A comprehensive revitalization of intelligent technology

Nowadays, intelligent branch of science and technology has flourished and flourished at home and abroad, such as fuzzy logic, genetic algorithm, neural network, expert system, human intelligence, rough set theory, matter-element extension method, knowledge engineering, model Identification, qualitative control, wavelet analysis, fractal geometry, chaos control, data fusion technology, etc., can be described as the Eight Immortals crossing the sea, recount. They have their own strengths, respectively, combinations, complement each other.

Artificial neural network is the basic technology of intelligent technology today. Its connection mechanism is parallel to the symbol reasoning mechanism of artificial intelligence, and becomes the two camps of intelligent technology. It simulates the anatomy and physiology of the human brain. It uses many parallel simple neurons to connect with a certain topology to form a network. It receives external information and stimulates each other. It specializes in distributed memory, associative memory, feedback refinement, black box mapping Self-organization, self-diagnosis, self-repairing ability, and the weight of the network nodes, the weights of the network nodes, Dynamic feedback, dynamic analysis, and language, audio-visual interface with the man-machine interface, can automatically obtain a wealth of knowledge and experience of human experts, and simulation of human brain logic reasoning, imagery thinking and inspiration emergent, appropriately handle various inaccuracies , Imperfect, uncertain information, reasoning to draw the correct conclusion.

Fuzzy logic to imitate the concept of the human brain to determine the uncertainty, reasoning thinking, for the model is unknown or uncertain description system, and strong nonlinear, large lag of the control object, fuzzy sets and fuzzy rules for reasoning, the transitional boundaries Or qualitative knowledge and experience, to simulate the human brain way, the implementation of fuzzy comprehensive judgment, reasoning to solve the conventional method is difficult to deal with the rule-based fuzzy information problem. Fuzzy logic is good at expressing qualitative knowledge and experience with unclear boundary. It makes use of the concept of membership function, distinguishes fuzzy set, deals with fuzzy relation, simulates human brain to implement rule-based reasoning, and solves the problems caused by logic break. Identify the problem.

Genetic Algorithm (GA) is a search method to suppress the computational explosion of the search space with the feature of "electron beam search". It can search the solution space with multi-points, and use genetic algorithm to repeatedly crossover and mutate the operation. Diversity and high adaptability to environmental change, which is characterized by strong operability, and can avoid falling into the local minimum point, the problem of global convergence quickly, is a class of information can be a number of global use of the autonomous decentralized system. Genetic algorithm (GA) and other evolutionary methods can be made of the evolutionary hardware (EHW), can produce more than the existing model of the technical integration and design capabilities of the new circuit, in particular GA unique global optimization performance, its self-learning, Adaptive, self-organizing and self-evolving capabilities. It can automatically integrate and extend large-scale parallel processing (MPP) in unmanned space, and configure and call EPG-based functional EHW in real time and flexibly. Space complexity of the uncertainty of the opening of the course.

Expert system is the application of human experts to collect the knowledge and experience, mimic experts to deal with knowledge and problem-solving methods, compiled into a computer intelligent software system, through the combination of continuous feedback information, real-time online rules, Model to implement independent decision-making of a problem-solving or control system. This computer-based intelligent system is enlightening, transparent and flexible. It is highly efficient, accurate, accurate, comprehensive, fast and tireless in terms of time, space and environment. Its ability to solve problems and knowledge Wide range can be more than human experts, but also to overcome the human experts because of negligence, forgetfulness, tension, fatigue and other interference factors caused by errors and errors, so its promotion, application has enormous economic and social benefits.

Pattern recognition is a kind of intelligent decision-making method and technology which imitate human brain image thinking and identify, judge and deal with things according to the characteristics, images or relations of things. It is widely used in scientific research and production and is a valuable technical method .

