Intelligent control system for automatic production line of electric bicycle

Abstract

The invention relates to the technical field of intelligent control, and particularly discloses an electric bicycle automatic production line intelligent control system comprising a data acquisition module used for acquiring detection data of stamping finished products; the data processing module is used for calculating a quality coefficient; the quality evaluation module is used for evaluating the quality of the stamped finished product, judging whether the quality of the stamped finished product is qualified or not and generating a quality evaluation signal; the relevance evaluation module is used for analyzing whether the quality of the stamped finished product is related to the stamping pressure or not and generating a relevance evaluation signal; the regulation and control module is used for regulating and controlling the stamping pressure in the stamping process based on the associated signals; according to the method, the quality of the stamped finished product is analyzed in multiple dimensions, so that the scientificity and the accuracy of quality evaluation are improved, and powerful technical support is provided for quality control of the manufacturing industry; the regulation and control strategy is based on real-time data feedback and predictive analysis, fine control over the stamping process is achieved, and the product quality and the production efficiency are improved.

Description

Technical Field

[0001] The invention relates to the technical field of intelligent control, and in particular to an intelligent control system for an automatic production line of electric bicycles. Background Art

[0002] Electric bicycles are a green, environmentally friendly and convenient means of transportation. In the electric bicycle automatic production line, electric bicycle accessories such as frames and handlebars need to be stamped. Stamping is a forming processing method that relies on presses and molds to apply external forces to plates, strips, pipes and profiles to cause plastic deformation or separation, thereby obtaining workpieces of the desired shape and size.

[0003] However, in the prior art, during the stamping production process of electric bicycle accessories, the quality inspection of the stamped finished products does not conduct multi-dimensional inspection and analysis through the degree of thickness abnormality and the distribution of thickness abnormality, resulting in low accuracy of quality assessment, which in turn affects the regulation of the stamping pressure in the subsequent stamping process and makes it impossible to achieve refined control of the stamping process. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent control system for an electric bicycle automatic production line to solve the technical problems in the above background.

[0005] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides an intelligent control system for an electric bicycle automatic production line, comprising: Data acquisition module: detects the abnormal thickness degree and distribution of the stamped finished product and obtains the inspection data of the stamped finished product; Among them, the detection data includes: abnormal degree coefficient and abnormal distribution coefficient; Data processing module: calculate the quality coefficient based on the detection data; Quality assessment module: Based on the quality coefficient, the quality of the stamped finished product is assessed to determine whether the quality of the stamped finished product is qualified and generate a quality assessment signal; Among them, the quality assessment signal includes: a quality qualified signal and a quality unqualified signal; Correlation evaluation module: Based on the historical data of the stamping process, the quality coefficient of all stamping products and the pressure fluctuation coefficient of the stamping pressure are obtained, and the correlation coefficient is calculated based on the quality coefficient and the pressure fluctuation coefficient. If the correlation coefficient is ≤ the correlation coefficient threshold, a correlation signal is generated; Control module: Based on the relevant signals, the pressure fluctuation warning value is preset. At the same time, in the subsequent stamping process, the stamping pressure is tested to obtain the real-time pressure fluctuation coefficient; When the real-time pressure fluctuation coefficient in the stamping process is equal to the pressure fluctuation warning value, the quality coefficient of the stamping product obtained in this stamping process is predicted to obtain the quality prediction coefficient; When the quality prediction coefficient is greater than the quality coefficient threshold, a quality failure prediction signal is generated; Based on the quality failure prediction signal, the fluctuation range of the stamping pressure is extracted, and at the same time, the fluctuation limit range of the stamping pressure in the subsequent stamping process is calculated.

