Linear DC power supply quality detection method and system

Abstract

The invention discloses a linear DC power supply quality detection method and system, and relates to the technical field of quality detection and evaluation, and the method comprises the following steps: collecting quality detection historical data related to the manufacturing of a linear DC power supply, carrying out the preprocessing of the data, and integrating the data in a data warehouse; according to the method, the historical quality detection data are learned and trained, the mapping relation between the quality of the linear direct-current power supply and each detection index is established to construct the quality evaluation model, and the key quality index data of the linear direct-current power supply are collected in real time in the actual production process; the method comprises the following steps: training a quality evaluation model, inputting into the trained quality evaluation model for prediction, outputting a quality detection evaluation coefficient, judging the quality grade of a power supply, and generating a detailed detection analysis report, so that related personnel can timely discover problems in a production process and take corresponding measures for improvement, thereby improving the overall quality of a product.

Description

Technical Field

[0001] The present invention relates to the technical field of linear direct current power supply detection, and in particular to a linear direct current power supply quality detection method and system. Background Art

[0002] A linear DC power supply refers to a DC regulated power supply in which the adjustment tube works in a linear state. The AC power is first reduced in voltage amplitude by a transformer, and then rectified by a rectifier circuit to obtain pulsed DC power, which is then filtered to obtain a DC voltage with a small ripple voltage. The working process of the main circuit of the linear DC power supply is that the input power is first subjected to preliminary AC stabilization by a pre-stabilization circuit, and then converted into a DC power supply through isolation rectification by the main working transformer, and then the linear adjustment element is finely adjusted under the intelligent control of the control circuit and the single-chip microprocessor controller to output a high-precision DC voltage source.

[0003] With the rapid development of electronic technology, various electronic devices and systems have increasingly higher requirements for power supply performance. At the same time, linear DC power supplies are an indispensable component of electronic devices and systems, and they have unique advantages such as low noise, low ripple, and high stability. These advantages make them irreplaceable in specific application scenarios. At the same time, since the stability, ripple, efficiency and other characteristics of linear DC power supplies directly affect the overall performance and reliability of the equipment, the quality of linear DC power supplies is very important. Therefore, in the manufacturing process of linear DC power supplies, strict quality testing and evaluation of linear DC power supplies are the key to ensuring stable equipment performance and improving product competitiveness.

[0004] In the prior art, the quality detection of linear DC power supplies involves many test parameters, but the incomplete coverage of the test parameters will affect the comprehensiveness of the evaluation results; therefore, how to comprehensively analyze the various test parameters of the linear DC power supply quality detection and improve the comprehensiveness and reliability of the linear DC power supply detection is a problem that technical personnel need to solve. Summary of the invention

[0005] The present invention aims to provide a linear DC power quality detection method and system to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides a linear DC power quality detection method in a first aspect, comprising the following steps:

[0007] Step 1: Collect historical quality inspection data related to linear DC power supply manufacturing, pre-process the data, and integrate it in the data warehouse;

[0008] Step 2, perform feature analysis on the preprocessed data to identify key detection and evaluation indicators. The detection and evaluation indicators are divided into electrical performance indicators, quality indicators, and heat dissipation performance indicators to obtain an evaluation indicator data set;

[0009] Step 3, based on the evaluation index data set, extract relevant data of the electrical performance index, quality index and heat dissipation performance index respectively, and analyze and obtain the electrical performance evaluation index, quality evaluation index and heat dissipation performance evaluation index according to the extracted data;

[0010] Step 4, using the pre-processed quality inspection historical data to perform model training, construct a quality evaluation model, and combine the electrical performance evaluation index, quality evaluation index and heat dissipation performance evaluation index to comprehensively analyze and calculate the quality inspection evaluation coefficient;

[0011] Step 5: deploy the quality evaluation model and collect real-time quality inspection data of the linear DC power supply manufacturing, evaluate the quality of the linear DC power supply in combination with the quality inspection evaluation coefficient, determine its quality level and simultaneously output the inspection analysis report.

[0012] In one implementation method, the data is preprocessed, comprising the following steps:

[0013] Step 101, collecting quality inspection history data of linear DC power supply manufacturing from a database and a quality inspection management system, covering the entire quality inspection process of linear DC power supply manufacturing, and verifying the collected data;

[0014] Step 102, preprocessing the verified quality inspection history data, including data cleaning, data formatting and data standardization steps;

[0015] Step 103, integrating the pre-processed quality inspection historical data into a unified data set, building a data warehouse, and loading the pre-processed data into the data warehouse to ensure the correctness and integrity of the data;

[0016] Step 104, mark the pre-processed quality inspection historical data in the data warehouse, mark the abnormal quality data and normal quality data therein, and back up the data to prevent data loss or damage, formulate a data recovery plan to ensure that the data can be restored in time when the data is lost or damaged.

