CN102540944B - Embedded multifunctional statistical process control device and method - Google Patents

Abstract

The invention discloses an embedded multifunctional statistical process control (SPC) device and method, wherein the device comprises a processor, an industrial control main board, an operating system, a basic information module, a data entry module, a data processing center, an SPC pattern formation module and an alarm module, wherein the data processing center is internally provided with a control limit computer and a comparator; the basic information module at least comprises hierarchy information of data and information of SPC control basis, quality control points and parameters; the control limit computer adopts an SPC technology to detect the distribution characteristics of collected data with a certain amount and in a certain time slot, calculates and stores the control limit; and the comparator compares the data and the control limit according to the judged rules of the SPC control basis. According to the invention, an SPC system is embedded in an embedded industrial control device, the seamless integration of software and hardware is realized, the system can rapidly and accurately complete quality data analysis based on the statistical principle, and adapts to the specific requirement of different manufacturing systems, and the quality control of a product process is realized.

Description

Embedded multifunctional statistical process control device and method

technical field

The invention relates to the technical field of data processing, and more specifically relates to an embedded multifunctional statistical process control device and method.

Background technique

The development of microelectronics technology and manufacturing application software technology has made mechanized, electrified, and automated production equipment in the traditional sense of industrialization have the characteristics of digitalization, intelligence, and networking. Manufacturing application software is the fusion agent of informatization and industrialization. At present, many enterprises regard informatization work as an important part of enterprise development. However, due to some technical development restrictions, in the process of enterprise informatization, most of the hardware and software are separated. Seriously, even As a result, the actual application is disconnected, the application software can not play a good role, and even the system is no longer used once it goes online. Even enterprises that have implemented a modern information management system, because of the characteristics of the manufacturing industry, can only enter production data manually or Common PCs need to be used in industrial production sites, coupled with factors such as the inherent instability of the current Windows operating system itself, making the use and maintenance costs of the system very high. In the internal production department of the enterprise, in order to control the overall production status of the product and establish a mechanism to prevent unqualified products, managers often use the statistical process control method (Statistical Process Control, SPC), which is based on a large amount of data and facts. Process information and analysis, with the help of mathematical statistics technology, to find abnormal conditions in time, so as to take measures to eliminate abnormalities and improve processes, and pursue perfect quality management methods. Although many softwares about SPC have emerged, in view of the above information and Due to the shortcomings of the combination of industrialization, the application is limited to a certain extent.

Contents of the invention

The object of the present invention is to overcome the shortcomings of the above-mentioned prior art, provide an embedded multifunctional statistical process control device and method, and provide a good tool for enterprise product quality management.

The technical scheme adopted in the present invention is: an embedded multifunctional statistical process control device, including a processor, an industrial control motherboard, and an operating system, and also includes:

Basic data module, at least including data level information, SPC control basis, quality control points and parameter information;

Data entry module, including one or more of manual data entry module, file or database import module, instrument and equipment data acquisition module;

The data processing center is equipped with a control limit calculator and a comparator. The control limit calculator applies statistical process control technology to detect the distribution characteristics of the data collected in a certain amount and within a time period, and calculates the control limit coexistence storage, the comparator compares the data with the control limit according to the discrimination rule of the SPC control basis;

The SPC graphics generation module marks the data on the control limit diagram of the corresponding level;

The alarm module is configured to alarm the abnormal data detected by the data processing center.

The above-mentioned embedded multifunctional statistical process control device also includes a report output module, and the data processing center processes the collected data and generates statistical reports for production management.

The above-mentioned embedded multifunctional statistical process control device, the industrial control main board includes a CPU module, an IDE hard disk, a memory, a display module, a power supply module, and also includes an Ethernet connection module, a wireless network module connected to WiFi, a touch screen interface module, and a USB interface. module, the Ethernet connection module, the wireless network module, the touch screen interface module, and the USB interface module are all connected to the CPU.