[0006] Beneficial effects of the present invention: (1) The present invention first obtains the thickness value of each position point of the stamped finished product through a thickness detection device, calculates the thickness difference ratio, and reveals the deviation degree of the local thickness of the stamped finished product; then, the region is divided based on the thickness difference ratio, and the thickness characterization value of each region is calculated; further, the overall abnormality degree and abnormal distribution of the finished product are quantified in multiple dimensions through two indicators: the abnormality degree coefficient and the abnormality distribution coefficient; the abnormality degree coefficient combines the regional thickness characterization value and the regional weight, reflecting the comprehensive abnormality level of the finished product in terms of thickness; and the abnormality distribution coefficient reveals the distribution characteristics of the abnormality on the surface of the finished product by comparing and analyzing the quantitative relationship between the region and the thickness difference ratio sub-range; finally, the above multi-dimensional information is integrated by using a preset quality coefficient formula to obtain a single quality coefficient as a comprehensive evaluation indicator of the quality of the finished product; by comparing with the quality coefficient threshold, the quality of the finished product is quickly determined to be qualified, thereby achieving efficient and accurate quality evaluation; the present invention analyzes the quality of the stamped finished product in multiple dimensions, which not only improves the scientificity and accuracy of quality evaluation, but also provides strong technical support for quality control in the manufacturing industry; (2) The present invention collects the quality coefficient of each stamped product and its corresponding stamping pressure data; then, calculates the pressure fluctuation coefficient of each stamping process to quantify the stability of the stamping pressure; then, compares the quality coefficient with the pressure fluctuation coefficient, and evaluates the correlation between the two by calculating the mass-pressure ratio and its variance (correlation coefficient); if the correlation coefficient is lower than the preset threshold, it indicates that the quality is closely related to the pressure fluctuation, and a correlation signal is issued; after confirming the correlation, the pressure control strategy is implemented in the subsequent stamping; a pressure fluctuation warning value is preset, the stamping pressure is monitored in real time and the real-time fluctuation coefficient is calculated; when the warning value is reached, the quality coefficient (quality prediction coefficient) of the current stamped product is predicted and compared with the quality coefficient threshold; if the predicted quality does not meet the standard, a quality failure prediction signal is generated, and the pressure fluctuation range in the stamping process is analyzed; in order to optimize the subsequent stamping, the limited range of pressure fluctuation is calculated according to the prediction results to limit the pressure fluctuation range in the future stamping process, thereby ensuring the quality pass rate of the stamped product; the control strategy of the present invention is based on real-time data feedback and predictive analysis to achieve refined control of the stamping process and improve product quality and production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The present invention will be further described below in conjunction with the accompanying drawings.

[0008] Figure 1 It is a flowchart of the present invention; Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0009] 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.

