KR20090102494A - Prognosis system using mems sensors driven by energy harvester - Google Patents

Abstract

PURPOSE: A prognosis system using MEMS sensors driven by an energy harvester is provided to use vibration energy generated in machine equipment as the driving energy of a sensor or a sensor module. CONSTITUTION: A prognosis system using MEMS sensors driven by an energy harvester comprises a MEMS sensor, a wireless sensor unit(409), an energy harvest and a base station(411). The MEMS sensor unites with objects. The wireless sensor unit receives the sensor signal of the analog type from one or more MEMS sensors and signal-processes the sensor signal and converts the sensor signal into the digital signals. The sensor signal converted into the digital signal is produced as packet data. The produced packet data are wirelessly transmitted to the base station. The energy harvest converts the vibration energy of the objects and supplies driving energy to operate one or more among the MEMS sensors and the wireless sensor unit. The base station receives and synthesizes the packet data. The base station analyzes the synthesized packet data and drives prognosis diagnosis based on the analyzed data and produces the prognosis diagnosis result.

Description

Prognosis system using MEMS sensors driven by energy harvester

The present invention relates to a prognostic diagnosis system using MEMS sensors driven by energy harvesters.

Conventionally, a method of diagnosing a state of a mechanical facility has been used for attaching various sensors to a mechanical facility for maintenance of a mechanical facility, and then analyzing a sensor signal generated from the sensor by an expert. Even in the conventional diagnostic method, the number of experts is small compared to the number of mechanical facilities, and the sensors are relatively expensive, and since many sensors are difficult to be installed in the equipment, there is a disadvantage that maintenance of efficient mechanical equipment is difficult.

Recently, in the diagnosis of mechanical equipment, a diagnostic system that can replace an expert has been used. However, based on sensor information about the current state of the machine, there are scientific techniques that can accurately predict future progress and predict the lead time or residual useful life to the next planned maintenance time. It is not generalized and therefore is not applied to industrial sites.

1 is a view schematically showing the configuration of a diagnostic system for the maintenance of the conventional mechanical equipment.

Referring to FIG. 1, in a conventional diagnosis system, at least one expensive acceleration sensor 103 and at least one expensive temperature sensor coupled to an object 101, such as a mechanical facility, for diagnosing a mechanical facility. 105, various sensors such as at least one clamp current sensor 107, a junction box 109, a signal processing device 111, a display device 113, etc. for integrating and branching the sensors are required. In the conventional automatic diagnosis system, various kinds of sensors are required to precisely monitor the state of a mechanical equipment. However, since these sensors are expensive, they are applicable to only expensive mechanical equipment. If all or most of the mechanical equipment can be managed by an automatic diagnostic system, the use of low-cost sensors is essential, because more reliable maintenance is possible. In addition, in the conventional diagnostic system, since various sensors, the junction box 109, the signal processing device 111, the display device 113, and the like are connected by wires, it is difficult to secure flexible mobility of the devices. There is a problem. In particular, when obstacles exist between the devices, a situation in which an acceleration signal cannot be obtained may result. In addition, in the conventional diagnostic system, only the diagnosis of the current state of the machine equipment is performed, but the fundamental problem that the prognosis of the future state of the machine equipment (for example, the life of the machine equipment) cannot be performed in real time. have. In addition, in the conventional diagnostic system, since a separate power source (for example, a battery, etc.) is required to drive various sensors and sensor modules for processing sensor signals, it is cumbersome to periodically replace the batteries. .

Disclosure of Invention The present invention has been made to overcome the problems of the prior art, and an object of the present invention is to provide a prognostic diagnosis system using a MEMS sensor driven by an energy harvester capable of prognosing a future condition such as the remaining useful life of a mechanical facility in real time. It is done.

Another object of the present invention is to provide a prognostic diagnosis system using a MEMS sensor driven by an energy harvester that can use vibration energy generated in a mechanical facility as a driving energy of a sensor or a sensor module.