Rough set theory is in the discrete normalization processing in the measurement of the data set, through the set element based on the indivisible relations of algebraic operation, the use of conditions and results attributes in a large number of useful features, effective data discovery knowledge, Then, the rule is simplified and the minimum decision algorithm is selected according to the requirement of the problem. In this way, the redundant attribute of a large amount of information is removed, and the dimension and the number of attributes of the information space are reduced. It can greatly simplify the network structure and the number of samples, shorten the training time, is a smart technology in a fundamental analysis. This method is based on the measurement data set and access to knowledge, so the intelligent development of virtual instruments is of great significance.

Chaotic motion is a highly unstable motion confined to finite phase space in deterministic systems. It is a disorder in order, which causes things to show some chaos on the surface for a long time. Chaos phenomenon is characterized by "non-periodic order behind the hidden" and "sensitive to the initial conditions of dependency", make full use of chaotic features, intelligent information processing in the implementation of nonlinear decision making and prediction, , Image data compression, high-performance confidentiality, multi-target search, and infinite rich, fantastic computer graphics and other magical applications.

Fractal theory of non-linear system generated by non-smooth and non-differentiable geometric shape and its internal structure of the proportion of self-similarity for the study of all the complex nature of the movement of things to provide a powerful tool and method.

Wavelet analysis is the backbone of the modern analytical mathematics tree and the most perfect crystallization. From the visual point of view, wavelet is that people can observe the shortest, most simple positive and negative the same, with the attenuation of the oscillation wave; and mathematically speaking, wavelet function f (t) is the center of the three conditions Window function, which can not only characterize the localization of signal in time domain and frequency domain, but also retain all the information of the signal completely, and has a zooming property, that is, for a high-frequency signal with a very short time Window, and in the low-frequency and has a wide range of different scales of transformation. The essence of wavelet analysis is the reflection of the wave-particle duality of the world of things and the dialectical relationship between the local and the whole multi-level representation. The most attractive feature is the time-frequency localization and multi-scale approximation capability. In adaptive control, robust control, Linear control, process identification, neural networks and many other fields have achieved fruitful results.

Fractals and chaos are essentially the same two aspects. Chaotic events show similar patterns of variation at different times, while fractal represent similarity under spatial scales. Chaos is concerned with its complex process of instability, divergence, convergence, while the fractal is to describe the chaotic movement of intuitive geometric language. The combination of chaos, fractal and wavelet analysis has rich connotation and profound philosophy, which will provide a powerful tool for the solution of material science and technology problems such as automatic assembly of materials, high-speed gene sequencing and efficient protein structure prediction. But also for the instrumentation of virtualization, network and intelligent to open up a brilliant future.

Matter-element extension method is based on the comparison and optimization of many known general decision-making, and breaks through the conventional and extensively creative decision-making according to the need of incompatible contradictory problems at all levels and stages. Skills, to seize the key strategies to maximize meet the main system, incompatible contradictions into compatible relationship, in order to achieve the overall best decision-making objectives. It is in the complex system to resolve the secondary contradictions, to resolve the main contradictions and key challenges of a powerful means of instrumentation will also be virtual, network and intelligent development process to make a significant contribution.

Data fusion technology is a kind of technical method for calculating the total optimal representation value of the characteristic attribute according to its importance and credibility distribution in the whole system with different weight proportion. It is a kind of technique to measure and represent the properties of complex things, which is of great significance to the research of high-tech development.

In short, the world of intelligent technology is rapidly and comprehensively forward.

Application of Intelligent Technology in Instrumentation and Measurement

Intelligent automation technology is fully integrated into the instrumentation industry.

(1) instrumentation in the structure, performance improvement in the application

First of all, intelligent automation technology for instrumentation and measurement of the relevant fields of application has opened up broad prospects. The use of intelligent hardware and software so that each instrument or instrument can at any time and accurately analyze and deal with current and previous data information, from the appropriate low, medium and high levels of abstraction of the measurement process in order to improve the existing measurement system Such as neural network, genetic algorithm, evolutionary computation, chaos control and other intelligent technology, so that instrumentation to achieve high-speed, high efficiency, multi-function, high flexibility and other properties of the motor.Therefore, the performance of the traditional measurement system,

Secondly, it can also be used in different instruments and meters distributed system micro-controller, micro-controller and other micro-chip technology, the design of fuzzy control procedures, set the threshold of various measurement data, fuzzy rules fuzzy reasoning technology, And various fuzzy relations to carry out various types of fuzzy decision. The advantage is that it does not need to establish the mathematical model of the controlled object, and do not need a lot of test data, just according to experience, summarize the appropriate control rules, application chip off-line calculation, on-site debugging, according to our needs and accuracy to produce accurate Analysis and timely control action.