[0010] Embodiment 1: See also Figure 1 , Figure 2 As shown, an intelligent control system for an electric bicycle automatic production line according to an embodiment of the present invention comprises: Data acquisition module: to detect stamping products and obtain the inspection data of stamping products; Among them, the detection data includes: abnormal degree coefficient and abnormal distribution coefficient; In some implementation schemes, the thickness of the stamped product is detected to obtain the thickness difference ratio of each position point of the stamped product; It should be explained that the position points are set on the surface of the stamped product, and all the position points cover the entire surface of the stamped product; Based on the thickness difference ratio of each position point, the stamped product is divided into regions to obtain multiple analysis regions, and the thickness characterization value of each analysis region is calculated; Based on the thickness characterization values ​​of each analysis area, the anomaly degree coefficient and the anomaly distribution coefficient are calculated respectively; First, specifically, the process of obtaining the thickness difference ratio is: Based on a single position point, a thickness detection device is used to obtain a thickness value of the position point, and a difference calculation is performed between the thickness value and a preset thickness of the position point, and an absolute value is taken to obtain a thickness difference value, and then a ratio calculation is performed between the thickness difference value and the preset thickness to obtain a thickness difference ratio; It should be explained that thickness detection equipment includes but is not limited to: ultrasonic thickness gauge, laser thickness gauge; The preset thickness is: the thickness value preset by technicians in this field for the corresponding position of the stamped product based on the stamping production process. This value is an empirical value. It needs to be further explained that the thickness difference ratio reflects the degree to which the thickness of the position point deviates from the preset thickness; the smaller the value of the thickness difference ratio, the closer the thickness of the position point is to the preset thickness and the smaller the deviation; the larger the value of the thickness difference ratio, the greater the deviation of the thickness of the position point from the preset thickness; Second, specifically, based on the thickness difference ratio of each position point, the stamped product is divided into regions, and the process of obtaining multiple analysis regions is as follows: Based on the thickness difference ratio of all position points, the maximum and minimum values ​​of the thickness difference ratio are extracted. Mark them as the maximum thickness difference ratio and the minimum thickness difference ratio respectively, and obtain the total range of thickness difference ratios of all position points (the lower limit of the total range of thickness difference ratios is the minimum thickness difference ratio, and the upper limit of the total range of thickness difference ratios is the maximum thickness difference ratio); Divide the total range of thickness difference ratio evenly into a number of thickness difference ratio sub-ranges; Based on each thickness difference ratio sub-range and according to the thickness difference ratio of each position point, a plurality of analysis regions are obtained (the basis for dividing the analysis regions is: dividing the adjacent position points whose thickness difference ratio is in the same thickness difference ratio sub-range into the same analysis region); Thirdly, specifically, the process of obtaining the thickness characterization value is as follows: Based on a single analysis area, the thickness difference ratios of all position points in the analysis area are averaged to obtain a thickness characterization value; Fourthly, the process of obtaining the abnormality degree coefficient is as follows: Based on a single analysis area, the thickness characterization value of the analysis area is multiplied by the number of position points of the analysis area (the number of position points is the number of position points in the analysis area) to obtain the regional thickness abnormal value; At the same time, based on the thickness difference ratio sub-range of the analysis area, the number of analysis areas that are all in the thickness difference ratio sub-range is extracted to obtain the number of similar areas, and the ratio of the number of similar areas to the total number of analysis areas (the total number of analysis areas is the number of analysis areas divided by the stamping product) is calculated to obtain the area weight; The regional thickness anomaly value of the analysis area is multiplied by the regional weight to obtain the regional anomaly coefficient; Then the regional anomaly coefficients of all analysis areas are summed up to obtain the anomaly degree coefficient; It should be explained that the abnormality degree coefficient reflects the overall abnormality degree of the thickness of the stamped product; it comprehensively considers the thickness abnormality of each analysis area and the weight of these areas in the whole; if the value of the abnormality degree coefficient is high, it means that there are more thickness abnormal areas in the stamped product, or the abnormality of some abnormal areas is more serious, and the overall quality is poor; on the contrary, if the value of the abnormality degree coefficient is low, it means that the thickness distribution of the stamped product is more uniform and the overall quality is better; Fifth, specifically, the process of obtaining the abnormal distribution coefficient is: Extract the number of thickness difference ratio sub-ranges to obtain the total number of sub-ranges; The difference between the total number of the analysis area and the total number of sub-ranges is calculated to obtain the difference in the number of areas, and then the ratio of the difference in the number of areas to the total number of sub-ranges is calculated to obtain the abnormal distribution coefficient; It should be explained that the abnormal distribution coefficient reflects the difference between the number of thickness difference ratio sub-ranges and the number of analysis areas. If the total number of analysis areas is much larger than the total number of thickness difference ratio sub-ranges, it means that there is a lack of concentration in the division of the analysis areas, which indirectly reflects the degree of discreteness of the thickness anomaly of the stamped product. The larger the value, the more dispersed the thickness anomaly distribution is; the smaller the value, the relatively concentrated the thickness anomaly is. Data processing module: calculate the quality coefficient based on the detection data; In some embodiments, by the formula: , calculate the quality coefficient ZL, where a1 and a2 are preset proportional factors, both greater than 0, CD is the abnormality degree coefficient, and FB is the abnormality distribution coefficient; Quality assessment module: Based on the quality coefficient, the quality of the stamped finished product is assessed to determine whether the quality of the stamped finished product is qualified and generate a quality assessment signal; Among them, the quality assessment signal includes: a quality qualified signal and a quality unqualified signal; In some embodiments, a quality coefficient threshold is preset, and the quality coefficient is compared and analyzed with the quality coefficient threshold; If the quality coefficient is greater than the quality coefficient threshold, it means that the thickness of the stamped product is abnormally large or dispersed, and the quality of the stamped product is judged to be unqualified, that is, a quality unqualified signal is generated; If the quality coefficient is less than or equal to the quality coefficient threshold, it means that the thickness of the stamped product is abnormally small and concentrated, and the quality of the stamped product is determined to be qualified, that is, a quality qualified signal is generated; The technical solution of the embodiment of the present invention is mainly as follows: first, the thickness value of each position point of the stamped finished product is obtained by a thickness detection device, and the thickness difference ratio is calculated to reveal the degree of deviation of the local thickness of the stamped finished product; then, the region is divided based on the thickness difference ratio, and the thickness characterization value of each region is calculated; further, the overall abnormality degree and abnormal distribution of the finished product are quantified in multiple dimensions through the two indicators of abnormality degree coefficient and abnormality distribution coefficient; the abnormality degree coefficient combines the regional thickness characterization value and the regional weight, reflecting the comprehensive abnormality level of the finished product in terms of thickness; and the abnormality distribution coefficient reveals the distribution characteristics of the abnormality on the surface of the finished product by comparing and analyzing the quantitative relationship between the region and the thickness difference ratio sub-range; finally, the above-mentioned multi-dimensional information is fused by using a preset quality coefficient formula to obtain a single quality coefficient as a comprehensive evaluation indicator of the quality of the finished product; by comparing with the quality coefficient threshold, it is quickly determined whether the quality of the finished product is qualified, thereby realizing efficient and accurate quality evaluation; the embodiment of the present invention analyzes the quality of the stamped finished product in multiple dimensions, which not only improves the scientificity and accuracy of quality evaluation, but also provides strong technical support for quality control in the manufacturing industry.