Another object of the present invention is to provide a prognostic diagnosis system using a MEMS sensor driven by an energy harvester capable of automatically prognosing a future state of a mechanical facility using a low-cost MEMS sensor.

It is another object of the present invention to provide a prognostic diagnosis system using a MEMS sensor driven by an energy harvester capable of automatically prognosing a future state of a mechanical facility using an SVR algorithm based on a sensor signal from a low-cost MEMS sensor. To provide.

Another object of the present invention is to wirelessly transmit and receive the sensor signal from the MEMS sensor coupled to the mechanical equipment in real time to automatically prognostic the future state of the mechanical equipment, to prevent economic loss or loss of life due to shutdown or failure It is to provide a prognostic diagnosis system using MEMS sensors driven by energy harvesters that can minimize and minimize maintenance costs and maximize economic benefits or energy efficiency.

According to an aspect of the present invention to achieve the above object, by receiving at least one MEMS (Micro-Electro Mechanical System) sensor coupled to the object, the sensor signal of the analog form from the at least one MEMS sensor, After the processing, converts into a digital signal, generates a sensor signal converted into a digital signal form the packet data, a wireless sensor unit for transmitting the generated packet data to the base station by wireless, by converting the vibration energy of the object Receives and combines an energy harvester and packet data for supplying driving energy for driving at least one of the at least one MEMS sensor and the wireless sensor unit, analyzes the synthesized packet data, and analyzes the analyzed data. Based on the prognostic diagnosis algorithm, the prognostic result is generated and reported. A MEMS sensor that is powered by the energy harvester including a device station can provide prognostic diagnosis system using.

In a preferred embodiment, the object is characterized in that the machine. The MEMS sensor may be at least one of a MEMS acceleration sensor, a MEMS temperature sensor, and a MEMS current sensor. The display apparatus may further include a display device for outputting the prognostic diagnosis result. In addition, the display device may include an alarm device corresponding to the prognostic diagnosis result. The wireless sensor unit also includes a high pass filter, an amplifier, and a low pass filter, wherein the signal processing includes the high pass filter, the amplifier, and the low pass filter. (Low Pass Filter) characterized in that performed by. The wireless sensor unit may include an analog / digital converter, and the conversion to the digital signal is performed by the analog / digital converter. In addition, the wireless sensor unit includes a wireless LAN bridge (Wireless LAN Bridge), characterized in that the wireless transmission of the packet data is performed by a wireless LAN bridge (Wireless LAN Bridge). In addition, the driving energy is characterized in that the electrical energy. The energy harvester may include a piezoelectric element in which displacement occurs in response to the vibration of the object, and a magnetic generator for generating the driving energy in response to the displacement of the piezoelectric element. The energy harvester may further include a length adjusting plate for adjusting the resonance region of the piezoelectric element, a micrometer for moving the length adjusting plate, and a magnet as an additional mass for increasing the resonance frequency of the piezoelectric element. In addition, the energy harvester is characterized in that it further comprises a battery for storing the drive energy. In addition, the prognostic diagnosis algorithm is characterized in that the support vector regression (SVR) algorithm. In addition, the prognostic diagnosis result is a life expectancy corresponding to the object.

According to another aspect of the present invention, at least one MEMS (Micro-Electro Mechanical System) sensor coupled to the object, a sensor module (Sensor Module) for signal processing the sensor signal of the analog form received from the at least one MEMS sensor and A wireless sensor unit converts the analog sensor signal into a digital signal and includes an interface module for wirelessly transmitting a packet signal corresponding to the sensor signal converted into a digital signal to a base station. An energy harvester for converting vibration energy to supply driving energy for driving at least one of the at least one MEMS sensor and the wireless sensor unit, and a server module configured to receive and synthesize the packet signal; Analysis data by analyzing the synthesis data generated by the synthesis Energy including a base station including a data analysis module to generate and a solver module for generating and reporting a prognostic diagnosis result corresponding to the object through a prognostic diagnosis algorithm based on the analysis data. Prognostic diagnosis system using MEMS sensors driven by harvesters can be provided.