The above-mentioned embedded multifunctional statistical process control device includes a CPU module, an IDE hard disk, a memory, a display module, and a power supply module, and is characterized in that it also includes an Ethernet connection module, a wireless network module connected to WiFi, a touch screen interface module, and a USB interface. module, the Ethernet connection module, the wireless network module, the touch screen interface module, and the USB interface module are all connected to the CPU.

In the above-mentioned embedded multifunctional statistical process control device, the CPU module is also connected with an infrared remote control module, a camera connection module, and an audio module.

The control method of the above-mentioned embedded multifunctional statistical process control device comprises the following steps:

1) Analyze the product manufacturing process, determine the quality control points and their parameters;

2) Set product varieties, procedures, quality control points, parameters, and bad code information in layers;

3) Select the appropriate control chart type according to the level;

4) Collect and analyze the basic data, and judge the stability of the data;

5) Calculate the control limit with the collected stable data;

6) The daily process data is collected and judged to be stable. If the process is stable, the monitoring will continue, and if the data is unstable, an alarm will be triggered.

In the above-mentioned multifunctional statistical process control method, the capability of the manufacturing process is determined after the steps of basic data collection, analysis and judgment.

Compared with the prior art, the present invention has the following advantages:

1) The present invention establishes a set of software application system with certain practicability. The SPC statistical process control system is embedded in the embedded industrial control equipment to realize the seamless integration of software and hardware. The system application is a highly integrated software and hardware framework. On top of that, it possesses the powerful functions of a PC and the advantages of an embedded system. When applied to manufacturing equipment in the manufacturing industry, it can quickly and accurately complete various quality data analyzes based on statistical principles, adapting to the specific requirements of different manufacturing systems. Requirements, to achieve the quality control of the product process.

2) In the present invention, the intelligent industrial control motherboard with the characteristics of strong functions, low power consumption, high performance, multiple interfaces, and high resolution is used as the support at the bottom of the complete system of industrial control equipment, and the Windows CE operating system is embedded. At the same time, SPC technology and The product quality control is organically combined, aiming at the practical application of SPC, the design of its simulation software is analyzed by using c# high-level programming language, and the control model is established.

3) The present invention has functions such as multi-language support, arbitrary customization of system layered information, unlimited expansion of product and quality control points, multiple alarm modes, adjustable rules for judging abnormalities, comprehensive authority management mechanism, and supports ERP system, etc., the system is powerful practical.

Description of drawings

Fig. 1 is the structural representation of the industrial control motherboard of the embedded multifunctional statistical process control device of the present invention;

Fig. 2 is a structural frame diagram of the embedded multifunctional statistical process control device of the present invention;

Fig. 3 is a working flowchart of the embedded multifunctional statistical process control device of the present invention.

Detailed ways

The present invention will be further described below through specific embodiments and in conjunction with the accompanying drawings.

See Figure 1-3. SPC statistical process control technology is applied to enterprise indicators expressed as a quantity using multidimensional data. The system can detect potential process capabilities and laws in the production process of enterprises, and monitor their fundamental changes. Therefore, it can Notify the production manager of the enterprise of the potential variation trend of a quality control point through the alarm, and improve the quality of the product through corrective and preventive measures. The embedded multifunctional statistical process control device of the present invention perfectly combines SPC control technology and computer control technology, and provides a good tool for product quality management of enterprises.

