Forging quality online detection method and system

Abstract

The invention provides a forge piece quality online detection method and system, and relates to the technical field of intelligent manufacturing, and the method comprises the steps: collecting online energy consumption data of a forging production line, and carrying out the feature extraction of the processing data, so as to obtain energy consumption feature data; defining a fuzzy set of energy consumption input variables and mass output variables, and fuzzifying the energy consumption input variables corresponding to the energy consumption input characteristic data set so as to convert the energy consumption input variables into fuzzy values; according to a fuzzy rule between the energy consumption input variable and the quality output variable, performing fuzzy reasoning by using the fuzzy incidence matrix, and determining an online fuzzy quality output value of the quality output result; deblurring the online fuzzy mass output value by using a gravity center method to obtain an in-shape mass output result of the forge piece, and establishing an energy consumption-mass database of the forge piece; time sequence analysis is carried out on the energy consumption-quality data, abnormal values are judged and marked, and the running state of the equipment is adjusted in time according to abnormal conditions. According to the invention, the forging work quality detection efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent manufacturing, and specifically to an online detection method and system for forging quality. Background Art

[0002] Forging is a high-volume, fast-paced production method. Due to factors such as materials, processes, equipment and operators, forgings such as automobile steering knuckles are prone to dimensional deviations, cracks, folds, pores and deformations, which seriously affect their mechanical properties and service life.

[0003] In order to ensure the processing quality of forgings, they need to be quality inspected. The commonly used quality inspection method in forging factories is to inspect the finished products manually. This inspection method requires a lot of manpower and time costs and is inefficient and has poor inspection accuracy. For large-scale, fast-paced forging production lines, it is impossible to manually inspect all finished products online, so only sampling inspection can be carried out. However, sampling inspection will result in the inability to ensure the stability of the finished product quality and is not conducive to the subsequent tracing of quality problems. Therefore, a method is needed to improve the efficiency of forging work quality inspection. Summary of the invention

[0004] The present application provides a forging quality online detection method and system, which can improve the efficiency of forging work quality detection.

[0005] In a first aspect of the present application, a method for online detection of forging quality is provided, the method comprising:

[0006] Collecting online energy consumption data of a forging production line, and preprocessing the online energy consumption data to obtain processed data;

[0007] Extracting features from the processed data to obtain energy consumption feature data;

[0008] The energy consumption characteristic data is converted into continuous data and normalized to obtain standard energy consumption characteristic data, and the energy consumption input characteristic data set is established using the standard energy consumption characteristic data;

[0009] Defining fuzzy sets of energy consumption input variables and quality output variables, and fuzzifying the energy consumption input variables corresponding to the energy consumption input feature data set to convert them into fuzzy values;

[0010] According to the fuzzy rules between the energy consumption input variable and the quality output variable, a fuzzy correlation matrix is ​​used to perform fuzzy reasoning to determine an online fuzzy quality output value of the quality output result;

[0011] Defuzzifying the online fuzzy quality output value by using the center of gravity method, obtaining the online forming quality output result of the forging, and establishing the energy consumption-quality database of the forging;

[0012] Conduct time series analysis on energy consumption-quality data, use time series change trends to determine and mark abnormal values, and adjust the operating status of equipment in a timely manner according to abnormal situations.

[0013] On the basis of the above technical solution, preferably, the energy consumption characteristic data includes product-level characteristic data, equipment-level characteristic data, and production line-level characteristic data, wherein the product-level characteristic data includes product quantity, heating temperature and time, solution temperature and time, aging temperature and time, material strain, and strain rate;

[0014] The feature extraction of the processed data to obtain energy consumption feature data specifically includes:

[0015] The infrared photoelectric sensor is used to count the products to obtain the product quantity data, and the heating temperature and time, the solution temperature and time, the aging temperature and time, the material strain and the strain rate are obtained through the product process parameter card, wherein the calculation formula of the product quantity is as follows:

[0016]

[0017] Among them, N(ΔT) is the number of products in time, M is the number of product channels of the production line, and N i (T 0 +ΔT) represents the i-th product channel at time T 0 The number of products within +ΔT, N i (T 0 ) represents the i-th product channel at time T 0 The number of products in

[0018] The total energy consumption of the equipment is calculated by collecting the equipment current, voltage and working time. The equipment-level characteristic data includes the working temperature and operating time of the heating furnace, the forging pressure and forging time of the forging press, and the operating time, current and voltage of each equipment. The total energy consumption of the equipment is calculated by the following formula:

[0019]

[0020] Among them, E e (ΔT) is the total energy consumption of the equipment, U ej (ΔT) is the jth phase current of the three-phase AC power of the equipment, I ej (ΔT) is the jth phase voltage of the three-phase AC power of the equipment;

[0021] The total energy consumption of the production line is calculated by collecting the running time, current and voltage of the production line. The characteristic data at the production line level includes the running time, current and voltage of the production line, which are calculated specifically by the following formula:

[0022]

[0023] Among them, E p (ΔT) is the total energy consumption of the production line, U pk (ΔT) is the kth phase current of the three-phase AC power of the production line, I pk (ΔT) is the k-th phase voltage of the three-phase AC power of the production line.

[0024] On the basis of the above technical solution, preferably, the energy consumption characteristic data is converted into continuous data and normalized to obtain standard energy consumption characteristic data, specifically including:

[0025] The discrete energy consumption characteristic data is converted into continuous energy consumption characteristic data by means of unique hot encoding, wherein the energy consumption characteristic data is divided into the continuous energy consumption characteristic data and the discrete energy consumption characteristic data, wherein the continuous energy consumption characteristic data includes current, voltage, temperature, forging pressure and material strain rate, and the discrete energy consumption characteristic data includes material strain and product quantity;

[0026] The energy consumption characteristic data is normalized to max-min, which is specifically expressed as follows:

[0027]

[0028] Among them, x i ' is the i-th standard energy consumption characteristic data, x i is the i-th energy consumption characteristic data, x max is the maximum value among the multiple energy consumption characteristic data, x min It is the minimum value among the multiple energy consumption characteristic data.

[0029] On the basis of the above technical solution, preferably, the fuzzy set defining the energy consumption input variable and the quality output variable, fuzzifying the energy consumption input variable corresponding to the energy consumption input feature data set to convert it into a fuzzy value, specifically includes:

[0030] Determine the types of the energy consumption input variable and the quality output variable, the energy consumption input variable is the data in the energy consumption input feature data set, and the quality output variable includes dimensional deviation, crack, fold, pore and deformation;

[0031] Determine the actual value ranges of the energy consumption input variable and the quality output variable, and divide the value ranges of the energy consumption input variable and the quality output variable into a plurality of segments, each value range corresponds to a fuzzy set, thereby forming fuzzy sets of the energy consumption input variable and the quality output variable;

[0032] According to the distribution of the energy consumption input variable and the quality output variable and domain knowledge, different membership functions are selected for each of the energy consumption input variable and the quality output variable, and the parameters of the membership function are determined. The expression of the membership function is as follows:

[0033]

[0034] Among them, γ(z) is the membership degree of the input variable z, h and σ are the parameters of the Gaussian membership function, and h is the mean parameter;

[0035] The membership degree of the energy consumption input variable in each of the fuzzy sets is calculated to obtain the fuzzy value of the energy consumption input variable.