In a preferred embodiment, the object is characterized in that the machine. The MEMS sensor may be at least one of a MEMS acceleration sensor, a MEMS temperature sensor, and a MEMS current sensor. The display apparatus may further include a display device for outputting the prognostic diagnosis result. In addition, the display device may include an alarm device corresponding to the prognostic diagnosis result. The sensor module may also include a mux, a high pass filter, an amplifier, and a low pass filter. In addition, the interface module may include an analog / digital converter, a memory, a wireless LAN bridge, a central processing unit (CPU), and an auxiliary storage device. In addition, the driving energy is characterized in that the electrical energy. The energy harvester may include a piezoelectric element in which displacement occurs in response to the vibration of the object, and a magnetic generator for generating the driving energy in response to the displacement of the piezoelectric element. The energy harvester may further include a length adjusting plate for adjusting the resonance region of the piezoelectric element, a micrometer for moving the length adjusting plate, and a magnet as an additional mass for increasing the resonance frequency of the piezoelectric element. In addition, the energy harvester is characterized in that it further comprises a battery for storing the drive energy. In addition, the prognostic diagnosis algorithm is characterized in that the support vector regression (SVR) algorithm. In addition, the prognostic diagnosis result is a life expectancy corresponding to the object.

According to the present invention, it is possible to provide a prognostic diagnosis system using a MEMS sensor driven by an energy harvester capable of prognosing a future condition such as the life of a mechanical facility in real time.

In addition, according to the present invention, it is possible to provide a prognostic diagnosis system using a MEMS sensor driven by an energy harvester that can use vibration energy generated in a mechanical facility as a driving energy of a sensor or a sensor module.

In addition, according to the present invention, it is possible to provide a prognostic diagnosis system using a MEMS sensor driven by an energy harvester capable of automatically prognosing a future state of a mechanical facility using a low-cost MEMS sensor.

Also, according to the present invention, a prognostic diagnosis system using a MEMS sensor driven by an energy harvester capable of automatically prognosing a future state of a mechanical facility using an SVR algorithm based on a sensor signal from a low-cost MEMS sensor can be provided. In addition, the present invention wirelessly transmits and receives a sensor signal from a MEMS sensor coupled to a mechanical facility in real time, and automatically prognoses the future state of the mechanical facility, thereby preventing economic loss and loss of life due to an operation stop or failure. It is possible to provide a prognostic diagnosis system using MEMS sensors driven by an energy harvester that can maximize the economic benefits or the energy efficiency through minimizing or minimizing the maintenance costs and reducing maintenance costs.

1 is a view schematically showing the configuration of a diagnostic system for the maintenance of conventional mechanical equipment.

2 is a view showing a capacitor type accelerometer according to the MEMS process combining a general semiconductor process (IC process) and a micromachining process (Micromachining Process).

3A and 3B are diagrams illustrating piezoresistive accelerometers according to a MEMS process combining a general semiconductor process and a micromachining process.

4 is a diagram schematically showing the configuration of a prognostic diagnosis system using MEMS sensors driven by an energy harvester according to a preferred embodiment of the present invention.

5 is a view showing in detail the configuration of the prognostic diagnosis system using the MEMS sensor driven by the energy harvester according to a preferred embodiment of the present invention.

6 is a view schematically showing the configuration of an energy harvester according to a preferred embodiment of the present invention.

7 is a schematic representation of a prognostic diagnosis algorithm in accordance with a preferred embodiment of the present invention.

8 is a graph showing an example of a prognostic diagnosis result according to a preferred embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings, the configuration and operation of the embodiment of the present invention will be described in detail. The drawings and the description of the drawings are examples of the present invention, which does not limit the scope of the present invention.