[0017] In one implementation method, the identification of key detection evaluation indicators includes the following steps:

[0018] Step 201, traverse the pre-processed quality inspection historical data, identify the key inspection areas that need to be paid attention to, and analyze the evaluation requirements of its quality inspection, which are electrical performance evaluation, quality evaluation and heat dissipation performance evaluation;

[0019] Step 202, combining the quality inspection and evaluation requirements of linear DC power supply manufacturing, performing feature analysis to identify key inspection and evaluation indicators, wherein the inspection and evaluation indicators are divided into electrical performance indicators, quality indicators, and heat dissipation performance indicators;

[0020] Step 203, analyzing the evaluation sub-indicators of the electrical performance index, which are input voltage, output voltage, output current and output voltage adjustment range;

[0021] Step 204, analyzing the evaluation sub-indicators of the quality indicator, which are stable voltage regulation rate, ripple voltage, equivalent output resistance and current regulation rate;

[0022] Step 205, analyzing the evaluation sub-indicators of the heat dissipation performance index, which are heat dissipation efficiency, temperature stability and thermal resistance;

[0023] Step 206 , sort out the identified detection evaluation indicators and their respective evaluation sub-indicators, and mark the associated data of each indicator to form an evaluation indicator data set.

[0024] In one implementation method, the analysis to obtain the electrical performance evaluation index, the quality evaluation index and the heat dissipation performance evaluation index comprises the following steps:

[0025] Step 301, filtering out data related to electrical performance indicators from the evaluation indicator data set, including evaluation sub-indicator data of input voltage, output voltage, output current, and output voltage adjustment range, presetting a reference value of each electrical performance evaluation sub-indicator, and analyzing the deviation function of each electrical performance evaluation sub-indicator, constructing a model to comprehensively analyze and calculate the electrical performance evaluation index, and evaluating the electrical performance test results of the linear DC power supply;

[0026] Step 302, filtering out data related to the quality index from the evaluation index data set, including evaluation sub-indicator data of stable voltage regulation rate, ripple voltage, equivalent output resistance, and current regulation rate, presetting a reference value of each quality evaluation sub-indicator, and analyzing the deviation function of each quality evaluation sub-indicator, building a model for comprehensive analysis and calculation of the quality evaluation index, and evaluating the quality test results of the linear DC power supply;

[0027] Step 303, filter out data related to the heat dissipation performance index from the evaluation index data set, including evaluation sub-indicator data of heat dissipation efficiency, temperature stability and thermal resistance, preset a baseline value for each heat dissipation performance evaluation sub-indicator, analyze the deviation function of each heat dissipation performance evaluation sub-indicator, build a model for comprehensive analysis and calculation of the heat dissipation performance evaluation index, and evaluate the heat dissipation performance test results of the linear DC power supply.

[0028] In one implementation method, the calculation formula of the electrical performance evaluation index is:

[0029]

[0030] Among them, EI is the electrical performance evaluation index, is the input voltage deviation function, is the output voltage deviation function, is the output current deviation function, is the output voltage regulation range deviation function, V in is the actual input voltage, V in，base is the reference input voltage, V out is the actual output voltage, V out，base is the reference output voltage, I out is the actual output current, I out，base is the reference output current, R out is the actual output voltage adjustment range, R out，base is the reference output voltage adjustment range, and the value range of EI is between 0 and 1;

[0031] The calculation formula of the quality evaluation index is:

[0032]

[0033] Among them, QI is the quality evaluation index, is the voltage regulation deviation function, is the ripple voltage deviation function, is the equivalent output resistance deviation function, is the current regulation rate deviation function, S v is the actual stable voltage regulation rate, S v，base is the reference stable voltage regulation rate, R v is the actual ripple voltage, R v，base is the reference ripple voltage, R eq is the actual equivalent output resistance, R eq，base is the reference equivalent output resistance, C a is the actual current regulation rate, C a，base is the reference current regulation rate, and the value range of QI is between 0 and 1;

[0034] The calculation formula of the heat dissipation performance evaluation index is:

[0035] TI=P TE ·P TS ·P TR ;

[0036]

[0037] Among them, TI is the heat dissipation performance evaluation index, P TE is the heat dissipation efficiency deviation function, P TSis the temperature stability deviation function, P TR is the thermal resistance deviation function, TE is the actual heat dissipation efficiency, TE base is the benchmark heat dissipation efficiency, TS is the actual temperature stability, TS base is the reference temperature stability, TR is the actual thermal resistance, TR base is the benchmark thermal resistance, and the value range of TI is between 0 and 1.

[0038] In one implementation method, the process of obtaining the quality detection evaluation coefficient includes the following steps:

[0039] Step 401, extracting the preprocessed quality inspection historical data, and obtaining the associated data of each inspection evaluation index in the evaluation index data set, training the neural network model, building a quality evaluation model, and testing the trained quality evaluation model to evaluate the model performance, so as to adjust the model parameters and optimize the model performance;

[0040] Step 402, based on the pre-processed quality inspection historical data, analyze and obtain the weight of each inspection evaluation index, and use the trained quality evaluation model to evaluate the quality inspection result of the linear DC power supply;

[0041] Step 403, according to the quality inspection result of the linear DC power supply, the electrical performance evaluation index, the quality evaluation index and the heat dissipation performance evaluation index are calculated, and the weight of each inspection evaluation index is matched to calculate the quality inspection evaluation coefficient;

[0042] Step 404, using the value of the quality detection evaluation coefficient and the evaluation requirements of the quality detection, different quality levels are set, namely excellent quality level, good quality level, medium quality level and poor quality level, and the quality evaluation threshold corresponding to each quality level is determined.