[0011] Embodiment 2: Based on Example 1, please refer to Figure 1 , Figure 2 As shown, the intelligent control system for an electric bicycle automatic production line according to an embodiment of the present invention further includes: Relevance evaluation module: Based on the historical data of the stamping process, it analyzes whether the quality of the stamped product is related to the stamping pressure and generates a relevance evaluation signal; The correlation evaluation signal includes: a correlation signal and an uncorrelated signal; It should be explained that the historical data includes: a fixed number of stamped products and the quality coefficient of each stamped product and the stamping pressure of the stamping process; In some embodiments, based on historical data of the stamping process, the quality coefficients of all stamped products and the pressure fluctuation coefficient of the stamping pressure are obtained; By formula: , calculate the correlation coefficient GL, where Expressed as the quality coefficient of the i-th stamping product, Expressed as the average quality coefficient of all stamping products, Expressed as the pressure fluctuation coefficient of the i-th stamping product, It is expressed as the average value of the pressure fluctuation coefficient of all stamping products, and i is a positive integer; It should be explained that the correlation coefficient represents the absolute value of the Pearson correlation coefficient between the quality coefficient of the stamped product and the pressure fluctuation coefficient during the stamping process; the smaller the value of the correlation coefficient, the higher the correlation between the quality coefficient and the pressure fluctuation coefficient, that is, the higher the correlation between the quality of the stamped product and the fluctuation of the stamping pressure; Preset a correlation coefficient threshold, and compare and analyze the correlation coefficient with the correlation coefficient threshold; If the correlation coefficient is less than or equal to the correlation coefficient threshold, it means that the quality of the stamped product is highly correlated with the fluctuation of the stamping pressure, that is, a correlation signal is generated; If the correlation coefficient is greater than the correlation coefficient threshold, it means that the quality of the stamped product is weakly correlated with the fluctuation of the stamping pressure, that is, an uncorrelated signal is generated; Specifically, the process of obtaining the pressure fluctuation coefficient is: Based on a single stamped product, the real-time stamping pressure of the stamped product during the stamping process is obtained, and a pressure fluctuation curve is drawn based on a two-dimensional coordinate system with time as the X-axis and the real-time stamping pressure as the Y-axis; At the same time, with the stamping pressure value as the upper and lower limits of pressure, draw a straight line parallel to the X-axis, and mark the interval between the straight line as the normal pressure interval. Based on the pressure fluctuation curve and the normal pressure interval, extract the pressure fluctuation curve outside the normal pressure interval to obtain multiple abnormal fluctuation curve segments; The area of ​​the region enclosed by each abnormal fluctuation curve segment and the normal pressure interval is calculated respectively to obtain the abnormal fluctuation characterization value of each abnormal fluctuation curve segment, and then the abnormal fluctuation characterization values ​​of all abnormal fluctuation curve segments are summed up to obtain the total abnormal fluctuation characterization value, and the ratio of the total abnormal fluctuation characterization value to the abnormal fluctuation characterization threshold is calculated to obtain the pressure fluctuation coefficient; It should be explained that the ratio of the total value of abnormal fluctuation characterization to the abnormal fluctuation characterization threshold is calculated. The abnormal fluctuation characterization threshold here is to remove the dimension of the total value of abnormal fluctuation characterization, so as to obtain a dimensionless value, namely, the pressure fluctuation coefficient; It needs to be further explained that the pressure fluctuation coefficient reflects the degree to which the stamping pressure deviates from the normal range during the stamping process; the smaller the value of the pressure fluctuation coefficient, the smaller the fluctuation of the stamping pressure during the stamping process, and the better the stamping pressure stability; Control module: Based on the relevant signals, the stamping pressure in the stamping process is controlled to improve the pass rate of the stamping product quality in the subsequent stamping process; In some implementation schemes, based on the associated signal, a pressure fluctuation warning value is preset, and at the same time, in the subsequent stamping process, the stamping pressure is detected to obtain a real-time pressure fluctuation coefficient; When the real-time pressure fluctuation coefficient in the stamping process is equal to the pressure fluctuation warning value, the quality coefficient of the stamped product obtained in this stamping process is predicted to obtain the quality prediction coefficient, and the quality prediction coefficient is compared and analyzed with the quality coefficient threshold; If the quality prediction coefficient is greater than the quality coefficient threshold, it is predicted that the quality of the stamped product obtained in this stamping process is unqualified, that is, a quality unqualified prediction signal is generated; If the quality prediction coefficient is less than or equal to the quality coefficient threshold, it is predicted that the quality of the stamped product obtained in this stamping process is qualified, that is, a quality qualified prediction signal is generated; Based on the quality failure prediction signal, the fluctuation range of the stamping pressure is extracted. At the same time, the fluctuation limit range of the stamping pressure in the subsequent stamping process is calculated, so as to limit the fluctuation range of the stamping pressure and prevent the pressure fluctuation coefficient of the stamping process from being too large, thereby improving the qualified rate of the stamping product quality in the stamping process; First, specifically, the process of obtaining the pressure fluctuation warning value is as follows: Based on a single stamped product, the mass coefficient of the stamped product and its pressure fluctuation coefficient are calculated to obtain the mass-pressure ratio; All mass pressure ratios are averaged to obtain a mass pressure average ratio, and then the mass coefficient threshold is ratio-calculated to the mass pressure average ratio to obtain a pressure fluctuation coefficient threshold, and a pressure fluctuation warning value is set based on the pressure fluctuation coefficient threshold; It should be explained that the pressure fluctuation warning value is pre-set by those skilled in the art based on the pressure fluctuation coefficient threshold value. The pressure fluctuation warning value is an empirical value and satisfies that the pressure fluctuation warning value is less than the pressure fluctuation coefficient threshold value. Second, specifically, the process of obtaining the quality prediction coefficient is: When the real-time pressure fluctuation coefficient is equal to the pressure fluctuation warning value, the completion time of the stamping process at this time is extracted, and the completion time is multiplied by several prediction cycles; Extract the real-time pressure fluctuation coefficient at the end time point of the prediction period and the real-time pressure fluctuation coefficient at the start time point, and perform difference calculation to obtain the coefficient growth value of the real-time pressure fluctuation coefficient of the prediction period; Sort the coefficient growth values ​​of all forecast periods in ascending order, extract the third quartile value in the sorting, and mark it as the coefficient growth value of the forecast period; After calculating the ratio of the prediction coefficient growth value to the prediction cycle duration, multiply it by the non-stamping duration to obtain the prediction coefficient growth value, and sum the prediction coefficient growth value with the pressure fluctuation warning value to obtain the pressure fluctuation prediction coefficient; Then, the pressure fluctuation prediction coefficient is multiplied by the mass pressure ratio to obtain the mass prediction coefficient; Third, specifically, the process of obtaining the fluctuation range is as follows: Extract the pressure fluctuation curve during the stamping process, and obtain the peak mean and trough mean of the pressure fluctuation curve respectively; It should be explained that the peak mean is the average value of the real-time stamping pressures corresponding to all the peaks in the pressure fluctuation curve, and the trough mean is the average value of the real-time stamping pressures corresponding to all the troughs in the pressure fluctuation curve; Based on the peak mean and the trough mean, the fluctuation range is obtained, the upper limit of the fluctuation range is the peak mean, and the lower limit of the fluctuation range is the trough mean; Fourthly, the process of obtaining the fluctuation limit interval is as follows: The quality prediction coefficient is calculated by difference with the quality coefficient threshold to obtain the quality coefficient excess value. The excess value of the mass coefficient is calculated by ratio with the mass pressure to obtain the pressure fluctuation control value; The difference between the peak mean value and the pressure fluctuation control value is calculated to obtain the fluctuation limit upper limit value; The trough average value and the pressure fluctuation control value are summed to obtain the fluctuation limit lower limit value; Based on the fluctuation limit upper limit value and the fluctuation limit lower limit value, a fluctuation limit interval is obtained; The technical scheme of the embodiment of the present invention is mainly as follows: first, the quality coefficient of each stamped product and its corresponding stamping pressure data are collected; then, the pressure fluctuation coefficient of each stamping process is calculated to quantify the stability of the stamping pressure; then, the quality coefficient is compared with the pressure fluctuation coefficient, and the correlation between the two is evaluated by calculating the mass-pressure ratio and its variance (correlation coefficient); if the correlation coefficient is lower than the preset threshold, it indicates that the quality is closely related to the pressure fluctuation, and a correlation signal is issued; after confirming the correlation, the pressure control strategy is implemented in the subsequent stamping; a pressure fluctuation warning value is preset, the stamping pressure is monitored in real time and the real-time fluctuation coefficient is calculated; when the warning value is reached, the quality coefficient (quality prediction coefficient) of the current stamped product is predicted and compared with the quality coefficient threshold; if the predicted quality does not meet the standard, a quality unqualified prediction signal is generated, and the pressure fluctuation range in the stamping process is analyzed; in order to optimize the subsequent stamping, the limited range of pressure fluctuation is calculated according to the prediction results to limit the pressure fluctuation range in the future stamping process, thereby ensuring the quality pass rate of the stamped product; the control strategy of the embodiment of the present invention is based on real-time data feedback and predictive analysis to achieve refined control of the stamping process and improve product quality and production efficiency.