According to a preferred embodiment of the present invention, a system for diagnosing a future state of a mechanical facility employs a MEMS sensor as a low-cost sensor. The MEMS sensor, in particular the MEMS acceleration sensor, will be described in detail.

MEMS (Micro-Electro Mechanical System) is used interchangeably as a micro system, a micro machine, and a micro mechatronics, and means a micro system or a micro machine. MEMS can be manufactured in the size of several micrometers to several hundred micrometers by using the semiconductor manufacturing process, mass production by the batch process, low power consumption, and a single chip for mechanical parts, sensors, electronic circuits, etc. For this reason, active research and industrial application are currently in progress for this reason.

Looking at the development of MEMS technology, in the 1960s, micromechanical elements such as valves, motors, and gears were fabricated on a silicon substrate in a two-dimensional plane. In the 1970s, various devices using anisotropic etching and optical devices of three-dimensional structure using the same, inkjet, holography, and the like were studied. In the 1980s, research on thin-dimensional three-dimensional structures in which a sacrificial thin film deposited on a substrate was etched without etching the substrate using surface micromachining technology was conducted. After that, various fabrication technologies such as LIGA (LIthography, Galvanoformung / electroplating, Abformung / plastic modeling), laser, and electric discharge have been developed. Thus, MEMS technology has been steadily developed.

MEMS technology is being actively applied to various sensors to replace human sense organs. Especially, collision detection sensors and pressure sensors for automobile airbags are world-class products. Among various sensors using MEMS technology, the acceleration sensor (Accelerometer) is frequently used. In this regard, FIG. 2 is a capacitor type accelerometer according to the MEMS process combining a general semiconductor process (IC process) and a micromachining process (Micromachining process), and FIG. 3A and FIG. Piezoresistive acceleration sensors (Piezoresistive Accelerometer) according to the MEMS process that combines is shown.

4 is a view schematically showing the configuration of a prognostic diagnosis system using a MEMS sensor driven by an energy harvester according to a preferred embodiment of the present invention.

4, a prognostic diagnosis system using MEMS sensors driven by an energy harvester according to the present invention includes at least one MEMS acceleration sensor 403 and at least one MEMS coupled to an object 401 such as a mechanical facility. And a temperature sensor 405, at least one MEMS current sensor 407, a wireless sensor unit 409, a base station 411, and a display device 413. The at least one MEMS acceleration sensor 403, the at least one MEMS temperature sensor 405, the at least one MEMS current sensor 407, and the wireless sensor unit 409 may convert vibration energy generated from a mechanical facility into electrical energy or the like. It can be driven by receiving energy from an energy harvester (415) to convert to form. The energy harvester 415 will be described in more detail with reference to FIG. 6. For convenience of description, the prognostic diagnosis system using the MEMS sensor driven by the energy harvester 415 according to the present invention includes the MEMS acceleration sensor 403, the MEMS temperature sensor 405, and the MEMS current sensor 407, but It will be apparent to those skilled in the art that the present invention may include a kind of sensor.

Each of the at least one MEMS acceleration sensor 403, the at least one MEMS temperature sensor 405, and the at least one MEMS current sensor 407 coupled to an object 401, such as a machine, generates a sensor signal, which It passes to the wireless sensor unit 409.

The wireless sensor unit 409 processes the sensor signal, which is an analog signal obtained from the MEMS sensor, using an amplifier, a low pass filter, and a high pass filter. Thereafter, the wireless sensor unit 409 converts the analog sensor signal processed by using an analog / digital converter (Analog / Digital, A / D Converter) into a digital signal, and generates packet data that is bundled into packets. The packet data is transmitted to the base station 411 based on a wireless communication protocol, particularly TCP / IP, using a wireless LAN bridge. Here, the packet data may be directly transmitted to the base station 411 by wireless, or may be transmitted to the base station 411 through wireless LAN communication using an access point (AP).