[0036] On the basis of the above technical solution, preferably, the determining of the parameters of the membership function specifically includes:

[0037] The material strain rate, forging pressure, and temperature data use a trapezoidal membership function, and the expression is as follows:

[0038]

[0039] Wherein, a, b, c, d are parameters of the trapezoidal membership function, a and d are boundary points with a membership degree of 0, b and c are boundary points with a membership degree of 1, x is the value of the input variable, a(x) is the membership degree of the input variable x, and the graph of the trapezoidal membership function presents a trapezoidal shape;

[0040] The material strain and product quantity data adopt the S-shaped membership function, which is expressed as follows:

[0041]

[0042] Among them, f and h are the parameters of the S-shaped membership function, f is the steepness parameter, and g is the center position parameter.

[0043] On the basis of the above technical solution, preferably, the fuzzy rules between the energy consumption input variable and the quality output variable are used to perform fuzzy reasoning using a fuzzy association matrix to determine the online fuzzy quality output value of the quality output result, specifically including:

[0044] The energy consumption input variables are grouped into a plurality of clusters using a fuzzy self-organizing map algorithm, and the cluster data are analyzed to establish fuzzy rules between the energy consumption input variables and the quality output variables;

[0045] By constructing a fuzzy association matrix according to the fuzzy sets of the energy consumption input variables and the quality output variables;

[0046] The energy consumption input variable is subjected to fuzzy reasoning according to the fuzzy rules and the fuzzy association matrix to obtain the online fuzzy quality output value.

[0047] On the basis of the above technical solution, preferably, the center of gravity method is used to defuzzify the online fuzzy quality output value to obtain the online forming quality output result of the forging. The specific calculation formula is as follows:

[0048]

[0049] Among them, m COG is the quality output result, the interval (a, b) represents the value range of the quality output variable, and λ(m) is the membership function of the quality output variable;

[0050] The online energy consumption data and the online fuzzy quality output value are connected to form a corresponding relationship, and the energy consumption-quality database is established using the energy consumption-quality data.

[0051] On the basis of the above technical solutions, preferably, the energy consumption-quality data is analyzed in time series, and the abnormal values ​​are judged and marked by using the time series change trend, and the operation status of the equipment is adjusted in time according to the abnormal situation, specifically including:

[0052] Sorting the data in the energy consumption-quality database in chronological order to obtain time series data of energy consumption-quality data;

[0053] Analyzing the time series data using a linear regression model to obtain a time series variation trend of the energy consumption-quality data;

[0054] Determine and mark abnormal values ​​in the time series data according to the time series change trend;

[0055] Checking the corresponding product process parameters and equipment operating parameters under the abnormal state according to the abnormal value, and judging the operating condition of the equipment according to the product process parameters and the equipment operating parameters;

[0056] The operating conditions are fed back to a control system, and the operating state of the equipment is quickly adjusted through the control system.

[0057] In a second aspect of the present application, a forging quality online detection system is provided, the system comprising an acquisition module, a processing module and an output module, wherein:

[0058] The acquisition module is used to collect online energy consumption data of the forging production line, pre-process the online energy consumption data, and obtain processed data;

[0059] The processing module is used to extract features from the processed data to obtain energy consumption feature data;

[0060] The processing module is used to convert the energy consumption characteristic data into continuous data and normalize it to obtain standard energy consumption characteristic data, and use the standard energy consumption characteristic data to establish an energy consumption input characteristic data set;

[0061] The processing module is used to define a fuzzy set of energy consumption input variables and quality output variables, and fuzzify the energy consumption input variables corresponding to the energy consumption input feature data set to convert them into fuzzy values;

[0062] The processing module is used to determine the online fuzzy quality output value of the quality output result by performing fuzzy reasoning using a fuzzy association matrix according to the fuzzy rules between the energy consumption input variable and the quality output variable;

[0063] The processing module is used to defuzzify the online fuzzy quality output value by using the center of gravity method, obtain the online forming quality output result of the forging, and establish the energy consumption-quality database of the forging;

[0064] The output module is used to perform time series analysis on the energy consumption-quality data, determine and mark abnormal values ​​using the time series change trend, and adjust the operating status of the equipment in time according to abnormal situations.

[0065] On the basis of the above technical solution, preferably, the processing module is used to extract features from the processed data to obtain energy consumption feature data, specifically including:

[0066] The processing module is used to count the products using an infrared photoelectric sensor to obtain the product quantity data, and obtain the heating temperature and time, the solution temperature and time, the aging temperature and time, the material strain and the strain rate through the product process parameter card, wherein the calculation formula of the product quantity is as follows:

[0067]

[0068] Among them, N(ΔT) is the number of products in time, M is the number of product channels of the production line, and N i (T 0 +ΔT) represents the i-th product channel at time T 0 The number of products within +ΔT, N i (T 0 ) represents the i-th product channel at time T 0 The number of products in

[0069] The processing module is used to calculate the total energy consumption of the equipment through the collected equipment current, voltage and working time, wherein the equipment level characteristic data includes the working temperature and operating time of the heating furnace, the forging pressure and forging time of the forging press, and the operating time, current and voltage of each equipment. The total energy consumption of the equipment is calculated specifically by the following formula:

[0070]

[0071] Among them, E e (ΔT) is the total energy consumption of the equipment, U ej (ΔT) is the jth phase current of the three-phase AC power of the equipment, I ej (ΔT) is the jth phase voltage of the three-phase AC power of the equipment;

[0072] The processing module is used to calculate the total energy consumption of the production line through the collected production line operation time, current and voltage. The production line level characteristic data includes the production line operation time, current and voltage, which are specifically calculated by the following formula:

[0073]

[0074] Among them, E p (ΔT) is the total energy consumption of the production line, U pk (ΔT) is the kth phase current of the three-phase AC power of the production line, I pk (ΔT) is the k-th phase voltage of the three-phase AC power of the production line.

[0075] On the basis of the above technical solution, preferably, the processing module is used to convert the discrete energy consumption characteristic data into continuous energy consumption characteristic data by means of unique hot encoding, and the energy consumption characteristic data is divided into the continuous energy consumption characteristic data and the discrete energy consumption characteristic data, the continuous energy consumption characteristic data includes current, voltage, temperature, forging pressure and material strain rate, and the discrete energy consumption characteristic data includes material strain and product quantity;

[0076] The processing module is used to perform max-min normalization processing on the energy consumption characteristic data, which is specifically expressed as follows:

[0077]

[0078] Among them, x i ' is the i-th standard energy consumption characteristic data, x i is the i-th energy consumption characteristic data, x max is the maximum value among the multiple energy consumption characteristic data, x min It is the minimum value among the multiple energy consumption characteristic data.

[0079] On the basis of the above technical solution, preferably, the processing module is used to determine the types of the energy consumption input variable and the quality output variable, the energy consumption input variable is the data in the energy consumption input feature data set, and the quality output variable includes dimensional deviation, crack, fold, pore and deformation;

[0080] The processing module is used to determine the actual value ranges of the energy consumption input variable and the quality output variable, and divide the value ranges of the energy consumption input variable and the quality output variable into a plurality of segments, each value range corresponding to a fuzzy set, thereby forming the fuzzy sets of the energy consumption input variable and the quality output variable;

[0081] The processing module is used to select different membership functions for each of the energy consumption input variables and the quality output variables according to the distribution of the energy consumption input variables and the quality output variables and domain knowledge, and determine the parameters of the membership function. The expression of the membership function is as follows:

[0082]

[0083] Among them, γ(z) is the membership degree of the input variable z, h and σ are the parameters of the Gaussian membership function, and h is the mean parameter;

[0084] The processing module is used to calculate the membership degree of the energy consumption input variable in each of the fuzzy sets to obtain the fuzzy value of the energy consumption input variable.