[0012] The size of the above threshold is set for the convenience of comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each group of sample data; for example: in the actual acquisition process, there are many groups of abnormal degree coefficients and abnormal distribution coefficients. Many groups of abnormal degree coefficients and abnormal distribution coefficients are processed to obtain the quality coefficients of the corresponding groups. The staff evaluates the quality of the stamped finished product based on so many groups of quality coefficients, thereby obtaining a corresponding relationship between a quality coefficient and the quality of the stamped finished product, and then derives and divides the threshold of the quality coefficient according to the quality of the stamped finished product, thereby obtaining the quality coefficient threshold, and compares the obtained abnormal coefficient with the quality coefficient threshold, that is, completes the identification of the quality of the stamped finished product corresponding to the quality coefficient; The above formulas are obtained by collecting a large amount of data and performing software simulation to select a formula close to the actual value. The factors in the formula are set by technicians in this field according to actual conditions; for example: Formula ; A technician in this field collects multiple groups of test data and sets a corresponding quality coefficient for each group of test data; Substitute the set quality coefficient and the collected test data into the formula, any two formulas constitute a binary linear equation system, and the calculated factors are screened and averaged to obtain the values ​​of a1 and a2, which are 2.65 and 2.23 respectively; The size of the factor is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the factor depends on the amount of test data and the preliminary setting of the corresponding quality coefficient for each set of test data by technical personnel in this field; as long as it does not affect the proportional relationship between the parameter and the quantified value, such as the quality coefficient is proportional to the value of the abnormality degree coefficient.