The base station 411 synthesizes the received packet data using the server module, analyzes the synthesized packet data using the data analysis module, and uses the solver module to support SVR (Support Vector Regression) based on the analyzed data. The algorithm performs the function of generating and reporting the prognostic diagnosis result for the future state (eg, the life of the machine) of the machine.

The display device 413 outputs a prognostic diagnosis result generated by the base station 411. The display device 413 may further include various alarms that may generate a warning signal according to a prognostic diagnosis result.

5 is a view showing in detail the configuration of the prognostic diagnosis system using the MEMS sensor driven by the energy harvester according to a preferred embodiment of the present invention.

Referring to FIG. 5, the wireless sensor unit 409 according to the present invention includes a sensor module 501, an interface module 503, a low-cost at least one MEMS acceleration sensor 403, a low-cost at least An energy harvester for powering one MEMS temperature sensor 405, at least one low-cost MEMS current sensor 407, and a wireless sensor unit 409 including the sensor module 501 and the interface module 503. (Energy Harvester, 505). In addition, the base station 411 according to the present invention includes a server module 527, a data analysis module 529, and a solver module 531.

The sensor module 501 includes a mux 507, a high pass filter 509 having a cutoff frequency of 0.7 Hz, an amplifier 511 for amplifying a signal, and an eighth order low band for limiting the bandwidth. A low pass filter 513. The sensor signals obtained from the low cost at least one MEMS acceleration sensor 403, the low cost at least one MEMS temperature sensor 405, and the low cost at least one MEMS current sensor 407 are high pass through the mux 507. It is input to the pass filter 509, the amplifier 511, and the low pass filter 513, and is signal-processed. The low pass filter 513 passes an acceleration signal below a set maximum frequency and is designed in consideration of anti-aliasing, and the high pass filter 509 is designed to filter a frequency band of 0.7 Hz or more. . The amplifier 511 has a gain amplification function of 0 dB, 10 dB, 20 dB and 30 dB.

The interface module 503 may include an analog / digital converter (Analog / Digital, A / D Converter, 517), an auxiliary storage device (for example, an SD memory card 519), a memory 521, and a wireless LAN bridge. LAN Bridge 525) and a central processing unit (CPU) 515. The sensor signal processed from the sensor module 501 is input to the interface module 503, and the input analog sensor signal is converted into a digital signal by the A / D converter 517 and then the memory ( 521 is used to form packet data 523 bundled into packets. The sensor signal converted into a digital signal by the A / D converter 517 may be separately stored in the auxiliary memory device (for example, the SD memory card 519). Thereafter, the packet data 523 is transmitted to the base station 411 through the wireless LAN bridge 525 based on a wireless communication protocol, in particular TCP / IP. The packet data 523 may be directly transmitted to the base station 411 by wireless, or may be transmitted to the base station 411 through wireless LAN communication using an access pointer (AP). The CPU 515 controls each component of the sensor module 501 and each component of the interface module 503 to adjust the amplification ratio, A / D conversion, memory storage, network monitoring, data restoration, Perform packet data conversion and data management functions.

The server module 527 includes means for wireless LAN communication with the interface module 503, and performs a function of receiving and synthesizing packet data received from the interface module 503. Thereafter, the server module 527 transfers the synthesized data to the data analysis module 529. The data analysis module 529 includes sensor identification, configuration information, data, time frequency display, and means for storing all information that needs to be stored in the sensor and corresponds to sensor signals transmitted from the server module 527. It analyzes synthesized data. The solver module 531 uses the SVR (Support Vector Regression) algorithm, which is a prognostic diagnosis algorithm, based on the analyzed sensor data and the time data from the data analysis module 529 (for example, the mechanical equipment). It generates prognostic results of the prognosis, and reports them. The SVR algorithm will be described in more detail with reference to FIG. 7.

6 is a view schematically showing the configuration of an energy harvester according to a preferred embodiment of the present invention.