Industrial controller mainboard includes CPU module 1, IDE hard disk, memory 9, display module 2, power supply module, Ethernet connection module 3, wireless network module 4 with external WiFi, touch screen interface module 12, USB interface module 5, infrared remote control Module 7, camera connection module 6, audio module 11, external memory 8, sound card 10, the above-mentioned modules or peripherals are all connected to the CPU. The Windows CE operating system is embedded in the IDE hard disk. The motherboard adopts the AU1250 processor of RMI Company, and the 3.5-inch (146mm x 102mm) motherboard is used as the bottom support of the whole industrial controller system. The embedded Windows CE operating system can be used for the development and use of various applications, and can be applied to various industries. All walks of life. RMI's AlchemyAu1250 series processor is a low-power, high-performance system-on-chip solution for multi-function portable media players (mPMP), mobile infotainment devices, home digital media adapters (DMA), navigation equipment, medical For electronic and industrial applications, the Au1250 processor supports MPEG-1, MPEG-2, 4wmv9, DivX and Xvid full D1 (720x480) video playback at 30fps smoothly, with rich processor interfaces and powerful functions, and supports a maximum resolution of 1280x1024, especially suitable for general For high-performance applications that cannot be met by ARM processors, its SoC architecture integrates most of the peripheral devices, thus greatly reducing the cost of materials. The following hardware is integrated on the industrial control motherboard: power supply, CPU, RTC, FLASH, DDR2 SDRAM, Ethernet, WiFi, USB2.0, RS232/RS485 serial port, CF/SDHC/MMC/xD/MS card, IDE hard disk, sound card, power amplifier , Touch screen controller, 8x8 keyboard, infrared remote control, VGA, LCD driver, backlight control, Camera, etc. The operating system adopts the Windows CE embedded operating system produced by Microsoft Corporation. This system is a compact, efficient, and scalable 32-bit operating system. It can be tailored according to actual needs and is mainly oriented to various embedded systems and products. Its multi-threading, multi-tasking, and fully preemptive features are specially designed for various hardware systems with very strict resource constraints. The following technical indicators can be achieved by the above scientific configuration, providing strong support for the SPC control system:

(1) Support handwriting touch screen of 7 to 15 inches;

(2) Anti-virus intrusion, easy maintenance;

(3) Low power consumption, which is about 1/10 of ordinary PC, which greatly saves electric energy. Compared with ordinary PC, each set can save about 2,000 yuan of electric energy for enterprises every year;

(4) The hardware is small in size and can be placed in the workbench or installed in the equipment, saving the space of the factory site;

(5) High stability and higher security protection level than ordinary PCs;

(6) Compared with ordinary PC, the maintenance cost is low, and the system has a self-protection mechanism;

(7) Through expansion, multiple devices can be connected at the same time, and it supports USB interface, Ethernet port, RS232/RS485, and wireless transmission function.

In order to statistically analyze the data scientifically, find out the factors affecting the quality of key control points, and prevent defective products, so as to achieve the purpose of reducing quality costs and improving product quality, the entire system consists of the following modules:

(1) Data entry module, including manual data entry, file or database import, instrument and equipment data collection, to complete the data collection and entry of each control point of the products produced by the enterprise.

(2) Basic data module, including:

Hierarchical information entry - complete the hierarchal information in the production process, so that multi-directional control can be carried out at different levels during process control, such as batches, equipment, personnel, shifts, customers and other information.

Quality control points and parameter entry - complete the entry of key control parameter information for each product.

User information entry - complete the entry of basic user information for system use, including system operation authority, etc.

SPC control basic data entry - Complete the entry of SPC theoretical guidance information, including the entry of basic information of control charts (specification limits, control limits, sample size, process capability judgment standards, etc.), discrimination rules and other information.

(3) SPC graphics generation - including the generation and viewing module of the operation diagram, to complete the viewing and analysis of the operation trend diagram; the control diagram generation and viewing module, to complete the viewing and analysis of the control diagram; the generation and viewing of the process capability analysis diagram module to complete the viewing and analysis of the process capability analysis chart.

(4) Report and guidance file output module - complete the results after summarization, statistics and analysis, and guide production.

(5) Real-time alarm module—complete the instant warning of production data in the production process to prevent the occurrence of quality accidents. The alarm module compares the data entered or obtained by the device against the control limit parameter value to determine whether there is a situation beyond the limit, and if the limit is exceeded, an alarm is triggered. The way of alarm is selected by the user and preset in the system, such as sending alarm e-mail, SMS service, turning on the alarm indicator light and so on.

There is also a maintenance module in the system, which is used to complete the maintenance of system data, such as initialization, data backup and data recovery.