[0085] On the basis of the above technical solution, preferably, the processing module is used to calculate the material strain rate, forging pressure and temperature data using a trapezoidal membership function, and the expression is as follows:

[0086]

[0087] Wherein, a, b, c, d are parameters of the trapezoidal membership function, a and d are boundary points with a membership degree of 0, b and c are boundary points with a membership degree of 1, x is the value of the input variable, a(x) is the membership degree of the input variable x, and the graph of the trapezoidal membership function presents a trapezoidal shape;

[0088] The processing module is used to adopt the S-shaped membership function material strain variable and product quantity data, and the expression is as follows:

[0089]

[0090] Among them, f and h are the parameters of the S-shaped membership function, f is the steepness parameter, and g is the center position parameter.

[0091] On the basis of the above technical solution, preferably, the energy consumption input variables are grouped into multiple clusters using a fuzzy self-organizing map algorithm, and the cluster data are analyzed to establish fuzzy rules between the energy consumption input variables and the quality output variables;

[0092] The processing module is used to construct a fuzzy association matrix according to the fuzzy sets of the energy consumption input variables and the quality output variables;

[0093] The processing module is used to perform fuzzy reasoning on the energy consumption input variable according to the fuzzy rules and the fuzzy association matrix to obtain the online fuzzy quality output value.

[0094] On the basis of the above technical solution, preferably, the processing module is used to use the centroid method to defuzzify the online fuzzy quality output value to obtain the online forming quality output result of the forging, and the specific calculation formula is as follows:

[0095]

[0096] Among them, m COG is the quality output result, the interval (a, b) represents the value range of the quality output variable, and λ(m) is the membership function of the quality output variable;

[0097] The processing module is used to establish a connection between the online energy consumption data and the online fuzzy quality output value to form a corresponding relationship, and to establish the energy consumption-quality database using the energy consumption-quality data.

[0098] On the basis of the above technical solution, preferably, the processing module is used to sort the data in the energy consumption-quality database in chronological order to obtain time series data of energy consumption-quality data;

[0099] The processing module is used to analyze the time series data using a linear regression model to obtain a time series variation trend of energy consumption-quality data;

[0100] The processing module is used to determine and mark abnormal values ​​in the time series data according to the time series change trend;

[0101] The processing module is used to check the corresponding product process parameters and equipment operation parameters under the abnormal state according to the abnormal value, and judge the operation status of the equipment according to the product process parameters and the equipment operation parameters;

[0102] The output module is used to feed back the operating status to the control system, and quickly adjust the operating status of the equipment through the control system.

[0103] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0104] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0105] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0106] 1. This application collects online energy consumption related data of the forging production line, uses fuzzy algorithm to establish a fuzzy relationship between forging energy consumption data and forming quality, and uses energy consumption data with a unified data structure and easy to obtain to characterize the forming quality of forgings. On the one hand, it reduces the complexity of calculating the forming quality model, making the model operation simpler and meeting the requirements of online detection of all finished products. On the other hand, it constructs a mapping relationship between forging energy consumption and forming quality, making the model output results more intuitive and meeting the requirements of online detection under large-scale and fast-beat production methods.

[0107] 2. This application uses a fuzzy algorithm to connect the input online energy consumption data with the output quality data to form a corresponding relationship, construct a forging energy consumption-quality database, and establish an online quality detection system with detectable quality, feedback on results, and traceability of problems through the mapping relationship of data in the forging energy consumption-quality database, which is conducive to the online monitoring of forming quality and subsequent tracing of quality problems.

[0108] 3. Based on the time series trend analysis of energy consumption-quality data, this application can timely discover operating problems of production line equipment and feedback them to the system. The system can quickly take corresponding measures to adjust the operating status of the equipment, thereby ensuring better forming quality of forgings, saving time and labor costs, and at the same time providing a theoretical basis for reducing energy consumption of forging production lines and optimizing process parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0109] Figure 1 It is a schematic diagram of a process of an online forging quality detection method disclosed in an embodiment of the present application;

[0110] Figure 2 It is a module schematic diagram of a forging quality online detection system disclosed in an embodiment of the present application;

[0111] Figure 3It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0112] Explanation of the reference numerals: 201, acquisition module; 202, processing module; 203, output module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0113] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0114] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0115] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0116] Forgings are prone to dimensional deviation, cracks, folds and other defects in large-scale, fast-paced production, which affect their mechanical properties and service life. Currently, forging plants mainly rely on manual inspection, but this method consumes a lot of manpower and time, has low inspection accuracy, and is difficult to achieve full inspection. Only sampling inspection can be performed, resulting in limited quality stability and problem traceability. Therefore, it is urgent to improve the efficiency and accuracy of forging quality inspection.

[0117] This embodiment discloses a method for online detection of forging quality. Figure 1 , including the following steps S110-S170:

[0118] S110, collecting online energy consumption data of the forging production line, preprocessing the online energy consumption data, and obtaining processed data.

[0119] The forging quality online detection method disclosed in the embodiment of the present application is applied to a server, and the server includes but is not limited to electronic devices such as mobile phones, tablet computers, wearable devices, PCs (Personal Computers), etc., and can also be a background server running a forging quality online detection method. The server can be implemented as an independent server or a server cluster composed of multiple servers.

[0120] The collected online energy consumption data of the forging production line include three levels: product, equipment, and production line: the energy consumption data collected at the product level include product quantity and product forging process parameters; the energy consumption data collected at the equipment level include equipment operating parameters such as equipment current, voltage, and working time; the energy consumption data collected at the production line level include parameters such as the voltage, current, and operating time of the production line. In the subsequent embodiments of this application, the forging is illustrated by taking the steering knuckle of an automobile as an example.

[0121] The online energy consumption data of the forging is preprocessed to obtain processed data. The preprocessing specifically includes the following steps:

[0122] Data cleaning: identifying missing values ​​in online energy consumption data, filling or deleting missing values, and removing duplicate values ​​and outliers in the data;

[0123] Data standardization, converting online energy consumption data into a unified format to ensure data consistency;

[0124] Data integration: integrating online energy consumption data from multiple data sources into a unified data set by connecting and merging;

[0125] Data denoising: using filtering to reduce noise in online energy consumption data and improve data quality.

[0126] S120, extracting features from the processed data to obtain energy consumption feature data.

[0127] Correspondingly, the energy consumption characteristic data include product-level characteristic data, equipment-level characteristic data, and production line-level characteristic data. Among them, the product-level characteristic data include product quantity, heating temperature and time, solution temperature and time, aging temperature and time, material strain, and strain rate.

[0128] Infrared photoelectric sensors are used to count the forgings such as automobile steering knuckles passing through the production line in real time, and the temperature and time information of the product such as heating, solution, aging, etc., as well as the material strain and strain rate are read from the process parameter management system. The temperature data of the heating furnace is obtained through the temperature sensor, and the forging pressure data of the forging press is obtained through the pressure sensor. The total energy consumption and energy consumption per unit time of the equipment are calculated and obtained through the collected equipment current, voltage and working time.