Referring to FIG. 6, an energy harvester converts vibration energy generated from a mechanical device into other energy such as electric energy, and converts the converted energy (for example, electric energy) into various sensors and wireless sensor units. Supply. That is, the energy harvester may convert the converted energy into a low-cost at least one MEMS acceleration sensor, a low-cost at least one MEMS temperature sensor, a low-cost at least one MEMS current sensor, and a wireless sensor unit including the sensor module and the interface module. Supply.

The energy harvester includes a micrometer 601, a length control plate 603, a piezoelectric element 605, a magnet 607 and a coil 609. The piezoelectric element 605 is a member that collects vibration energy of a mechanical facility, and displacement of the piezoelectric element is generated by vibration of the mechanical facility. It is effective to adjust the resonance region of the piezoelectric element in accordance with the resonance region when the machine equipment vibrates. The length adjusting plate 603 is moved using the micrometer 601 to adjust the resonance region of the piezoelectric element. In addition, when the resonant frequency of the piezoelectric element is high, the energy generation efficiency is increased. In order to increase the resonant frequency of the piezoelectric element, the magnet 607 is used as an additional mass. In order to maximize the energy collection efficiency of the energy harvester, the coil 609 is used as a magnetic generator that generates electrical energy by using the displacement of the piezoelectric element. Electrical energy generated by the piezoelectric element 605 and the coil 609 is stored in a battery (not shown) and supplied to various sensors and wireless sensor units.

7 schematically illustrates a prognostic diagnosis algorithm according to a preferred embodiment of the present invention.

Referring to FIG. 7, a support vector regression (SVR) algorithm, which is a prognostic diagnosis algorithm, may be divided into a data level and a knowledge level. The solver module performs the data acquisition process at the data level and performs data distribution, prediction, and prognostic diagnosis at the knowledge level.

In the data level data acquisition process, trend data having a history until a defect occurs is used. First, the solver module receives the analyzed sensor data from the data analysis module (step 701). The solver module then stores the received analyzed sensor data in a database (step 703). Thereafter, the solver module calculates the feature value from the sensor data stored in the database and extracts the feature value which is the trend data (steps 705 and 707).

At the knowledge level, the solver module distributes the trend data, which is the extracted feature values, to training data, verification data and test data (step 709). The training data and verification data are used to generate a machine failure prognosis diagnostic model (especially life prediction model), and the test data is used to test the verified machine failure prognosis diagnostic model (especially life prediction model). If the test results are valid, then the tested mechanical failure prognostic model (especially life prediction model) can be used. The solver module generates a mechanical failure prognostic model (especially a life prediction model) based on the training data and the validation data (step 711). The solver module then predicts future data (eg, life data) based on the generated mechanical failure prognostic model (especially life prediction model) (step 713). The solver module evaluates the performance based on the root-means square error (RMS) and the correlation coefficient R (step 715).

The RMSE provides the accuracy of global prognostic diagnosis, and is represented by Equation 1 below.

Where N is the total number of data points in the test set, y is the observed value and y ^ is the prognostic value.

The correlation coefficient R measures a linear correlation between the prognostic value and the actual value, as shown in Equation 2 below.

Where Cov (y, y ^) represents the cross-correlation between the prognosed and actual values (see Equation 3), Is the standard deviation of the actual value (see equation 4), Denotes the standard deviation of the predicted value (see Equation 5).

here, Is the mean of the actual values, Represents the mean of the prognostic values.

The solver module then reports the final prognostic result (step 717).

8 is a graph showing an example of a prognostic diagnosis result according to a preferred embodiment of the present invention.

Referring to FIG. 8, information on the remaining useful life of the target machine may be displayed until a defect of the target machine occurs. The prognostic diagnosis result may be output to a display device coupled to the base station. According to another embodiment of the present invention, when the remaining useful life is within a predetermined period, various alarms included in the display device may generate a warning signal.

The present invention is not limited to the above embodiments, and many variations are possible by those skilled in the art within the spirit of the present invention.