The present invention uses c# as the programming language, Visual Studio 2008 is the development environment, and Microsoft SQL Server 2000 is the background database. Because WinCE development is the VC++ environment familiar to everyone, it is convenient and fast. With Microsoft's strong technical support, its development difficulty is much lower than that of embedded Linux. Transplanted to embedded systems; the latest Visual Studio 2008 already supports the development of WindowsCE applications based on Compact Framework, plus Windows CE can be easily migrated from Win32 applications, and the already researched SPC basic framework based on Windows Programs and basic classes can be transplanted to Windows CE applications through the development of Visual Studio 2008, and then make necessary adjustments and improvements according to the actual requirements of applications under Windows CE, and the expansion of functions can be formed. Through the above schemes, the SPC statistical process control system is embedded in the embedded industrial control equipment, and the application of seamless integration of software and hardware is achieved. In addition, after the program development is completed, the program system and some required components, Windows CE operating system and related components are compiled, packaged and integrated together, and burned into the embedded industrial control equipment at one time. This makes the system easy to maintain, free from interference from external factors such as viruses, and thus work stably.

The software design of the present invention establishes a mathematical model, and when the system receives data from various input modules, the SPC data processing center processes the data and generates an SPC control chart, thereby providing managers with accurate and intuitive quality control means. The following is a detailed introduction to the generation of the SPC control chart of this system and the criterion of discrimination. SPC control chart is mainly divided into variable (Variablechart) chart and attribute (Attribute chart) chart and prediction chart (Capabilitychart). There are three control lines in the control chart, namely the upper control line USL, the control center line SL and the lower control limit LSL, using the following calculation formula:

$\left\{\begin{array}{c}\mathrm{USL}=\mathrm{\μ}=3\mathrm{\σ}\\ \mathrm{SL}=\mathrm{\μ}\\ \mathrm{LSL}=\mathrm{\μ}-3\mathrm{\σ}\end{array}\right.$ μ: mean or standard value σ: standard deviation

The system uses SPC technology to calculate control limits to measure and analyze process changes. Usually, the control limit calculator is used to complete this function. First, the system will use at least 100 data points within a period of time to calculate the upper control limit, lower control limit, and average value through SPC technology, and generate a control chart. The control chart is usually Display three horizontal lines of upper control limit, lower control limit and center line (average value), and mark the position of each data point on the graph. The commonly used charts for the SPC control chart set by the system include: defect rate control chart, defect number control chart, mean value control chart, standard deviation control chart, range control chart, etc.

Among them, the defect number control chart is based on the Poisson process, which can conveniently monitor the number of "defects" of equal samples, and is suitable for graphical representation of the number of defective reports. The mean and range control charts are commonly used in daily production process control.

The data of a certain interval is firstly taken as the preliminary data, and the number of data N is preferably more than 100. Divide N into groups of k=20-25 with n=4 or 5 according to time and form a table, Xmax is the maximum value in the data group, Xmin is the minimum value in the data group, and the data is calculated according to the following formula :

The design of the data structure unified the five measured values as a set of data groups. In order to make the drawing and the final calculation more convenient, it is necessary to store all the data set contents. Objects can be added dynamically using containers. In addition, in order to facilitate drawing and provide hotspot query (click on the data point on the coordinate axis to display the detailed information of the data group where the data point is located), write a data class (DetailData) to save the detailed information of the data group, including data The 5 basic data values in the group, the mean value of the data group (that is, the value of the data point in this data group), the extreme difference value of the data group, the date and time of collection, and the description of the abnormal situation. Then write an attribute class (Pointer), which describes the data point attributes, including the abscissa value x, ordinate value y, abnormal situation, whether it meets the specification, and the object of the specific information of the data group.

When a set of data is successfully measured, first generate an object of the data class and record the information of the data set, including 5 basic data, the mean and range of the data set, the date and time of collection, and abnormal conditions (if none, it is normal) . Then calculate the corresponding coordinate value according to the mean value of the data group, and if it meets the specification, an object of attribute class is generated and added to the container. If there is an abnormal situation, an object of the data class is also generated, and the basic information of the data group is set to 5 basic data, the mean value and the range are both 0, and the collection date and time and the corresponding abnormal situation are set. The ordinate value is the minimum value of the ordinate axis, if it is set to not meet the specifications, an object of attribute class is also generated and added to the container. In this way, subsequent drawing and statistical calculations can be performed relatively simply and conveniently.