[0129] At fixed intervals (such as 1 hour), record the product counter value of each channel, and calculate the actual product quantity in the time period by comparing the cumulative product quantity changes within the time interval based on the difference method of cumulative counting according to the product quantity difference at different time points. The product quantity calculation formula is as follows:

[0130]

[0131] Among them, N(ΔT) is the number of products in time, M is the number of product channels of the production line, and N i (T 0 +ΔT) represents the i-th product channel at time T 0 The number of products within +ΔT, N i (T 0 ) represents the i-th product channel at time T 0 The number of products within.

[0132] In a multi-channel production line, the number of products depends not only on a single channel, but also on the synchronous operation of multiple channels. Therefore, it is necessary to accumulate the counting results of each channel to avoid missing and miscounting.

[0133] The data acquisition module (DAQ) is used to collect the three-phase voltage and current of the equipment every 1 second, and the power in each time period is integrated and calculated to obtain the cumulative energy consumption during the operation of the equipment. The data is stored in the industrial control server (PLC / SCADA) to form an energy consumption report. The total energy consumption of the equipment is then calculated based on the collected equipment current, voltage and working time. The equipment-level characteristic data includes the operating temperature and operating time of the heating furnace, the forging pressure and forging time of the forging press, and the operating time, current and voltage of each equipment. The total energy consumption of the equipment is calculated using the following formula:

[0134]

[0135] Among them, E e (ΔT) is the total energy consumption of the equipment, U ej (ΔT) is the jth phase current of the three-phase AC power of the equipment, I ej (ΔT) is the j-th phase voltage of the three-phase AC power of the equipment.

[0136] In a three-phase AC system, the total power is the sum of the instantaneous power of the three phases. The formula integrates the power of each phase over time to obtain the cumulative power consumption of the equipment in the time period ΔT.

[0137] Forging equipment such as heating furnaces and forging presses have obvious periodic load changes. Real-time monitoring of three-phase voltage and current can accurately reflect the energy consumption characteristics of equipment operation and support energy efficiency management and maintenance decisions.

[0138] Furthermore, the energy consumption per unit time of the equipment is calculated based on the total energy consumption of the equipment, which is specifically calculated using the following formula:

[0139]

[0140] Among them, E e (T) is the energy consumption per unit time of the equipment, E e (ΔT) is the total energy consumption of the equipment.

[0141] Smart meters are used to collect three-phase voltage and current at high frequency with a collection cycle of 1 second. The overall energy consumption is calculated in real time based on the production line operation time. The data is stored in the factory database for production optimization and cost control analysis. The total energy consumption of the production line is calculated by collecting the production line operation time, current and voltage. The characteristic data at the production line level includes the production line operation time, current and voltage, which are calculated using the following formula:

[0142]

[0143] Among them, E p (ΔT) is the total energy consumption of the production line, U pk (ΔT) is the kth phase current of the three-phase AC power of the production line, I pk (ΔT) is the k-th phase voltage of the three-phase AC power of the production line.

[0144] Similar to the equipment energy consumption formula, the total energy consumption formula for the production line is also based on the instantaneous power integration of three-phase electricity. The difference is that its scope of application is the entire production line. The production line is usually composed of multiple devices, and its total power is the accumulation of the power of each device.

[0145] Furthermore, the energy consumption per unit time of the production line is calculated based on the total energy consumption of the production line, which is specifically calculated using the following formula:

[0146]

[0147] Among them, E p (T) is the energy consumption per unit time of the production line, E p (ΔT) is the total energy consumption of the production line.

[0148] S130, converting the energy consumption characteristic data into continuous data and normalizing it to obtain standard energy consumption characteristic data, and using the standard energy consumption characteristic data to establish an energy consumption input characteristic data set.

[0149] Energy consumption characteristic data are divided into continuous energy consumption characteristic data and discrete energy consumption characteristic data. Continuous energy consumption characteristic data include current, voltage, temperature, forging pressure and material strain rate. These data have continuous values ​​and do not require special processing. Discrete energy consumption characteristic data include material strain and product quantity, which have discrete properties and need to be converted into continuous data for subsequent analysis and modeling.

[0150] One-hot encoding converts discrete data into multidimensional vectors, with each possible discrete value corresponding to an independent encoding bit. Example: Material strain variables have three categories: small, medium, and large. After one-hot encoding, they are represented as: small: [1,0,0], medium: [0,1,0], large: [0,0,1].

[0151] Determine the total number of categories of discrete features (such as different levels of material strain), and generate a one-hot encoded vector representation based on the actual discrete data category values ​​collected. Combine the encoding result with the continuous data to form a unified data matrix.

[0152] The energy consumption characteristic data is normalized by max-min to obtain the standard energy consumption characteristic data, which is specifically expressed as follows:

[0153]

[0154] Among them, x i ' is the i-th standard energy consumption characteristic data, x i is the i-th energy consumption characteristic data. The energy consumption characteristic data is composed of n data and is expressed as X=(x 1 ,x 2 ,…,x n ); x max is the maximum value among multiple energy consumption characteristic data, x min The minimum value among multiple energy consumption feature data. Standardize data of different scales to avoid feature data being over-magnified or ignored in model training, and improve the robustness and accuracy of the model.

[0155] Collect the original energy consumption characteristic data of the forging production line, construct the original data set, traverse the entire data set, and find the maximum value x of each characteristic max and the minimum value x min . Perform normalization calculations on each data point to obtain a standardized data matrix. Store the standardized data in a database for subsequent analysis, modeling, or production control. For example, assume that the following original energy consumption characteristic data is collected: current: I = [10, 15, 20, 25] (unit: A), voltage: U = [220, 230, 240, 250] (unit: V), number of products: [100, 200, 300] (discrete). The number of products (assuming possible values ​​are 100, 200, 300) is uniquely encoded: 100 gets [1, 0, 0], 200 gets [0, 1, 0], and 300 gets [0, 0, 1]. Then perform Max-Min normalization. Taking current as an example, the maximum value xmax = 25, the minimum value x min =10.

[0156] The normalized result of the current is:

[0157]

[0158]

[0159]

[0160]

[0161] S140, defining fuzzy sets of energy consumption input variables and quality output variables, and fuzzifying the energy consumption input variables corresponding to the energy consumption input feature data set to convert them into fuzzy values.

[0162] In a possible implementation, fuzzy sets of energy consumption input variables and quality output variables are defined, and the energy consumption input variables corresponding to the energy consumption input feature data set are fuzzified to be converted into fuzzy values, specifically including: determining the types of energy consumption input variables and quality output variables, the energy consumption input variables are data in the energy consumption input feature data set, and the quality output variables include dimensional deviation, cracks, folds, pores and deformation; determining the actual value ranges of the energy consumption input variables and the quality output variables, and dividing the value ranges of the energy consumption input variables and the quality output variables into several sections, each value range corresponds to a fuzzy set, thereby forming a fuzzy set of energy consumption input variables and quality output variables; according to the distribution of energy consumption input variables and quality output variables and domain knowledge, different membership functions are selected for each energy consumption input variable and quality output variable, and the parameters of the membership function are determined, and the membership degree of the energy consumption input variable in each fuzzy set is calculated to obtain the fuzzy value of the energy consumption input variable.