[0043] In one implementation method, the calculation formula of the quality detection evaluation coefficient is:

[0044]

[0045] Among them, C is the quality inspection evaluation coefficient, EI is the electrical performance evaluation index, QI is the quality evaluation index, TI is the heat dissipation performance evaluation index, and w 1 is the weight of the electrical performance evaluation index, w 2 is the weight of the quality evaluation index, w 3 is the weight of the heat dissipation performance evaluation index, and the value range of C is between 0 and 1.

[0046] In one implementation method, the plurality of quality levels correspond to a plurality of quality evaluation thresholds, wherein the quality evaluation thresholds include an upper threshold and a lower threshold;

[0047] The multiple quality levels and the multiple quality evaluation thresholds satisfy the following relationship:

[0048] Excellent quality grade C y ≤C≤1;

[0049] Good quality grade C l ≤C <C y ;

[0050] Medium quality grade C c ≤C <C l ;

[0051] Poor quality level 0≤C <C c ;

[0052] Among them, C is the quality inspection evaluation coefficient, C y is the lower threshold corresponding to the excellent quality level and the upper threshold corresponding to the good quality level, C l is the lower threshold corresponding to the good quality level and the upper threshold corresponding to the medium quality level, C c is the lower threshold corresponding to the medium quality level and the upper threshold corresponding to the poor quality level, C y =0.9, C l =0.8, C c =0.7.

[0053] In one implementation method, in step 5, the output process of the detection and analysis report is:

[0054] Step 501, deploying the trained quality evaluation model to the quality inspection management system, collecting quality inspection data in the linear DC power supply manufacturing process in real time, and using the quality evaluation model to evaluate the inspection quality of the linear DC power supply;

[0055] Step 502, based on the evaluation result of the quality evaluation model and in combination with the weight of each detection evaluation index, the quality detection evaluation coefficient is calculated, and the quality grade of the linear DC power supply is determined according to the value of the calculated quality detection evaluation coefficient;

[0056] Step 503, based on the real-time collected data and quality evaluation results, automatically generates a test analysis report, which includes the quality grade of each linear DC power supply, the value of each evaluation index and a detailed analysis of the quality evaluation, and outputs the generated test analysis report to the server of the quality inspection management system for relevant personnel to view and analyze.

[0057] In a second aspect, the present application provides a linear DC power supply manufacturing quality inspection and evaluation system for executing the linear DC power supply quality inspection method as described above, comprising a server, the server being communicatively connected to a data acquisition module, a feature analysis module, a quality index grading evaluation module, a quality evaluation model module, a quality grade determination module, and a test report generation module, wherein the modules are electrically signal-connected;

[0058] The data acquisition module is used to obtain key quality indicator data in the linear DC power supply manufacturing process, including electrical performance, quality, and heat dissipation performance indicators, and clean, format and standardize the collected data to ensure data quality;

[0059] This feature analysis module, based on the pre-processed quality index data, screens the key indicators for quality inspection and evaluation of linear DC power supply manufacturing, helping the model focus on the most relevant data and improving the accuracy and efficiency of the evaluation model;

[0060] The quality index grading evaluation module calculates the electrical performance evaluation index, quality evaluation index and heat dissipation performance evaluation index based on the pre-processed quality index data and the extracted key quality inspection and evaluation indicators, and provides quantitative indicators to evaluate the performance of the power supply in different aspects, providing a basis for the final quality evaluation;

[0061] The quality evaluation model module trains the quality evaluation model based on machine learning technology, performs quality detection and evaluation on the linear DC power supply, and outputs the quality detection and evaluation coefficient to determine the quality level of the linear DC power supply;

[0062] The quality grade determination module determines the quality grade of the linear DC power supply according to the quality detection evaluation coefficient and the preset quality evaluation threshold, thereby realizing automation and standardization of quality evaluation;

[0063] The test report generation module automatically generates a test analysis report based on the real-time collected data and quality evaluation results. The report content includes basic information of the power supply, test index data, quality evaluation coefficient and quality grade, etc., which is convenient for relevant personnel to view and analyze.

[0064] Due to the adoption of the above technical solution, the present invention has the following technical advances compared with the prior art:

[0065] 1. The present invention provides a linear DC power quality detection method and system. By learning and training historical quality detection data, a mapping relationship between the quality of the linear DC power supply and various detection indicators is established to construct a quality evaluation model. In the actual production process, the key quality indicator data of the linear DC power supply is collected in real time and input into the trained quality evaluation model for prediction, and the quality detection evaluation coefficient is output to judge the quality level of the power supply, and a detailed detection analysis report is generated, so that relevant personnel can timely discover problems in the production process and take corresponding measures to improve them, thereby improving the overall quality of the product.