Especially in the sensor measurement, intelligent automation technology is more widely used. It is an effective way to simplify the hardware, improve the signal-to-noise ratio and improve the dynamic characteristics of the sensor, but need to determine the dynamic mathematical model of the sensor, and the high-order (high-order) Fourier transform is used to realize the signal filtering, such as fast Fourier transform, short-time Fourier transform, The real-time performance of the filter is poor. Using neural network technology, can achieve high-performance auto-correlation filtering and adaptive filtering. Self-organizing, associating, memorizing, and black-box mapping between the input and output of nonlinear complex relations. In the aspects of applicability, fast real-time and so on, it can make full use of the artificial neural network technology. Will greatly exceed the complex function type, can take full advantage of multi-sensor resources, integrated access to more accurate and more credible conclusions. The real-time and non-real-time, fast-changing and slow-changing, fuzzy and deterministic data information may support each other or contradict each other. At this time, the extraction, fusion and decision- Will become difficult. So neural network or fuzzy logic will become the most worthy of choice. For example, the gas sensor array is used for the mixed gas identification. In the signal processing method, the self-organizing map network and the BP network can be used to classify and re-identify the components. The whole process of the traditional method is transformed into To reduce the complexity of the algorithm to improve the recognition rate. Another example, the taste of food taste detection and identification of the difficulty of the research and development units was once a major obstacle. Now the data compression and feature extraction can be done by wavelet transform, and then the data input to the fuzzy neural network trained by genetic algorithm can greatly improve the recognition rate of simple composite flavor. For example, in the field of intelligent evaluation of fabric quality, flexible manipulator's handling of haptic signals, and fault diagnosis of machines, intelligent automation technology has also achieved a lot of successful examples.

(2) in the virtual instrument structure design

Instrument and measurement technology and computer technology, not only greatly improve the measurement accuracy and intelligent automation level, especially computer hardware softening and software modularization of the rapid development of virtual instrumentation, and its network system resources procedures and the unity of the Optimize the performance of the configuration, intelligent instrumentation for the rapid increase in the level of creating more and more favorable conditions.

In the instrument structure design, instrument manufacturers in the past are in the form of source code to provide users with intelligent virtual instrument plug-and-play instrument driver, in order to simplify the end-user operation and development process, and continuously improve operational efficiency, and programming quality and Programming flexibility, the relevant instrument manufacturers in the VXI plug-and-play bus instrument driver based on the standard made a new set of intelligent instrument driver software specifications, the virtual instrument structure and performance of the following improvements in many ways.

First, consider the need to take into account the user's intuitive, easy to use and maximize operational efficiency, and maintain the original VXI bus plug-and-play standard high-level programming interface to provide the same function function call format.

Secondly, based on the latest development tools of Labwindows / CVI 5.0, intelligent instruments can be used to generate the instrument driver code of intelligent virtual instrument (IVI), which can simplify the programming work. Volume, and unified the drive code programming structure and style, but also greatly facilitate the use of different levels of user and maintenance.

Once again, a series of intelligent techniques are used to identify, track and manage all instrument status and settings, allowing users to go directly to all low-level settings and through intelligent state management, allowing users to test and develop, Run "switch between the two modes at random. In Test Development mode, the drive intelligently and automatically performs a series of status checks to help you discover various programming errors. When the program is ready for normal use, the user can switch to the "normal operation" mode, so that the driver software to run at high speed. This will not only ensure the safety and reliability of the instrument, but also make the software at any time into high-speed operation, as far as possible to improve its operating efficiency.

