With the development of high-performance computer measurement and control systems, traditional sensors are no longer compatible with their measurement and control systems. The control system requires the sensor to have strong information processing and self-management capabilities to achieve information collection and information pre-processing, reduce the data processing burden of the control computer and improve the reliability of the entire measurement and control system. However, computers focus on high-level processing and processing of information, which greatly improves the performance of the system under existing hardware conditions and simplifies the structure of the system. The intelligent sensor system is a new research direction proposed to better adapt to the development of computer measurement and control systems.
A pressure sensor, when the pressure parameter is constant and the temperature changes, its output value also changes, that is, the pressure sensor has a composite sensitivity of the temperature parameter. When the sensor has a composite sensitivity, it will cause its performance to be unstable and the measurement accuracy to be low. Multi-sensor information fusion technology is to improve the measurement accuracy of each parameter by monitoring multiple parameters and adopting certain information processing methods. In the case where only one target parameter is required to be measured, in order to improve the measurement accuracy of the measured target parameter, the other parameters are regarded as the interference amount, the influence is eliminated, and the accuracy of each parameter measurement is improved, so the sensor information fusion technology is developed more. Functional sensor systems open the way. This paper mainly discusses the application of intelligent pressure sensor technology and data fusion technology.
1 Hardware design of intelligent pressure sensorFigure 1 shows the hardware block diagram of the intelligent pressure sensor. In this design, the hardware system is divided into two parts:
(1) The pre-processing part of the sensor output signal. It is mainly composed of signal conditioning circuits, including the design of constant voltage source circuits for static pressure and temperature sensors, and filter amplification of sensor output signals.
(2) The signal analysis processing section converts the analog signal into a digital signal and completes the analysis and processing of the signal.
In the signal processing part, this paper focuses on the design of hardware circuit using ADuC812 single-chip microcomputer, which is simple in structure and small in size. Its functional block diagram is shown in Figure 2. Figure 2 shows that the ADuC812 microcontroller has different features from other microcontrollers: it integrates high-performance 8-bit (8051-compatible) MCUs that can be reprogrammed with non-volatile flash/electrically erased program memory, and includes high performance. Self-calibrating 8-channel ADC and 2-channel 12-bit DAC; like all 8051-compatible devices, the ADuC812 has separate address spaces for program and data memory, such as 64 KB external program address space and 16 MB external data address space. But unlike other devices, it includes on-chip flash memory technology that provides users with 8 KB of flash/electrically erased program memory and 640 KB of flash/electrically erased data memory; the chip integrates all Auxiliary function blocks that fully support the programmable data acquisition core. These auxiliary function blocks include the Watchdog Timer (WDT), Power Monitor (PSM), and ADCDMA functions. In addition, 32 programmable I/O lines, IZC-compatible SPIs, and standard UART serial ports are available for multiprocessor interfaces and I/O expansion.
In this paper, P1.0, P1.1 and P1.2 of P1 port of ADuC812 MCU are used as three-channel signal input channels. One of the input temperature signals, one input static pressure signal, one way to ground, this way can be used with the corresponding software to reduce temperature drift and system error; P1.7 port is connected to LED, used to monitor whether the MCU works normally: P2 port P2 .0 and P2.1 are used as the input pulse and data terminal of the liquid crystal display; P3.0 (RXD) and P3.1 (TXD) of P3 port are connected to a MAX232, and level conversion is performed to realize communication with the PC. This design uses an external clock generation method. The crystal frequency is 11.059 2 MHz and an internal reference is used to connect a 0.1μF capacitor between the 7-pin (CRER) and AGND. The power reset circuit is reset using the MAX708 chip. The biggest highlight of the hardware design is the simple hardware circuit. The ADuC812 microcontroller does not need external A/D and D/A converters, does not occupy a lot of space, and has reprogrammable non-volatile flash/electrical erase program memory. simple.
2 Selection of multi-sensor data fusion algorithmAt present, there are two main fusion algorithms applied to intelligent pressure sensors, namely multidimensional regression analysis and BP neural network.
2.1 Multidimensional regression analysis
The basic idea of ​​the regression analysis method is to use the multidimensional regression equation to establish the correspondence between the measured target parameters and the sensor output. Different from the one-dimensional experimental calibration/calibration of classical sensors, multi-dimensional calibration/calibration experiments are required, and the coefficients in the regression equation are calculated from the experimental calibration/calibration data by the principle of least squares. Thus, when the sensor output value is measured, the corresponding input target parameter can be calculated from the multidimensional regression equation of the known coefficient. The specific algorithm is that the pressure sensor output voltage U represents pressure information, and the other temperature sensor output voltage Ut represents temperature information, then the pressure parameter P' can be expressed by U and Ut binary functions, namely:
The coefficients a0 to a5 are obtained by the principle of least squares and substituted into the two-dimensional regression equation to determine the binary input-output characteristics of the detected pressure P' and the output U. When the output values ​​U and Ut of the two sensors are collected, the measured parameter P' of the sensor can be calculated in the substitution mode.
2.2 BP neural network method
The connections between neurons are different, and the topology of the network is different. According to the connection between neurons, the neural network structure can be divided into two categories, hierarchical structure and interconnected structure. The hierarchical structure of the neural network divides the neurons into layers, such as the input layer, the middle layer (also called the hidden layer), and the output layer, and the layers are sequentially connected. The input layer neurons are responsible for accepting input information from the outside world and transmitting them to the middle hidden layer neurons; the hidden layer is the internal information processing layer of the neural network, responsible for information transformation; and finally the hidden layer is passed to the information of each neuron in the output layer. After a further processing, the information processing result is output from the output layer to the outside world. Figure 3 shows a single hidden layer BP neural network model with two inputs, each input connected to the next layer by an appropriate weight w, the network output can be expressed as equation (1), and f is the input. /Transfer function of the output relationship.
Application: Automobile Wire Harness
Place of Origin: Dongguan, China (Mainland)
Connector: Molex, JST, TYCO, AMP, JAM, KET,Amphenol, Wago,Weidmuller, Phoenix,
Wires & Cables: UL, VDE standards
Gauge: AWG 28 to AWG 16
Length: Customized
Inspection: 100% inspection before delivery
Certification UL, IATF16949, CE
We can provide first price for products produced in our factory; for sourcing products,we can also provide most competitive price because we have
large relationship among our friend factory.
Automotive Wiring Harness,Auto Wiring Harness,Universal Automotive Wiring Harness,Effect Assurance Cable Wire Harness
Dongguan YAC Electric Co,. LTD. , https://www.yacentercn.com