[0013] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An intelligent control system for an electric bicycle automatic production line, characterized in that: include: Data acquisition module: detects the abnormal thickness degree and distribution of the stamped finished product and obtains the inspection data of the stamped finished product; Among them, the detection data includes: abnormal degree coefficient and abnormal distribution coefficient; Data processing module: calculate the quality coefficient based on the detection data; Quality assessment module: Based on the quality coefficient, the quality of the stamped finished product is assessed to determine whether the quality of the stamped finished product is qualified and generate a quality assessment signal; Among them, the quality assessment signal includes: a quality qualified signal and a quality unqualified signal; Correlation evaluation module: Based on the historical data of the stamping process, the quality coefficient of all stamping products and the pressure fluctuation coefficient of the stamping pressure are obtained, and the correlation coefficient is calculated based on the quality coefficient and the pressure fluctuation coefficient. If the correlation coefficient is ≤ the correlation coefficient threshold, a correlation signal is generated; Control module: Based on the relevant signals, the pressure fluctuation warning value is preset. At the same time, in the subsequent stamping process, the stamping pressure is tested to obtain the real-time pressure fluctuation coefficient; When the real-time pressure fluctuation coefficient in the stamping process is equal to the pressure fluctuation warning value, the quality coefficient of the stamping product obtained in this stamping process is predicted to obtain the quality prediction coefficient; When the quality prediction coefficient is greater than the quality coefficient threshold, a quality failure prediction signal is generated; Based on the quality failure prediction signal, the fluctuation range of the stamping pressure is extracted, and at the same time, the fluctuation limit range of the stamping pressure in the subsequent stamping process is calculated.