[0163] Specifically, determine the types of energy consumption input variables and quality output variables, and clarify which variables will be used in the definition of fuzzy sets. Energy consumption input variables are data in the energy consumption input feature data set, while quality output variables include quality indicators such as dimensional deviation, cracks, folds, pores, and deformation. First, it is necessary to collect and identify all energy consumption data and quality indicators related to forging quality. These data will serve as input and output variables of fuzzy sets.

[0164] Determine the actual value range of energy consumption input variables and quality output variables. The value range of these variables is determined to divide them into several segments, each of which corresponds to a fuzzy set. Through statistical analysis of historical data, the maximum and minimum values ​​of each variable are determined, and then the range of the variable is divided into several intervals based on these values, each of which corresponds to a fuzzy set.

[0165] Select different membership functions for energy consumption input variables and quality output variables, and determine the parameters. The membership function is used to describe the degree to which an element belongs to a fuzzy set. Different membership functions can better reflect the characteristics of different variables. According to the distribution of variables and domain knowledge, select the appropriate membership function. For example, for energy consumption input variables, you may choose a Gaussian membership function, and for quality output variables, you may choose a triangular or trapezoidal membership function. Then, determine the parameters of the membership function through data analysis or expert experience. The expression of the membership function is as follows:

[0166]

[0167] Among them, γ(z) is the membership of the input variable z, h and σ are the parameters of the Gaussian membership function, and h is the mean parameter. The Gaussian membership function is a commonly used membership function, and its shape is similar to the normal distribution curve. The characteristic of this function is that when the input variable z is close to the mean h, the membership γ(z) is close to 1; when z is far away from h, the membership decreases rapidly. The parameter σ determines the width of the curve, that is, the larger σ is, the wider the curve is, and the more gradual the change of the membership is; the smaller σ is, the narrower the curve is, and the steeper the change of the membership is.

[0168] The material strain rate, forging pressure, and temperature data use a trapezoidal membership function, and the expression is as follows:

[0169]

[0170] Among them, a, b, c, d are the parameters of the trapezoidal membership function, a and d are the boundary points with a membership degree of 0, b and c are the boundary points with a membership degree of 1, x is the value of the input variable, a(x) is the membership degree of the input variable x, and the graph of the trapezoidal membership function is trapezoidal in shape. The trapezoidal membership function is a membership function used in fuzzy logic, which defines the degree of membership of a variable value to a specific fuzzy set. This function has four parameters, which represent the four boundaries of the trapezoid. These parameters can be used to adjust the shape and position of the trapezoid to adapt to different fuzzy sets.

[0171] The material strain and product quantity data adopt the S-shaped membership function, which is expressed as follows:

[0172]

[0173] Among them, f and h are the parameters of the S-shaped membership function, f is the steepness parameter, which is used to describe the rate of change of the membership function, g is the center position parameter, which is used to control the center position of the membership function. When the input variable is equal to g, the membership is 0.5, β(y) is the membership of the input variable y, and the graph of the S-shaped membership function presents an S-shaped shape. When the input variable y increases from 0 to the maximum value, the membership increases linearly from 0 to 1.

[0174] Finally, the membership of the energy consumption input variable in each fuzzy set is calculated, and the actual energy consumption data is converted into fuzzy values ​​for fuzzy reasoning. Using the determined membership function and parameters, the membership of each energy consumption data point in each fuzzy set is calculated. This will generate a fuzzy value that represents the degree of membership of each data point in different fuzzy sets.

[0175] S150, according to the fuzzy rules between the energy consumption input variable and the quality output variable, using the fuzzy association matrix to perform fuzzy reasoning to determine the online fuzzy quality output value of the quality output result.

[0176] In a possible implementation, according to the fuzzy rules between the energy consumption input variables and the quality output variables, fuzzy reasoning is performed using a fuzzy association matrix to determine the online fuzzy quality output value of the quality output result, specifically including: using a fuzzy self-organizing mapping algorithm to group the energy consumption input variables into multiple clusters, analyzing the cluster data to establish fuzzy rules between the energy consumption input variables and the quality output variables; constructing a fuzzy association matrix based on fuzzy sets of energy consumption input variables and quality output variables; and performing fuzzy reasoning on the energy consumption input variables according to the fuzzy rules and the fuzzy association matrix to obtain an online fuzzy quality output value.

[0177] Specifically, data of energy consumption input variables are collected, which may include material strain rate, forging pressure, temperature, etc. In the self-organizing map algorithm, cluster centers are initialized, which will represent different categories or clusters of energy consumption input variables. The weight nodes are set according to the dimension of the input data and the expected number of clusters. For example, if the input data is two-dimensional and we hope to eventually cluster it into 4 categories, the weight matrix may be a 2x4 matrix. The weight vector is the initial value of each node, which can be set by random initialization or based on some prior knowledge.

[0178] The distance between the input data point and each node is calculated using the Euclidean distance formula, which is one of the most commonly used distance measurement methods. Based on the calculated Euclidean distance and weight vector, an appropriate fuzzy membership function (such as the Gaussian function) is used to calculate the fuzzy membership of the input data to each node. As a neighborhood function, the Gaussian function can define the size and shape of the neighborhood and is usually used with the learning rate to update the weight vector. The learning rate determines the magnitude of the weight update. The initial learning rate can be set based on experience and can be dynamically adjusted based on the number of iterations.

[0179] According to the fuzzy membership and learning rate, the weight vector of each node is updated. The update rule can be W new =W old +αΔW, where α is the learning rate and ΔW is the weight change. Set the maximum number of iterations to ensure that the algorithm can converge within a certain period of time. As the number of iterations increases, dynamically reduce the learning rate and neighborhood width to ensure the convergence and accuracy of the algorithm. After the iteration, according to the final membership matrix, each data point is assigned to the cluster with the highest membership, thereby obtaining the clustering result. After clustering is completed, analyze the data in each cluster to establish fuzzy rules between energy consumption input variables and quality output variables.

[0180] For energy consumption input variables and quality output variables, their fuzzy sets are defined, and each fuzzy set corresponds to a membership function. Based on domain knowledge and expert experience, fuzzy rules between energy consumption input variables and quality output variables are established. Based on the fuzzy rules, a fuzzy association matrix is ​​constructed. This matrix will represent the relationship between the fuzzy sets of different energy consumption input variables and the fuzzy sets of quality output variables.

[0181] Convert the actual energy consumption input variable value to the corresponding fuzzy value, which involves calculating the membership of the input variable in each fuzzy set. Use the fuzzy association matrix and fuzzy rules to infer the fuzzified input variables and obtain the fuzzified quality output value. This process may involve the maximum and minimum set method, that is, taking the minimum membership of each rule as the output membership of the rule. Aggregate the outputs of all rules into a single fuzzy set, usually using the maximum value as the aggregation method. Finally, convert the aggregated fuzzy output into a specific value, which can be achieved through a variety of methods, such as the centroid method, the maximum membership method, etc.

[0182] Through the above steps, an online detection of forging quality can be realized, which effectively converts energy consumption data into decision support for quality control. This method can improve the automation and intelligence level of the forging process and optimize production efficiency and product quality.