[0066] 2. The present invention provides a linear DC power supply quality detection method and system, which monitors various performance indicators of the linear DC power supply during the manufacturing process, quickly analyzes and evaluates the data, and promptly feeds back the evaluation results, so that deviations in the production process can be discovered and corrected in a timely manner, thereby avoiding the production of unqualified products, reducing rework and scrap rates, and helping to prevent potential quality problems. Through continuous quality control, the stability and reliability of product quality are ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0068] Figure 1 is a flow chart of the method of the present invention;

[0069] Figure 2 A flow chart for obtaining the electrical performance evaluation index, the quality evaluation index and the heat dissipation performance evaluation index of the present invention;

[0070] Figure 3 This is a flow chart for obtaining the quality detection evaluation coefficient of the present invention;

[0071] Figure 4 It is a module structure diagram of the present invention. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0073] Embodiment 1, as Figure 1 , Figure 2 As shown, the present invention provides a linear DC power quality detection method, comprising the following steps:

[0074] Step 1: Collect historical quality inspection data related to linear DC power supply manufacturing, pre-process the data, and integrate it in the data warehouse. Collect historical quality inspection data of linear DC power supply manufacturing from the database and quality inspection management system, covering the entire quality inspection process of linear DC power supply manufacturing, and verify the collected data. Pre-process the verified historical quality inspection data, including data cleaning, data formatting and data standardization steps. Among them, data cleaning uses data deduplication technology to ensure that the data set does not contain duplicate records. For missing values, fill, delete or interpolate methods are used according to the specific situation to check and correct errors or outliers in the data. Formatting is used to unify data formats and ensure that all data follow the same format and standards to facilitate subsequent analysis and processing. Data standardization is used to standardize data and eliminate the impact of different dimensions on data analysis. The pre-processed quality inspection historical data is integrated into a unified data set and a data warehouse is built. The pre-processed data is loaded into the data warehouse to ensure the correctness and integrity of the data. The pre-processed quality inspection historical data is marked in the data warehouse, and abnormal quality data and normal quality data are marked. The data is backed up to prevent data loss or damage, and a data recovery plan is formulated to ensure that the data can be restored in time when it is lost or damaged.

[0075] Step 2, perform feature analysis on the preprocessed data to identify key detection and evaluation indicators. The detection and evaluation indicators are divided into electrical performance indicators, quality indicators and heat dissipation performance indicators, and obtain an evaluation indicator data set. Traverse the preprocessed quality detection historical data, clarify the key detection areas that need to be paid attention to, and analyze the evaluation requirements of its quality detection, which are electrical performance evaluation, quality evaluation and heat dissipation performance evaluation. Combined with the quality detection and evaluation requirements of linear DC power supply manufacturing, perform feature analysis to identify key detection and evaluation indicators. Among them, the detection and evaluation indicators are divided into electrical performance indicators, quality indicators and heat dissipation performance indicators. For the electrical performance indicators, analyze its evaluation sub-indicators, which are input voltage, output voltage, output current and output voltage adjustment range. The input voltage is the input voltage range of the linear DC power supply, which determines the upper and lower limits of the voltage input that the power supply can receive. The output voltage is the output voltage value that the power supply can provide under normal working conditions. The output current is the maximum current value that the power supply can provide at a specific output voltage, which reflects the load capacity of the power supply. The output voltage adjustment range is the range in which the power supply can keep the output voltage stable when the input voltage changes, reflecting the adaptability of the power supply to input voltage changes. For the quality indicators, analyze its evaluation sub-indicators, which are stability Voltage regulation, ripple voltage, equivalent output resistance and current regulation. Stable voltage regulation is the relative change in output voltage caused by the relative change in input voltage when the load current and ambient temperature remain unchanged. It measures the stability of the power supply when the input voltage changes. Ripple voltage is the AC voltage component superimposed on the output voltage. Its peak value is measured using an oscilloscope in millivolts. The size of the ripple voltage directly affects the stability and output quality of the power supply. The equivalent output resistance reflects the fluctuation of the output voltage when the load changes. The smaller the internal resistance, the better the stability of the power supply when the load changes. The current regulation The rate reflects the ability of the power supply to suppress output voltage fluctuations caused by changes in load current when the input voltage is constant. The smaller the current regulation rate, the better the stability of the power supply. For the heat dissipation performance index, its evaluation sub-indicators are analyzed, namely heat dissipation efficiency, temperature stability and thermal resistance. Heat dissipation efficiency is the ability of the power supply to transfer heat from the inside to the outside during operation. Temperature stability is the stability of the output voltage of the power supply when the temperature changes. Thermal resistance is the resistance of heat transfer inside the power supply. The identified detection evaluation indicators and their respective evaluation sub-indicators are sorted out, and the associated data of each indicator are marked to form an evaluation indicator data set;

[0076] Step 3, based on the evaluation index data set, extract the relevant data of the electrical performance index, quality index and heat dissipation performance index respectively, and analyze and obtain the electrical performance evaluation index, quality evaluation index and heat dissipation performance evaluation index according to the extracted data, and filter out the data related to the electrical performance index from the evaluation index data set, including the evaluation sub-indicator data of input voltage, output voltage, output current and output voltage adjustment range, preset the baseline value of each electrical performance evaluation sub-indicator, and analyze the deviation function of each electrical performance evaluation sub-indicator, build a model for comprehensive analysis and calculation of the electrical performance evaluation index, evaluate the electrical performance test results of the linear DC power supply, and filter out the data related to the quality index from the evaluation index data set , including evaluation sub-indicator data of stable voltage regulation rate, ripple voltage, equivalent output resistance and current regulation rate, presetting the baseline value of each quality evaluation sub-indicator, analyzing the deviation function of each quality evaluation sub-indicator, building a model for comprehensive analysis and calculation of the quality evaluation index, evaluating the quality test results of the linear DC power supply, screening out data related to the heat dissipation performance index from the evaluation indicator data set, including evaluation sub-indicator data of heat dissipation efficiency, temperature stability and thermal resistance, presetting the baseline value of each heat dissipation performance evaluation sub-indicator, analyzing the deviation function of each heat dissipation performance evaluation sub-indicator, building a model for comprehensive analysis and calculation of the heat dissipation performance evaluation index, evaluating the heat dissipation performance test results of the linear DC power supply;