In addition, due to the use of a variety of intelligent methods to enable multi-threaded multi-threaded safe operation, multi-threaded parallel test; at the same time, the drive also has a powerful simulation capabilities, can not connect the actual instrument case, the development test program.

The last feature is that the drive is only related to the test function, independent of the interface bus mode used by the instrument. The instrument interface bus and the area are differentiated only by an initialization function In it with Options.

In short, the virtual instrument using a series of intelligent automation means completely changed the past VXI bus plug-and-play standard instrument driver operating efficiency is low, the programming structure, style inconsistencies, programming difficulties, low quality, heavy workload, use, Maintenance, trouble and so a series of shortcomings, which in high efficiency, high quality, safe, reliable, easy to use, flexible to achieve a comprehensive and unified operation under the intelligent automation technology for virtual instrumentation and instrumentation industry as a whole the rapid development of far-reaching impact.

(3) instrumentation network in the application

Because the instrument and the computer once compose the network, can take advantage of intelligent software and hardware (such as pattern recognition, neural network self-study, self-adaptation, self-organization and association memory function), give full play to flexible call and reasonable configuration of various computers and instruments Instrumentation of their respective resource characteristics and potential, resulting in a combination of 1 +1> 2 advantage. For example, you can now use a digital multimeter connected to the Web and oscilloscopes, through the Internet and pattern recognition software distinguishes different time and space conditions and instrumentation of the category characteristics and the measured threshold, to make a different characteristic response; can also use distributed data The acquisition system replaces the previously used data acquisition devices alone, enabling remote measurement and acquisition of data and storage and application of categorized data across Ethernet or other networks.

Network-based intelligent measurement environment will be online various types of different tasks of the computer and instrumentation organically linked together to complete various forms of task requirements, such as data collected in a place sent to a variety of needs where these data, The need to copy the same data on multiple copies, sent to the needs of the department; or regularly sent to the remote database measurement results saved, when needed to call at any time. And multiple users can simultaneously monitor the same process, such as the engineering and technical personnel, quality control personnel and competent leaders at the same time separately in the remote monitoring and control of the same production and transportation process, do not have to visit the scene and timely collection All aspects of data, decision-making or the establishment of databases, analysis of the phenomenon of law. In the event of problems, can immediately show immediate or reconfigured, or immediately discuss the decision to take immediate action.

In addition, the intelligent reconstruction of information processing technology will also create a broader instrument for the activities of the stage. Reconfigurable computers, which combine the advantages of computer and application-specific integrated circuits (ASIC), not only need to flexibly configure a large number of programmable logic cell arrays (FPGAs) according to different computing tasks, their instruction level, bit level and pipeline level As well as task-level parallel computing, it runs at speeds up to hundreds of times more than general-purpose computers.

In summary, with the intelligent automation technology applications and the growing depth of the scope and scale of the continuous expansion of China's instrumentation industry, the level of development will quickly move towards a higher stage.

Future Prospects of Intelligent Automation of Three Instruments
Intelligent technology in the instrumentation applications are rapidly developing rapidly, many other areas of new technologies are constantly integrated. For example, in the full play of the highest speed of optical properties of the beam, based on the increasingly intelligent human brain. And actively use the human brain mechanism and bio-DNA chip organic intelligence, and electronic, photon computing speed of inorganic intelligent high efficiency, active advantage of combining, and intelligent materials, and virtual interaction, and jointly improve. Today, optical interconnect technology is to a very high temporal and spatial bandwidth, a very small electromagnetic interference and a small interconnect power consumption and a series of unique physical properties, to overcome the physical nature of the interconnection technology limit for the dynamic , Flexible, high-speed, real-time reconstruction of network interconnection structure, greatly improving the parallel processing power, and create a new world. This will create a variety of mankind, open man-machine combination system, and colorful anthropomorphic high intelligent and efficient automation system to lay a solid foundation, which will continue to push the human social productivity to a new higher realm, so that human life toward the intelligent world Happy and beautiful tomorrow big step forward!