[0183] S160, using the centroid method to defuzzify the online fuzzy quality output value, obtain the online forming quality output result of the forging, and establish the energy consumption-quality database of the forging.

[0184] The Center of Gravity Method is a commonly used defuzzification method. Its core idea is to convert the fuzzy set into a specific value, which represents the "center of gravity" or "center of mass" of the fuzzy set. The online fuzzy quality output value is defuzzified using the center of gravity method to obtain the online forming quality output result of the forging. The specific calculation formula is as follows:

[0185]

[0186] Among them, m COG is the quality output result, i.e., the center of gravity of the fuzzy set. The interval (a, b) represents the value range of the quality output variable, i.e., the definition domain of the membership function. λ(m) is the membership function of the quality output variable, which describes the degree to which the quality output variable m belongs to a certain fuzzy set.

[0187] The centroid of the fuzzy set is determined by calculating the weighted average of the membership function λ(m) over the entire range of values. The numerator is the integral of the product of the membership function and the variable m, which represents the contribution of each possible value of m to the centroid; the denominator is the integral of the membership function, which represents the total "weight" of the membership function over the entire range of values. In this way, an accurate value representing the overall trend of the fuzzy set can be obtained.

[0188] Finally, the online energy consumption data is connected with the online fuzzy quality output value to form a corresponding relationship. These data are used to establish an energy consumption-quality database, which can be used for further analysis and optimization of the production process.

[0189] S170, perform time series analysis on the energy consumption-quality data, use the time series change trend to determine and mark abnormal values, and adjust the operating status of the equipment in time according to abnormal situations.

[0190] In one possible implementation, a time series analysis is performed on the energy consumption-quality data, and outliers are judged and marked using the time series change trend, and the operating status of the equipment is adjusted in time according to abnormal situations, specifically including: sorting the data in the energy consumption-quality database in chronological order to obtain time series data of the energy consumption-quality data; analyzing the time series data using a linear regression model to obtain the time series change trend of the energy consumption-quality data; judging and marking outliers in the time series data according to the time series change trend; checking the corresponding product process parameters and equipment operating parameters under the abnormal state according to the outliers, and judging the operating status of the equipment according to the product process parameters and equipment operating parameters; feeding back the operating status to the control system, and quickly adjusting the operating status of the equipment through the control system.

[0191] Specifically, first, extract data from the energy consumption-quality database and sort the data by timestamp. This will form a series of data points arranged in chronological order, laying the foundation for time series analysis. Select a linear regression model to analyze time series data. The linear regression model can help us understand the relationship between energy consumption and quality and predict future trends. Before applying linear regression, perform data cleaning, including processing missing values, outliers, etc., as well as feature engineering, such as feature transformation or creating new features.

[0192] Then use the training data set to train the linear regression model, and use the validation set or test set to evaluate the performance of the model. Through the parameter estimation of the model, we can understand the relationship between energy consumption and quality, and predict the trend of quality changing with energy consumption. Use statistical methods, such as Grubbs'test or PCA, to detect and identify outliers. Once an outlier is detected, it is annotated in the time series data for further analysis and processing.

[0193] Check the product process parameters and equipment operating parameters corresponding to the abnormal values, and analyze whether the abnormalities of these parameters are caused by equipment failure, operating errors, or other reasons. Based on the changes in these parameters, determine whether the operating status of the equipment is normal and whether there are any problems that need adjustment or maintenance. Feedback the analysis results to the control system, and the control system quickly adjusts the operating parameters of the equipment based on the feedback information to restore normal operation or optimize the production process. The control system may include algorithms such as PID control and fuzzy control, and adjust the working status of the actuator according to real-time data to ensure the stability of system parameters.

[0194] Through the above steps, it is possible to perform time series analysis of energy consumption-quality data, timely discover and process abnormal values, and adjust the equipment operation status according to the analysis results, thereby optimizing production efficiency and product quality.

[0195] This embodiment also discloses an online forging quality detection system. Figure 2 The system includes an acquisition module 201, a processing module 202 and an output module 203, wherein:

[0196] The acquisition module 201 is used to collect online energy consumption data of the forging production line, pre-process the online energy consumption data, and obtain processed data.

[0197] The processing module 202 is used to extract features from the processed data to obtain energy consumption feature data.

[0198] The processing module 202 is used to convert the energy consumption characteristic data into continuous data and normalize it to obtain standard energy consumption characteristic data, and use the standard energy consumption characteristic data to establish an energy consumption input characteristic data set.

[0199] The processing module 202 is used to define fuzzy sets of energy consumption input variables and quality output variables, and fuzzify the energy consumption input variables corresponding to the energy consumption input feature data set to convert them into fuzzy values.

[0200] The processing module 202 is used to determine the online fuzzy quality output value of the quality output result by performing fuzzy reasoning based on the fuzzy rules between the energy consumption input variable and the quality output variable using the fuzzy association matrix.

[0201] The processing module 202 is used to defuzzify the online fuzzy quality output value by using the center of gravity method, obtain the online forming quality output result of the forging, and establish the energy consumption-quality database of the forging.

[0202] The output module 203 is used to perform time series analysis on the energy consumption-quality data, determine and mark abnormal values ​​using the time series change trend, and adjust the operating status of the equipment in time according to abnormal situations.

[0203] In a possible implementation, the processing module 202 is used to extract features from the processed data to obtain energy consumption feature data, specifically including:

[0204] The processing module 202 is used to count the products using infrared photoelectric sensors to obtain product quantity data, and obtain the heating temperature and time, solution temperature and time, aging temperature and time, material strain and strain rate through the product process parameter card, wherein the calculation formula of the product quantity is as follows:

[0205]

[0206] Among them, N(ΔT) is the number of products in time, M is the number of product channels of the production line, and N i (T 0 +ΔT) represents the i-th product channel at time T 0 The number of products within +ΔT, N i (T 0) represents the i-th product channel at time T 0 The number of products within.

[0207] The processing module 202 is used to calculate the total energy consumption of the equipment through the collected equipment current, voltage and working time, wherein the equipment level characteristic data includes the working temperature and operating time of the heating furnace, the forging force and forging time of the forging press, and the operating time, current and voltage of each equipment. Specifically, the total energy consumption of the equipment is calculated by the following formula:

[0208]

[0209] Among them, E e (ΔT) is the total energy consumption of the equipment, U ej (ΔT) is the jth phase current of the three-phase AC power of the equipment, I ej (ΔT) is the j-th phase voltage of the three-phase AC power of the equipment.

[0210] The processing module 202 is used to calculate the total energy consumption of the production line through the collected operation time, current and voltage of the production line. The characteristic data at the production line level includes the operation time, current and voltage of the production line, which is specifically calculated through the following formula:

[0211]

[0212] Among them, E p (ΔT) is the total energy consumption of the production line, U pk (ΔT) is the kth phase current of the three-phase AC power of the production line, I pk (ΔT) is the k-th phase voltage of the three-phase AC power of the production line.

[0213] In one possible implementation, the processing module 202 is used to convert discrete energy consumption characteristic data into continuous energy consumption characteristic data by means of one-hot encoding. The energy consumption characteristic data is divided into continuous energy consumption characteristic data and discrete energy consumption characteristic data. The continuous energy consumption characteristic data includes current, voltage, temperature, forging pressure and material strain rate, and the discrete energy consumption characteristic data includes material strain and product quantity.