[0077] Furthermore, the calculation formula of the electrical performance evaluation index is:

[0078]

[0079]

[0080] Among them, EI is the electrical performance evaluation index, is the input voltage deviation function, is the output voltage deviation function, is the output current deviation function, is the output voltage regulation range deviation function, V in is the actual input voltage, V in，base is the reference input voltage, V out is the actual output voltage, V out，base is the reference output voltage, I out is the actual output current, I out，base is the reference output current, R out is the actual output voltage adjustment range, R out，baseis the reference output voltage adjustment range, and the value range of EI is between 0 and 1. When the actual value is exactly the same as the reference value, each item in the bracket approaches 1, so EI approaches 1, indicating the best electrical performance. When the actual value differs more from the reference value, each item in the bracket approaches 0, so EI approaches 0, indicating poor electrical performance.

[0081] The calculation formula of the quality evaluation index is:

[0082]

[0083] Among them, QI is the quality evaluation index, is the voltage regulation deviation function, is the ripple voltage deviation function, is the equivalent output resistance deviation function, is the current regulation rate deviation function, S v is the actual stable voltage regulation rate, S v，base is the reference stable voltage regulation rate, R v is the actual ripple voltage, R v，base is the reference ripple voltage, R eq is the actual equivalent output resistance, R eq，base is the reference equivalent output resistance, C a is the actual current regulation rate, C a，base is the reference current regulation rate. The value range of QI is between 0 and 1. When the actual value is exactly the same as the reference value, each item in the bracket approaches 1, so QI approaches 1, indicating the best quality. When the actual value is more different from the reference value, each item in the bracket approaches 0, so QI approaches 0, indicating poor quality.

[0084] The calculation formula of the heat dissipation performance evaluation index is:

[0085] TI=P TE ·P TS ·P TR ;

[0086]

[0087] Among them, TI is the heat dissipation performance evaluation index, P TE is the heat dissipation efficiency deviation function, P TS is the temperature stability deviation function, P TR is the thermal resistance deviation function, TE is the actual heat dissipation efficiency, TE base is the benchmark heat dissipation efficiency, TS is the actual temperature stability, TR base is the reference temperature stability, TR is the actual thermal resistance, TR baseis the benchmark thermal resistance. The value range of TI is between 0 and 1. When the actual value is exactly the same as the benchmark value, each item in the bracket approaches 1, so TI approaches 1, indicating the best heat dissipation performance. When the actual value is more different from the benchmark value, each item in the bracket approaches 0, so TI approaches 0, indicating poor heat dissipation performance.

[0088] Step 4, using the pre-processed quality inspection historical data to perform model training, construct a quality evaluation model, and combine the electrical performance evaluation index, quality evaluation index and heat dissipation performance evaluation index to comprehensively analyze and calculate the quality inspection evaluation coefficient;

[0089] Step 5: deploy the quality evaluation model and collect real-time quality inspection data of the linear DC power supply manufacturing, evaluate the quality of the linear DC power supply in combination with the quality inspection evaluation coefficient, determine its quality level and simultaneously output the inspection analysis report.

[0090] Embodiment 2, as Figure 3 As shown, based on Example 1, the present invention provides a technical solution: Preferably, in step 4, the process of obtaining the quality detection evaluation coefficient is:

[0091] Extract the preprocessed quality inspection historical data, and obtain the associated data of each inspection and evaluation index in the evaluation index data set, train the neural network model, build a quality evaluation model, test the trained quality evaluation model, evaluate the model performance, so as to adjust the model parameters and optimize the model performance, analyze and obtain the weight of each inspection and evaluation index based on the preprocessed quality inspection historical data, and use the trained quality evaluation model to evaluate the quality inspection results of the linear DC power supply, and calculate the quality inspection evaluation coefficient according to the quality inspection results of the linear DC power supply, in combination with the calculation of the electrical performance evaluation index, the quality evaluation index and the heat dissipation performance evaluation index, and in conjunction with the weights of each inspection and evaluation index, calculate the quality inspection evaluation coefficient, and use the value of the quality inspection evaluation coefficient and the evaluation requirements of the quality inspection to set different quality levels, namely, excellent quality level, good quality level, medium quality level and poor quality level, and determine the quality evaluation threshold corresponding to each quality level;

[0092] Furthermore, the calculation formula of the quality inspection evaluation coefficient is:

[0093]

[0094] Among them, C is the quality inspection evaluation coefficient, EI is the electrical performance evaluation index, QI is the quality evaluation index, TI is the heat dissipation performance evaluation index, and w 1 is the weight of the electrical performance evaluation index, w 2 is the weight of the quality evaluation index, w 3is the weight of the heat dissipation performance evaluation index. The value range of C is between 0 and 1. When EI, QI and TI are all equal to 1, C reaches the maximum value of 1, indicating the best quality. When any of EI, QI or TI approaches 0, C approaches 0, indicating poor quality. By considering the minimum and maximum values ​​in the evaluation index, a nonlinear adjustment factor is introduced. The adjustment factor penalizes the low evaluation value (the difference from the maximum value), thereby increasing the sensitivity of the formula, so that the poor performance of any evaluation index will significantly affect the final C value.

