Research Of Improved Particle Swarm Fuzzy Neural Network And Its Application In Electronic Load

In the21century, all kinds of precision equipment emerge in endlessly, and therequirements of the device to control the ascension, then the output characteristics ofpower demand more and more. But now the detection of equipment mostly remain intraditional load, and the traditional load has its unable to overcome shortcomings, thatis because the traditional load of the resistance or fixed has fixed the loadcharacteristics of, the test form a single, and use have level adjustment, is in the actualload demand higher for, such as requirement is dynamic load, load is always orfrequency changes. Therefore, in power test the production process in order toincrease the production efficiency, electronic load reference on the significantparticularly important. Electronic load is in an electronic circuit to simulate all kindsof load, it can accurately detect the load voltage, precisely adjust load current, at thesame time it can realize the load simulation circuit.Because of electronic load small volume, convenient, efficiency, test automationdegree is high, it has set up a file in the car power, space power supply, battery andcommunications are very good application, and gradually become the power in theinspection of the hot research object.This article is from electronic load research direction and development trend began to loadthe first introduced, and then the dc electronic load can realize the function of a detailedintroduction, and then to electronic load the software and hardware realize mention the designplan and eventually in the electronic load to realization of software and hardware design; Thehardware of the United States on the controller TMS320F2812processor, based on DSP dcelectronic load, is in the traditional circuit on the basis of digital link alternativesimulation link, realize the real-time and accurate control object purpose; which in turnthe keyboard circuit module design, display circuit module, current and voltage acquisitionmodule, communication module and power supply module; Software design for neural networkPID, fuzzy PID robustness and stability of shortage, design improved particle swarm optimizationfuzzy neural network PID; Finally using MATLAB, and CCSLink will CCS DSP developmentboard is connected, has completed the experimental work.