2. According to claim 1, an intelligent control system for an electric bicycle automatic production line is characterized in that: The process of acquiring detection data is as follows: Detect the thickness of the stamped product and obtain the thickness difference ratio of each position point of the stamped product; Based on the thickness difference ratio of each position point, the stamped product is divided into regions to obtain multiple analysis regions, and the thickness characterization value of each analysis region is calculated; Based on the thickness characterization values ​​of each analysis area, the anomaly degree coefficient and the anomaly distribution coefficient are calculated respectively; Among them, the process of obtaining the thickness difference ratio is: Based on a single position point, a thickness detection device is used to obtain a thickness value of the position point, and a difference calculation is performed between the thickness value and a preset thickness of the position point, and an absolute value is taken to obtain a thickness difference value, and then a ratio calculation is performed between the thickness difference value and the preset thickness to obtain a thickness difference ratio; Based on the thickness difference ratio of each position point, the stamped product is divided into regions, and the process of obtaining multiple analysis regions is as follows: Based on the thickness difference ratios of all position points, the maximum and minimum values ​​of the thickness difference ratios are extracted, which are marked as the maximum thickness difference ratio and the minimum thickness difference ratio, respectively, to obtain the total range of the thickness difference ratios of all position points; Divide the total range of thickness difference ratio evenly into a number of thickness difference ratio sub-ranges; Based on each thickness difference ratio sub-range, multiple analysis areas are obtained according to the thickness difference ratio of each position point; The process of obtaining the thickness characterization value is as follows: Based on a single analysis area, the thickness difference ratios of all position points in the analysis area are averaged to obtain a thickness characterization value; The process of obtaining the abnormality degree coefficient is as follows: Based on a single analysis area, the thickness characterization value of the analysis area is multiplied by the number of position points of the analysis area to obtain the regional thickness abnormal value; At the same time, based on the thickness difference ratio sub-range of the analysis area, the number of analysis areas that are all in the thickness difference ratio sub-range is extracted to obtain the number of similar areas, and the ratio of the number of similar areas to the total number of analysis areas is calculated to obtain the area weight; The regional thickness anomaly value of the analysis area is multiplied by the regional weight to obtain the regional anomaly coefficient; Then the regional anomaly coefficients of all analysis areas are summed up to obtain the anomaly degree coefficient; The process of obtaining the abnormal distribution coefficient is: Extract the number of thickness difference ratio sub-ranges to obtain the total number of sub-ranges; The difference between the total number of analysis areas and the total number of sub-ranges was calculated to obtain the difference in the number of areas, and then the ratio of the difference in the number of areas to the total number of sub-ranges was calculated to obtain the abnormal distribution coefficient.

3. According to claim 1, an intelligent control system for an electric bicycle automatic production line is characterized in that: The process of obtaining the quality coefficient is: By formula: , calculate the quality coefficient ZL, where a1 and a2 are preset proportional factors, and both are greater than 0, CD is the abnormality degree coefficient, and FB is the abnormality distribution coefficient.

4. The intelligent control system for an electric bicycle automatic production line according to claim 1 is characterized in that: The process of generating the quality assessment signal is: Preset a quality coefficient threshold, and compare and analyze the quality coefficient with the quality coefficient threshold; If the quality coefficient > the quality coefficient threshold, a quality failure signal is generated; If the quality coefficient is less than or equal to the quality coefficient threshold, a quality qualified signal is generated.