Before simulating the actual measurement, the system first calculates the mathematical expectation (mu) of the data according to the data range, the user inputs the standard deviation (sig), uses the function method to generate data with a standard normal distribution, and then the system generates the corresponding non-standard Data that obeys a normal distribution with mathematical expectation mu and variance sig squared.

After measuring 100 normally distributed data, the system automatically calculates the corresponding statistical data and performance indicators such as the average value and standard deviation of the data and displays them. This step verifies the correctness of the program. The subsequent data is randomly generated by the system. The range of the measured value is between the value plus the difference and the value minus the difference. There is a function method in C# to generate random numbers. The design uses a random value that is greater than 0 and less than the upper difference, and that is greater than the lower difference and less than 0. A random value of , and the value is added to the two to obtain the measured value. This design can generate as many values in the middle range as possible, and make the data normally distributed as much as possible.

Discrimination rules and control limit adjustment rules must also be set in the basic module of SPC control. The Control Limits Calculator looks for trends in existing data if:

●Individual points are above the upper control limit or below the lower control limit

●Consecutive 9 points are on the same side of the center line

●There are six data points with consecutive rises or falls

●Consecutive 14 adjacent points alternate up and down

●All seven points are above or below the average

● Data loops or other non-random patterns occur

In the above situation, the control limits need to be recalculated. For example, when individual over-limits occur, suspicious points should be eliminated first. After recalculation, the new control limits and the average value will have a better distribution of data points (there are approximately equal points on both sides of the average value). amount of data points), it can be considered that the new control limits are more appropriate and have higher reliability. If there are seven data points with continuous rise or fall, it can usually be considered that the data has changed significantly, and more attention is required from the enterprise management personnel.

Claims (6)

1. embedded type multifunctional statistical process control method is characterized in that may further comprise the steps:

1) analytic product manufacture process is determined Quality Control point and parameter thereof;

2) product variety, operation, Quality Control point, parameter, bad code information layering are arranged;

3) press the suitable control chart type of hierarchy selection;

4) basic data acquisition and analysis, data are declared surely;

5) stability number that collects be it is calculated that the control limit;

6) daily process data collection and declare steady, if process stabilization then continues monitoring, if then trigger alarm of data shakiness occurs.

2. embedded type multifunctional statistical process control method according to claim 1 is characterized in that: basic data acquisition, analyze and declare steady step after determine the processing procedure process capability.

3. embedded type multifunctional statistical Process Control device of implementing embedded type multifunctional statistical process control method as claimed in claim 1 is characterized in that: comprise processor, industrial control mainboard, operating system, also comprise:

The basic data module includes the hierarchical information of data, the information that SPC controls basis, Quality Control point and parameter at least;

Data typing module comprises that manual data typing module, file or database import one or more of module, instrument and equipment data acquisition module;

Data processing centre (DPC), in be provided with control limit counter and comparer, the distribution character that described control limit counter detects data to the some that collects and the market demand statistical process control technology in the time period, calculate control limit and storage, described comparer compares data and control limit by the different rule of declaring that SPC controls the basis;

SPC figure generation module indicates the control limit figure of data at corresponding level;

Alarm module is reported to the police to detected unusual data cases by data processing centre (DPC).

4. embedded type multifunctional statistical Process Control device according to claim 3 is characterized in that: also comprise the report output module, processed the data that collect and generated the statistical report form that production management is used by data processing centre (DPC).

5. embedded type multifunctional statistical Process Control device according to claim 4, it is characterized in that: described industrial control mainboard comprises CPU module, IDE hard disk, internal memory, display module, power module, also comprise the Ethernet link block, the wireless network module of external WiFi, touch screen interface module, usb interface module, described Ethernet link block, wireless network module, touch screen interface module, usb interface module all are connected with CPU.

6. embedded type multifunctional statistical Process Control device according to claim 5, it is characterized in that: described CPU module also is connected with infrared remote control module, camera link block, audio-frequency module.

Images