[0095] Furthermore, the multiple quality levels correspond to multiple quality evaluation thresholds, wherein the quality evaluation threshold includes an upper threshold and a lower threshold;

[0096] Multiple quality levels and multiple quality evaluation thresholds satisfy the following relationship:

[0097] Excellent quality grade C y ≤C≤1; when the quality inspection evaluation coefficient reaches or exceeds C y When the linear DC power supply is rated at an excellent level, it means that all performance indicators of the power supply are close to or have reached the best state, and the performance is stable and meets high standards.

[0098] Good quality grade C l ≤C <C y ; When the quality inspection evaluation coefficient is C l to C y When the linear DC power supply is between 0 and 1, the quality of the linear DC power supply is considered to be at a good level, indicating that the performance of the power supply is good and most performance indicators have reached a satisfactory level;

[0099] Medium quality grade C c ≤C <C l ; When the quality inspection evaluation coefficient is C c to C l When the linear DC power supply is between , the quality is considered to be at a medium level, indicating that the performance of the power supply is average, and some performance indicators are not at the best state, but are still within an acceptable range;

[0100] Poor quality level 0≤C <C c ; When the quality inspection evaluation coefficient is lower than C c When the linear DC power supply is rated at a poor quality, it is considered that there are obvious problems with the performance of the power supply and further inspection is required, or it may not be suitable for use;

[0101] Among them, C is the quality inspection evaluation coefficient, C y is the lower threshold corresponding to the excellent quality level and the upper threshold corresponding to the good quality level, C l is the lower threshold corresponding to the good quality level and the upper threshold corresponding to the medium quality level, Cc is the lower threshold corresponding to the medium quality level and the upper threshold corresponding to the poor quality level, C y =0.9, C l =0.8, C c =0.7;

[0102] In step 5, the output process of the detection and analysis report is as follows:

[0103] The trained quality evaluation model is deployed to the quality inspection management system, and the quality inspection data of the linear DC power supply manufacturing process is collected in real time. The inspection quality of the linear DC power supply is evaluated using the quality evaluation model. Based on the evaluation results of the quality evaluation model and the weights of each inspection and evaluation index, the quality inspection evaluation coefficient is calculated, and the quality grade of the linear DC power supply is determined according to the value of the calculated quality inspection evaluation coefficient. According to the real-time collected data and quality evaluation results, an inspection analysis report is automatically generated, which includes the quality grade of each linear DC power supply, the value of each evaluation index and a detailed analysis of the quality evaluation. The generated inspection analysis report is output to the server of the quality inspection management system for relevant personnel to view and analyze.

[0104] Embodiment 3, as Figure 4 As shown, on the basis of Embodiment 1-2, the present invention further provides a quality detection and evaluation system for manufacturing a linear DC power supply, which is used to implement a linear DC power supply quality detection method, including a server, the server is communicatively connected with a data acquisition module, a feature analysis module, a quality index grading evaluation module, a quality evaluation model module, a quality grade determination module and a test report generation module, wherein the modules are connected by electrical signals;

[0105] The data acquisition module is used to obtain key quality indicator data in the linear DC power supply manufacturing process, including electrical performance, quality, and heat dissipation performance indicators, and clean, format, and standardize the collected data to ensure data quality;

[0106] The feature analysis module selects key indicators for quality inspection and evaluation of linear DC power supply manufacturing based on preprocessed quality indicator data, helping the model focus on the most relevant data and improving the accuracy and efficiency of the evaluation model;

[0107] The quality index grading evaluation module calculates the electrical performance evaluation index, quality evaluation index and heat dissipation performance evaluation index based on the pre-processed quality index data and the extracted key indicators of quality inspection and evaluation, and provides quantitative indicators to evaluate the performance of the power supply in different aspects, providing a basis for the final quality evaluation;

[0108] The quality evaluation model module trains the quality evaluation model based on machine learning technology, performs quality detection and evaluation on the linear DC power supply, and outputs the quality detection and evaluation coefficient to determine the quality level of the linear DC power supply;

[0109] The quality grade determination module determines the quality grade of the linear DC power supply according to the quality detection evaluation coefficient and the preset quality evaluation threshold, thereby realizing the automation and standardization of quality evaluation;

[0110] The test report generation module automatically generates a test analysis report based on the real-time collected data and quality evaluation results. The report content includes basic information of the power supply, test index data, quality evaluation coefficient and quality grade, etc., which is convenient for relevant personnel to view and analyze.