5. The intelligent control system for an electric bicycle automatic production line according to claim 1 is characterized in that: The process of obtaining the pressure fluctuation coefficient is: Based on a single stamped product, the real-time stamping pressure of the stamped product during the stamping process is obtained, and a pressure fluctuation curve is drawn based on a two-dimensional coordinate system with time as the X-axis and the real-time stamping pressure as the Y-axis; At the same time, with the stamping pressure value as the upper and lower limits of pressure, draw a straight line parallel to the X-axis, and mark the interval between the straight line as the normal pressure interval. Based on the pressure fluctuation curve and the normal pressure interval, extract the pressure fluctuation curve outside the normal pressure interval to obtain multiple abnormal fluctuation curve segments; The area of ​​the region enclosed by each abnormal fluctuation curve segment and the normal pressure interval is calculated respectively to obtain the abnormal fluctuation characterization value of each abnormal fluctuation curve segment, and then the abnormal fluctuation characterization values ​​of all abnormal fluctuation curve segments are summed up to obtain the total abnormal fluctuation characterization value, and the ratio of the total abnormal fluctuation characterization value to the abnormal fluctuation characterization threshold is calculated to obtain the pressure fluctuation coefficient.

6. The intelligent control system for an electric bicycle automatic production line according to claim 1 is characterized in that: The process of obtaining the correlation coefficient is: By formula: , calculate the correlation coefficient GL, where Expressed as the quality coefficient of the i-th stamping product, Expressed as the average quality coefficient of all stamping products, Expressed as the pressure fluctuation coefficient of the i-th stamping product, It is expressed as the average value of the pressure fluctuation coefficient of all stamping products, and i is a positive integer.

7. The intelligent control system for an electric bicycle automatic production line according to claim 1 is characterized in that: The process of obtaining the pressure fluctuation warning value is as follows: Based on a single stamped product, the mass coefficient of the stamped product and its pressure fluctuation coefficient are calculated to obtain the mass-pressure ratio; The average of all mass-pressure ratios is calculated to obtain the mass-pressure average ratio, and then the ratio of the mass coefficient threshold to the mass-pressure average ratio is calculated to obtain the pressure fluctuation coefficient threshold. The pressure fluctuation warning value is set based on the pressure fluctuation coefficient threshold.

8. The intelligent control system for an electric bicycle automatic production line according to claim 1 is characterized in that: The process of obtaining the quality prediction coefficient is: When the real-time pressure fluctuation coefficient is equal to the pressure fluctuation warning value, the completion time of the stamping process at this time is extracted, and the completion time is multiplied by several prediction cycles; Extract the real-time pressure fluctuation coefficient at the end time point of the prediction period and the real-time pressure fluctuation coefficient at the start time point, and perform difference calculation to obtain the coefficient growth value of the real-time pressure fluctuation coefficient of the prediction period; Sort the coefficient growth values ​​of all forecast periods in ascending order, extract the third quartile value in the sorting, and mark it as the coefficient growth value of the forecast period; After calculating the ratio of the prediction coefficient growth value to the prediction cycle duration, multiply it by the non-stamping duration to obtain the prediction coefficient growth value, and sum the prediction coefficient growth value with the pressure fluctuation warning value to obtain the pressure fluctuation prediction coefficient; Then, the pressure fluctuation prediction coefficient is multiplied by the mass pressure ratio to obtain the mass prediction coefficient.

9. The intelligent control system for an electric bicycle automatic production line according to claim 1 is characterized in that: The process of obtaining the fluctuation range is as follows: Extract the pressure fluctuation curve during the stamping process, and obtain the peak mean and trough mean of the pressure fluctuation curve respectively; Based on the peak mean and the trough mean, the fluctuation range is obtained, the upper limit of the fluctuation range is the peak mean, and the lower limit of the fluctuation range is the trough mean.

10. The intelligent control system for an electric bicycle automatic production line according to claim 1, characterized in that: The process of obtaining the fluctuation limit range is as follows: The quality prediction coefficient is calculated by difference with the quality coefficient threshold to obtain the quality coefficient excess value, and the quality coefficient excess value is calculated by ratio with the quality pressure ratio to obtain the pressure fluctuation control value; The difference between the peak mean value and the pressure fluctuation control value is calculated to obtain the fluctuation limit upper limit value; The trough average value and the pressure fluctuation control value are summed to obtain the fluctuation limit lower limit value; Based on the fluctuation limit upper limit value and the fluctuation limit lower limit value, a fluctuation limit interval is obtained.

Images