[0214] The processing module 202 is used to perform max-min normalization processing on the energy consumption characteristic data, which is specifically expressed as follows:

[0215]

[0216] Among them, x i ' is the i-th standard energy consumption characteristic data, x i is the i-th energy consumption characteristic data, x max is the maximum value among multiple energy consumption characteristic data, x min It is the minimum value among multiple energy consumption characteristic data.

[0217] In a possible implementation, the processing module 202 is used to determine the types of energy consumption input variables and quality output variables, where the energy consumption input variables are data in the energy consumption input feature data set, and the quality output variables include dimensional deviation, cracks, folds, pores, and deformation.

[0218] The processing module 202 is used to determine the actual value range of the energy consumption input variable and the quality output variable, and divide the value range of the energy consumption input variable and the quality output variable into several sections, each of which corresponds to a fuzzy set, thereby forming a fuzzy set of the energy consumption input variable and the quality output variable.

[0219] The processing module 202 is used to select different membership functions for each energy consumption input variable and quality output variable according to the distribution of energy consumption input variables and quality output variables and domain knowledge, and determine the parameters of the membership function. The expression of the membership function is as follows:

[0220]

[0221] Among them, γ(z) is the membership degree of the input variable z, h and σ are the parameters of the Gaussian membership function, and h is the mean parameter.

[0222] The processing module 202 is used to calculate the membership degree of the energy consumption input variable in each fuzzy set to obtain the fuzzy value of the energy consumption input variable.

[0223] In a possible implementation, the processing module 202 is used to calculate the material strain rate, forging pressure, and temperature data using a trapezoidal membership function, and the expression is as follows:

[0224]

[0225] Among them, a, b, c, and d are the parameters of the trapezoidal membership function, a and d are the boundary points with a membership degree of 0, b and c are the boundary points with a membership degree of 1, x is the value of the input variable, a(x) is the membership degree of the input variable x, and the graph of the trapezoidal membership function presents a trapezoidal shape.

[0226] The processing module 202 is used to use the S-shaped membership function material strain and product quantity data, and the expression is as follows:

[0227]

[0228] Among them, f and h are the parameters of the S-shaped membership function, f is the steepness parameter, and g is the center position parameter.

[0229] In a possible implementation, the energy consumption input variables are grouped into a plurality of clusters using a fuzzy self-organizing map algorithm, and the cluster data are analyzed to establish fuzzy rules between the energy consumption input variables and the quality output variables.

[0230] The processing module 202 is used to construct a fuzzy association matrix according to the fuzzy sets of energy consumption input variables and quality output variables.

[0231] The processing module 202 is used to perform fuzzy reasoning on the energy consumption input variable according to the fuzzy rules and the fuzzy association matrix to obtain an online fuzzy quality output value.

[0232] In a possible implementation, the processing module 202 is used to defuzzify the online fuzzy quality output value by using the centroid method to obtain the online forming quality output result of the forging. The specific calculation formula is as follows:

[0233]

[0234] Among them, m COG is the quality output result, the interval (a, b) represents the value range of the quality output variable, and λ(m) is the membership function of the quality output variable.

[0235] The processing module 202 is used to establish a connection between the online energy consumption data and the online fuzzy quality output value to form a corresponding relationship, and to establish an energy consumption-quality database using the energy consumption-quality data.

[0236] In a possible implementation, the processing module 202 is used to sort the data in the energy consumption-quality database in chronological order to obtain time series data of the energy consumption-quality data.

[0237] The processing module 202 is used to analyze the time series data using a linear regression model to obtain the time series variation trend of the energy consumption-quality data.

[0238] The processing module 202 is used to determine and mark abnormal values ​​in the time series data according to the time series change trend.

[0239] The processing module 202 is used to check the corresponding product process parameters and equipment operation parameters under the abnormal state according to the abnormal value, and determine the operation status of the equipment according to the product process parameters and equipment operation parameters.

[0240] The output module 203 is used to feed back the operating status to the control system, and quickly adjust the operating status of the equipment through the control system.

[0241] It should be noted that: when the system provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0242] This embodiment also discloses an electronic device, referring to Figure 3 The electronic device may include: at least one processor 301 , at least one communication bus 302 , a user interface 303 , a network interface 304 , and at least one memory 305 .

[0243] The communication bus 302 is used to realize the connection and communication between these components.

[0244] The user interface 303 may include a display screen (Display) and a camera (Camera). The optional user interface 303 may also include a standard wired interface and a wireless interface.

[0245] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0246] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0247] Among them, the memory 305 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also be at least one storage device located away from the aforementioned processor 301. The memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface 303 module, and an application for an online detection method for forging quality.

[0248] exist Figure 3 In the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 301 can be used to call an application program for an online forging quality detection method stored in the memory 305. When executed by one or more processors 301, the electronic device executes one or more methods in the above-mentioned embodiments.

[0249] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0250] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0251] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0252] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0253] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0254] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory 305 and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory 305 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0255] The present application also discloses a computer-readable storage medium, wherein the computer-readable storage medium stores instructions. When executed by one or more processors 301, the electronic device executes one or more of the methods described in the above embodiments.

[0256] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, it will be easy for those skilled in the art to think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for online detection of forging quality, characterized in that: The method comprises: Collecting online energy consumption data of a forging production line, and preprocessing the online energy consumption data to obtain processed data; Extracting features from the processed data to obtain energy consumption feature data; The energy consumption characteristic data is converted into continuous data and normalized to obtain standard energy consumption characteristic data, and the energy consumption input characteristic data set is established using the standard energy consumption characteristic data; Defining fuzzy sets of energy consumption input variables and quality output variables, and fuzzifying the energy consumption input variables corresponding to the energy consumption input feature data set to convert them into fuzzy values; According to the fuzzy rules between the energy consumption input variable and the quality output variable, a fuzzy correlation matrix is ​​used to perform fuzzy reasoning to determine an online fuzzy quality output value of the quality output result; Defuzzifying the online fuzzy quality output value by using the center of gravity method, obtaining the online forming quality output result of the forging, and establishing the energy consumption-quality database of the forging; Conduct time series analysis on energy consumption-quality data, use time series change trends to determine and mark abnormal values, and adjust the operating status of equipment in a timely manner according to abnormal situations.

2. The method for online detection of forging quality according to claim 1, characterized in that: The energy consumption characteristic data includes product-level characteristic data, equipment-level characteristic data, and production line-level characteristic data, wherein the product-level characteristic data includes product quantity, heating temperature and time, solution temperature and time, aging temperature and time, material strain, and strain rate; The feature extraction of the processed data to obtain energy consumption feature data specifically includes: The infrared photoelectric sensor is used to count the products to obtain the product quantity data, and the heating temperature and time, the solution temperature and time, the aging temperature and time, the material strain and the strain rate are obtained through the product process parameter card, wherein the calculation formula of the product quantity is as follows: Among them, N(ΔT) is the number of products in time, M is the number of product channels of the production line, and N i (T0+ΔT) represents the number of products in the i-th product channel within the time T0+ΔT, N i (T0) represents the number of products in the i-th product channel at time T0; The total energy consumption of the equipment is calculated by collecting the equipment current, voltage and working time. The equipment-level characteristic data includes the working temperature and operating time of the heating furnace, the forging pressure and forging time of the forging press, and the operating time, current and voltage of each equipment. The total energy consumption of the equipment is calculated by the following formula: Among them, E e (ΔT) is the total energy consumption of the equipment, U ej (ΔT) is the jth phase current of the three-phase AC power of the equipment, I ej (ΔT) is the jth phase voltage of the three-phase AC power of the equipment; The total energy consumption of the production line is calculated by collecting the running time, current and voltage of the production line. The characteristic data at the production line level includes the running time, current and voltage of the production line, which are calculated specifically by the following formula: Among them, E p (ΔT) is the total energy consumption of the production line, U pk (ΔT) is the kth phase current of the three-phase AC power of the production line, I pk (ΔT) is the k-th phase voltage of the three-phase AC power of the production line.