[0111] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A linear DC power quality detection method, characterized in that: The following steps are involved: Step 1, collect historical quality inspection data related to linear DC power supply manufacturing, pre-process the data, and integrate it in the data warehouse; Step 2, perform feature analysis on the preprocessed data to identify key detection and evaluation indicators. The detection and evaluation indicators are divided into electrical performance indicators, quality indicators, and heat dissipation performance indicators to obtain an evaluation indicator data set; Step 3, based on the evaluation index data set, extract relevant data of the electrical performance index, quality index and heat dissipation performance index respectively, and analyze and obtain the electrical performance evaluation index, quality evaluation index and heat dissipation performance evaluation index according to the extracted data; Step 4, using the pre-processed quality inspection historical data to perform model training, construct a quality evaluation model, and combine the electrical performance evaluation index, quality evaluation index and heat dissipation performance evaluation index to comprehensively analyze and calculate the quality inspection evaluation coefficient; Step 5: deploy the quality evaluation model and collect real-time quality inspection data of the linear DC power supply manufacturing, evaluate the quality of the linear DC power supply in combination with the quality inspection evaluation coefficient, determine its quality level and simultaneously output the inspection analysis report.

2. A linear DC power quality detection method according to claim 1, characterized in that: The data is preprocessed, comprising the following steps: Step 101, collecting quality inspection history data of linear DC power supply manufacturing from a database and a quality inspection management system, covering the entire quality inspection process of linear DC power supply manufacturing, and verifying the collected data; Step 102, preprocessing the verified quality inspection history data, including data cleaning, data formatting and data standardization steps; Step 103, integrating the preprocessed quality inspection historical data into a unified data set, building a data warehouse, and loading the preprocessed data into the data warehouse; Step 104: Mark the pre-processed quality inspection historical data in the data warehouse, mark the abnormal quality data and normal quality data therein, and back up the data.

3. A linear DC power quality detection method according to claim 1, characterized in that: The identification of key detection and evaluation indicators includes the following steps: Step 201, traverse the pre-processed quality inspection historical data, identify the key inspection areas that need to be paid attention to, and analyze the evaluation requirements of its quality inspection, which are electrical performance evaluation, quality evaluation and heat dissipation performance evaluation; Step 202, combining the quality inspection and evaluation requirements of linear DC power supply manufacturing, performing feature analysis to identify key inspection and evaluation indicators, wherein the inspection and evaluation indicators are divided into electrical performance indicators, quality indicators, and heat dissipation performance indicators; Step 203, analyzing the evaluation sub-indicators of the electrical performance index, which are input voltage, output voltage, output current and output voltage adjustment range; Step 204, analyzing the evaluation sub-indicators of the quality indicator, which are stable voltage regulation rate, ripple voltage, equivalent output resistance and current regulation rate; Step 205, analyzing the evaluation sub-indicators of the heat dissipation performance index, which are heat dissipation efficiency, temperature stability and thermal resistance; Step 206 , sort out the identified detection evaluation indicators and their respective evaluation sub-indicators, and mark the associated data of each indicator to form an evaluation indicator data set.

4. A linear DC power quality detection method according to claim 1, characterized in that: The analysis to obtain the electrical performance evaluation index, the quality evaluation index and the heat dissipation performance evaluation index comprises the following steps: Step 301, filtering out data related to electrical performance indicators from the evaluation indicator data set, including evaluation sub-indicator data of input voltage, output voltage, output current, and output voltage adjustment range, presetting a reference value of each electrical performance evaluation sub-indicator, and analyzing the deviation function of each electrical performance evaluation sub-indicator, constructing a model to comprehensively analyze and calculate the electrical performance evaluation index, and evaluating the electrical performance test results of the linear DC power supply; Step 302, filtering out data related to the quality index from the evaluation index data set, including evaluation sub-indicator data of stable voltage regulation rate, ripple voltage, equivalent output resistance, and current regulation rate, presetting a reference value of each quality evaluation sub-indicator, and analyzing the deviation function of each quality evaluation sub-indicator, building a model for comprehensive analysis and calculation of the quality evaluation index, and evaluating the quality test results of the linear DC power supply; Step 303, filter out data related to the heat dissipation performance index from the evaluation index data set, including evaluation sub-indicator data of heat dissipation efficiency, temperature stability and thermal resistance, preset a baseline value for each heat dissipation performance evaluation sub-indicator, analyze the deviation function of each heat dissipation performance evaluation sub-indicator, build a model for comprehensive analysis and calculation of the heat dissipation performance evaluation index, and evaluate the heat dissipation performance test results of the linear DC power supply.

5. A linear DC power quality detection method according to claim 4, characterized in that: The calculation formula of the electrical performance evaluation index is: Among them, EI is the electrical performance evaluation index, is the input voltage deviation function, is the output voltage deviation function, is the output current deviation function, is the output voltage regulation range deviation function, V in is the actual input voltage, V in，base is the reference input voltage, V out is the actual output voltage, V out，base is the reference output voltage, I out is the actual output current, I out，base is the reference output current, R out is the actual output voltage adjustment range, R out，base is the reference output voltage adjustment range, and the value range of EI is between 0 and 1; The calculation formula of the quality evaluation index is: Among them, QI is the quality evaluation index, is the voltage regulation deviation function, is the ripple voltage deviation function, is the equivalent output resistance deviation function, is the current regulation rate deviation function, S v is the actual stable voltage regulation rate, S v，base is the reference stable voltage regulation rate, R v is the actual ripple voltage, R v，base is the reference ripple voltage, R eq is the actual equivalent output resistance, R eq，base is the reference equivalent output resistance, C a is the actual current regulation rate, C a，base is the reference current regulation rate, and the value range of QI is between 0 and 1; The calculation formula of the heat dissipation performance evaluation index is: TI=P TE ·P TS ·P TR ; Among them, TI is the heat dissipation performance evaluation index, P TE is the heat dissipation efficiency deviation function, P TS is the temperature stability deviation function, P TR is the thermal resistance deviation function, TE is the actual heat dissipation efficiency, TE base is the benchmark heat dissipation efficiency, TS is the actual temperature stability, TS base is the reference temperature stability, TR is the actual thermal resistance, TR base is the benchmark thermal resistance, and the value range of TI is between 0 and 1.

