reference normal index sequence, the abnormal index of the protector is calculated, that is, the relational expression is satisfied as:

[0071]

[0072] In the formula, is the abnormal index of the th protector, is the normal index sequence of the th protector, is the reference normal index sequence of the th protector, represents the Pearson correlation coefficient.

[0073] Among them, by comparing the correlation between the normal index vector of the th protector and the standard reference normal index sequences of all previous qualified protectors, the performance of the th protector is evaluated whether it is normal. If has a large value, it indicates that the performance of the th protector is consistent with the change trend of the standard reference normal index, that is, under different short-circuit current levels, the change ratio of the normal index relationship of its short-circuit response time is the same as that of the standard reference normal index. Therefore, it can be considered that the performance of this protector is normal. On the contrary, if has a small value, it indicates that the performance of the th protector is inconsistent with the change trend of the standard reference normal index, and there may be abnormalities.

[0074] Further, according to the above formula for calculating the abnormal index of the th protector, the abnormal indices of all protectors for sampling detection can be obtained in the same way.

[0075] S3: When the abnormal index of the protector is greater than the preset abnormal threshold, it is determined that the current sampling detection is abnormal.

[0076] In one embodiment, the abnormal threshold is set to 0.96. When the abnormal index of the protector is greater than the preset abnormal threshold, it is determined that the current sampling detection is abnormal.

[0077] In one embodiment, the above abnormal result is further confirmed, and the accuracy of the result is ensured through secondary detection. The normal index vector of the th protector is compared with the normal index vectors of known qualified protectors. Through comparison, the quality of the current protector can be evaluated whether it is qualified.

[0078] Specifically, calculate the quality evaluation of the protector. When the quality evaluation of the protector is less than the preset quality threshold, determine that the protector is an abnormal protector. Among them, through quality testing of protectors under different batches and production conditions, and collecting a large amount of data for a series of experiments and tests, a suitable quality threshold is obtained.

[0079] The above quality evaluation satisfies the following relationship:

[0080]

[0081] In the formula, is the quality evaluation of the th protector, is the normal index sequence of the th protector, is the normal index sequence of the th qualified protector before the th protector, is the number of qualified protectors before the th protector, represents the cosine similarity.

[0082] In another embodiment, the above quality evaluation satisfies the following relationship:

[0083]

[0084] In the formula, is the quality evaluation of the th protector, is the normal index sequence of the th protector, is the normal index sequence of the th qualified protector before the th protector, is the number of qualified protectors before the th protector, represents the cosine similarity, is the weight of the th qualified protector for the th protector; can be obtained by analyzing the performance of qualified protectors in historical data and the degree of association between them and subsequent protectors.

[0085] Furthermore, process the abnormal protector, such as performing fault analysis and debugging.

[0086] The system includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the method for detecting the finished product quality of the protector based on data analysis according to the first aspect of the present invention.

[0087] The system further includes a communication bus, a communication interface, and other components well-known to those skilled in the art. Their settings and functions are known in the art, so they will not be elaborated herein.

[0088] It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent shall be subject to the appended claims.

Claims

1. A quality inspection method for finished protectors based on data analysis, characterized in that, Including: Obtain the production serial number corresponding to the extracted protector; extract a set number of protectors for detection and obtain their short - circuit response times, calculate the normal index of each short - circuit response time, and the normal index is obtained by comparing the short - circuit response time of a single protector with the overall average response time of all protectors; the relationship satisfied by the normal index is: ; wherein, is the normal exponent of the short - circuit response time of the th protector, is the short - circuit response time of the first protector, is the short - circuit response time of the th protector, is the short - circuit response time of the th protector, represents taking the average value; The normal index further includes: calculating the mutation adjustment factor of the normal index corresponding to the protector, and taking the product of the mutation adjustment factor and the normal index as the adjusted normal index of the protector; the relationship satisfied by the mutation adjustment factor is: ; wherein, is the mutation adjustment factor of the normal index corresponding to the th protector, is the short - circuit response time of the th protector, is the short - circuit response time of the th protector, is the short - circuit response time of the th protector, is the average number of normally produced protectors in the batch where the abnormal response time protector is located under the same current condition, is the production serial number corresponding to the th protector, is the production serial number corresponding to the th protector, is the production serial number corresponding to the th protector; Set multiple short - circuit current levels, sort the normal indexes of the protector at each current level according to the current level, and construct the normal index sequence of the protector; obtain the mean value of the normal indexes of all historical qualified protectors at all set current levels, and construct the reference normal index sequence; calculate the abnormal index of the protector based on the correlation between the normal index sequence of the protector and the reference normal index sequence; When the abnormal index of the protector is greater than the preset abnormal threshold, it is determined that the current sampling detection is abnormal.

2. The quality inspection method for the finished product of the protector based on data analysis according to claim 1, characterized in that The relationship satisfied by the abnormal index is: ; wherein, is the anomaly index of the th protector, is the normal index sequence of the th protector, is the reference normal index sequence of the th protector, represents the Pearson correlation coefficient.

3. The quality inspection method for the finished product of the protector based on data analysis according to claim 2, characterized in that, The determination that the current sampling detection is abnormal further includes: Calculate the quality evaluation of the protector. When the quality evaluation of the protector is less than the preset quality threshold, it is determined that the protector is an abnormal protector; the relationship satisfied by the quality evaluation is: ; where, is the quality evaluation of the th protector, is the normal index sequence of the th protector, is the normal index sequence of the th qualified protector before the th protector, is the number of qualified protectors before the th protector, represents the cosine similarity.

4. A quality inspection method for finished protectors based on data analysis according to claim 1, characterized in that, The process of obtaining the normal index is: Calculate the mean value and standard deviation of the short - circuit response times of all the extracted protectors, and take the ratio of the difference between the short - circuit response time of each protector and the mean value to the standard deviation as the normal index of the short - circuit response time of the corresponding protector.

5. The quality inspection method for the finished product of the protector based on data analysis according to claim 3, characterized in that, The quality evaluation also satisfies the relationship: ; where, is the quality evaluation of the th protector, is the normal index sequence of the th protector, is the normal index sequence of the th qualified protector before the th protector, is the number of qualified protectors before the th protector, represents the cosine similarity, is the weight of the th qualified protector for the th protector.

6. A finished product quality inspection system for a protector based on data analysis, characterized in that, Including: A processor and a memory, where the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the quality detection method for the finished product of the protector based on data analysis according to any one of claims 1 - 5 is implemented.

Images