3. The method for online detection of forging quality according to claim 1, characterized in that: The converting the energy consumption characteristic data into continuous data and normalizing it to obtain standard energy consumption characteristic data specifically includes: The discrete energy consumption characteristic data is converted into continuous energy consumption characteristic data by means of unique hot encoding, wherein the energy consumption characteristic data is divided into the continuous energy consumption characteristic data and the discrete energy consumption characteristic data, wherein the continuous energy consumption characteristic data includes current, voltage, temperature, forging pressure and material strain rate, and the discrete energy consumption characteristic data includes material strain and product quantity; The energy consumption characteristic data is normalized to max-min, which is specifically expressed as follows: Among them, x i ' is the i-th standard energy consumption characteristic data, x i is the i-th energy consumption characteristic data, x max is the maximum value among the multiple energy consumption characteristic data, x min It is the minimum value among the multiple energy consumption characteristic data.

4. The method for online detection of forging quality according to claim 1, characterized in that: The fuzzy set defining the energy consumption input variable and the quality output variable, fuzzifying the energy consumption input variable corresponding to the energy consumption input feature data set to convert it into a fuzzy value, specifically includes: Determine the types of the energy consumption input variable and the quality output variable, the energy consumption input variable is the data in the energy consumption input feature data set, and the quality output variable includes dimensional deviation, crack, fold, pore and deformation; Determine the actual value ranges of the energy consumption input variable and the quality output variable, and divide the value ranges of the energy consumption input variable and the quality output variable into a plurality of segments, each value range corresponds to a fuzzy set, thereby forming fuzzy sets of the energy consumption input variable and the quality output variable; According to the distribution of the energy consumption input variable and the quality output variable and domain knowledge, different membership functions are selected for each of the energy consumption input variable and the quality output variable, and the parameters of the membership function are determined. The expression of the membership function is as follows: Among them, γ(z) is the membership degree of the input variable z, h and σ are the parameters of the Gaussian membership function, and h is the mean parameter; The membership degree of the energy consumption input variable in each of the fuzzy sets is calculated to obtain the fuzzy value of the energy consumption input variable.

5. A forging quality online detection method according to claim 4, characterized in that: The determining of the parameters of the membership function specifically includes: The material strain rate, forging pressure, and temperature data use a trapezoidal membership function, and the expression is as follows: Wherein, a, b, c, d are parameters of the trapezoidal membership function, a and d are boundary points with a membership degree of 0, b and c are boundary points with a membership degree of 1, x is the value of the input variable, a(x) is the membership degree of the input variable x, and the graph of the trapezoidal membership function presents a trapezoidal shape; The material strain and product quantity data adopt the S-shaped membership function, which is expressed as follows: Among them, f and h are the parameters of the S-shaped membership function, f is the steepness parameter, and g is the center position parameter.

6. The method for online detection of forging quality according to claim 1, characterized in that: The method of performing fuzzy reasoning based on the fuzzy rules between the energy consumption input variable and the quality output variable using a fuzzy association matrix to determine the online fuzzy quality output value of the quality output result specifically includes: The energy consumption input variables are grouped into a plurality of clusters using a fuzzy self-organizing map algorithm, and the cluster data are analyzed to establish fuzzy rules between the energy consumption input variables and the quality output variables; By constructing a fuzzy association matrix according to the fuzzy sets of the energy consumption input variables and the quality output variables; The energy consumption input variable is subjected to fuzzy reasoning according to the fuzzy rules and the fuzzy association matrix to obtain the online fuzzy quality output value.

7. The method for online detection of forging quality according to claim 1, characterized in that: The center of gravity method is used to defuzzify the online fuzzy quality output value to obtain the online forming quality output result of the forging. The specific calculation formula is as follows: Among them, m COG is the quality output result, the interval (a, b) represents the value range of the quality output variable, and λ(m) is the membership function of the quality output variable; The online energy consumption data and the online fuzzy quality output value are connected to form a corresponding relationship, and the energy consumption-quality database is established using the energy consumption-quality data.

8. The method for online detection of forging quality according to claim 1, characterized in that: Perform time series analysis on energy consumption-quality data, use time series change trends to determine and mark abnormal values, and adjust the operating status of equipment in a timely manner according to abnormal situations, including: Sorting the data in the energy consumption-quality database in chronological order to obtain time series data of energy consumption-quality data; Analyzing the time series data using a linear regression model to obtain a time series variation trend of the energy consumption-quality data; Determine and mark abnormal values ​​in the time series data according to the time series change trend; Checking the corresponding product process parameters and equipment operating parameters under the abnormal state according to the abnormal value, and judging the operating condition of the equipment according to the product process parameters and the equipment operating parameters; The operating conditions are fed back to a control system, and the operating state of the equipment is quickly adjusted through the control system.

9. An online forging quality detection system, characterized in that: The system comprises an acquisition module (01), a processing module (202) and an output module (203), wherein: The acquisition module (01) is used to collect online energy consumption data of the forging production line, pre-process the online energy consumption data, and obtain processed data; The processing module (202) is used to extract features from the processed data to obtain energy consumption feature data; The processing module (202) is used to convert the energy consumption characteristic data into continuous data and normalize it to obtain standard energy consumption characteristic data, and use the standard energy consumption characteristic data to establish an energy consumption input characteristic data set; The processing module (202) is used to define a fuzzy set of energy consumption input variables and quality output variables, and to fuzzify the energy consumption input variables corresponding to the energy consumption input feature data set so as to convert them into fuzzy values; The processing module (202) is used to perform fuzzy reasoning using a fuzzy association matrix according to the fuzzy rules between the energy consumption input variable and the quality output variable to determine an online fuzzy quality output value of the quality output result; The processing module (202) is used to use the centroid method to defuzzify the online fuzzy quality output value, obtain the online forming quality output result of the forging, and establish the energy consumption-quality database of the forging; The output module (203) is used to perform time series analysis on the energy consumption-quality data, determine and mark abnormal values ​​using the time series change trend, and adjust the operating status of the equipment in a timely manner according to abnormal situations.

10. An electronic device, characterized in that: The electronic device comprises a processor (301), a communication bus (302), a user interface (303), a network interface (304) and a memory (305), wherein the memory (305) is used to store instructions, the user interface (303) and the network interface (304) are both used to communicate with other devices, the communication bus (302) is used to realize connection and communication between components in the electronic device, and the processor (301) is used to execute the instructions stored in the memory (305) so that the electronic device executes the method according to any one of claims 1 to 8.

Images