6. A linear DC power quality detection method according to claim 1, characterized in that: In step 4, the process of obtaining the quality detection evaluation coefficient includes the following steps: Step 401, extracting the preprocessed quality inspection historical data, and obtaining the associated data of each inspection evaluation index in the evaluation index data set, training the neural network model, building a quality evaluation model, and testing the trained quality evaluation model to evaluate the model performance; Step 402, based on the pre-processed quality inspection historical data, analyze and obtain the weight of each inspection evaluation index, and use the trained quality evaluation model to evaluate the quality inspection result of the linear DC power supply; Step 403, according to the quality inspection result of the linear DC power supply, the electrical performance evaluation index, the quality evaluation index and the heat dissipation performance evaluation index are calculated, and the weight of each inspection evaluation index is matched to calculate the quality inspection evaluation coefficient; Step 404, using the value of the quality detection evaluation coefficient and the evaluation requirements of the quality detection, different quality levels are set, namely excellent quality level, good quality level, medium quality level and poor quality level, and the quality evaluation threshold corresponding to each quality level is determined.

7. A linear DC power quality detection method according to claim 6, characterized in that: The calculation formula of the quality detection evaluation coefficient is: Among them, C is the quality inspection evaluation coefficient, EI is the electrical performance evaluation index, QI is the quality evaluation index, TI is the heat dissipation performance evaluation index, w1 is the weight of the electrical performance evaluation index, w2 is the weight of the quality evaluation index, w3 is the weight of the heat dissipation performance evaluation index, and the value range of C is between 0 and 1.

8. A linear DC power quality detection method according to claim 7, characterized in that: The multiple quality levels correspond to the multiple quality evaluation thresholds, wherein the quality evaluation thresholds include an upper threshold and a lower threshold; The multiple quality levels and the multiple quality evaluation thresholds satisfy the following relationship: Excellent quality grade C y ≤C≤1; Good quality grade C l ≤C <C y ; Medium quality grade C c ≤C <C l ; Poor quality level 0≤C <C c ; Among them, C is the quality inspection evaluation coefficient, C y is the lower threshold corresponding to the excellent quality level and the upper threshold corresponding to the good quality level, C l is the lower threshold corresponding to the good quality level and the upper threshold corresponding to the medium quality level, C c is the lower threshold corresponding to the medium quality level and the upper threshold corresponding to the poor quality level, C y =0.9, C l =0.8, C c =0.

7.

9. A linear DC power quality detection method according to claim 1, characterized in that: In step 5, the output process of the detection and analysis report includes: Step 501, deploying the trained quality evaluation model to the quality inspection management system, collecting quality inspection data in the linear DC power supply manufacturing process in real time, and using the quality evaluation model to evaluate the inspection quality of the linear DC power supply; Step 502, based on the evaluation result of the quality evaluation model and in combination with the weight of each detection evaluation index, the quality detection evaluation coefficient is calculated, and the quality grade of the linear DC power supply is determined according to the value of the calculated quality detection evaluation coefficient; Step 503, automatically generating a test analysis report based on the real-time collected data and quality evaluation results, the report including the quality grade of each linear DC power supply, the value of each evaluation index and the analysis of the quality evaluation, and outputting the generated test analysis report to the server of the quality inspection management system for viewing and analysis.

10. A linear DC power supply manufacturing quality detection and evaluation system, used to execute the linear DC power supply quality detection method as described in any one of claims 1 to 9, comprising a server, characterized in that: The server is communicatively connected with a data acquisition module, a feature analysis module, a quality index grading evaluation module, a quality evaluation model module, a quality grade determination module and a test report generation module, wherein the modules are electrically connected with each other; The data acquisition module is used to obtain key quality indicator data in the linear DC power supply manufacturing process, including electrical performance, quality, and heat dissipation performance indicators, and clean, format and standardize the collected data; The feature analysis module selects key indicators for quality inspection and evaluation of linear DC power supply manufacturing based on the preprocessed quality indicator data; The quality index grading evaluation module calculates the electrical performance evaluation index, the quality evaluation index and the heat dissipation performance evaluation index according to the preprocessed quality index data and the extracted quality inspection and evaluation key indicators; The quality evaluation model module trains the quality evaluation model based on machine learning technology, performs quality detection and evaluation on the linear DC power supply, and outputs a quality detection and evaluation coefficient to determine the quality level of the linear DC power supply; The quality grade determination module determines the quality grade of the linear DC power supply according to the quality detection evaluation coefficient and the preset quality evaluation threshold; The test report generation module automatically generates a test analysis report based on the real-time collected data and quality evaluation results.

Images