Claims (27)

At least one MEMS (Micro-Electro Mechanical System) sensor coupled to the object; Receives an analog sensor signal from the at least one MEMS sensor, performs signal processing, converts the signal into a digital signal, generates a sensor signal converted into a digital signal form as packet data, and wirelessly generates the generated packet data. A wireless sensor unit for transmitting to the base station; An energy harvester for converting vibration energy of the object to supply driving energy for driving at least one of the at least one MEMS sensor and the wireless sensor unit; And A base station that receives and synthesizes packet data, analyzes the synthesized packet data, and runs a prognostic diagnosis algorithm based on the analyzed data to generate and report prognostic results. Prognostic diagnosis system using a MEMS sensor driven by the energy harvester comprising a. The method of claim 1, The object is a prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that the mechanical equipment. The method of claim 1, The MEMS sensor is a prognostic diagnosis system using a MEMS sensor driven by an energy harvester, characterized in that at least one of MEMS acceleration sensor, MEMS temperature sensor and MEMS current sensor. The method of claim 1, Display device for outputting the prognostic diagnosis results Prognostic diagnosis system using a MEMS sensor driven by the energy harvester further comprising. The method of claim 4, wherein The display device is a prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that it comprises an alarm device corresponding to the prognostic result. The method of claim 1, The wireless sensor unit includes a high pass filter, an amplifier, and a low pass filter, The signal processing is a prognostic diagnosis system using a MEMS sensor driven by an energy harvester, characterized in that performed by the high pass filter, the amplifier and the low pass filter. The method of claim 1, The wireless sensor unit includes an A / D converter (Analog / Digital Converter), The conversion into the digital signal is a prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that performed by the A / D converter (Analog / Digital Converter). The method of claim 1, The wireless sensor unit includes a wireless LAN bridge, Wireless transmission of the packet data is a prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that performed by a wireless LAN bridge (Wireless LAN Bridge). The method of claim 1, The driving energy is a prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that the electrical energy. The method of claim 1, The energy harvester is A piezoelectric element in which displacement occurs in response to vibration of the object; And Magnetic generator for generating the driving energy in response to the displacement of the piezoelectric element Prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that it comprises a. The method of claim 10, The energy harvester is A length adjusting plate for adjusting a resonance region of the piezoelectric element; A micrometer for moving the length control plate; And Magnet as an added mass for raising the resonance frequency of the piezoelectric element Prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that it further comprises. The method of claim 11, The energy harvester is A battery for storing the driving energy Prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that it further comprises. The method of claim 1, The prognostic diagnosis algorithm is a prognostic diagnosis system using a MEMS sensor driven by an energy harvester, characterized in that the support vector regression (SVR) algorithm. The method of claim 1, The prognosis diagnosis result is a prognostic diagnosis system using a MEMS sensor driven by an energy harvester, characterized in that the life prediction corresponding to the object. At least one MEMS (Micro-Electro Mechanical System) sensor coupled to the object; A sensor module for processing an analog sensor signal received from the at least one MEMS sensor and a packet corresponding to the sensor signal converted into a digital signal, converting the analog sensor signal into a digital signal. A wireless sensor unit including an interface module for wirelessly transmitting a signal to a base station; An energy harvester for converting vibration energy of the object to supply driving energy for driving at least one of the at least one MEMS sensor and the wireless sensor unit; And A server module for receiving and synthesizing the packet signal, a data analysis module for generating analysis data by analyzing the synthesis data generated by the synthesis, and a prognostic diagnosis algorithm based on the analysis data. Base station including a solver module (Solver Module) for generating and reporting the prognostic diagnosis results corresponding to the target through Prognostic diagnosis system using a MEMS sensor driven by the energy harvester comprising a. The method of claim 15, The object is a prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that the mechanical equipment. The method of claim 15, The MEMS sensor is a prognostic diagnosis system using a MEMS sensor driven by an energy harvester, characterized in that at least one of MEMS acceleration sensor, MEMS temperature sensor and MEMS current sensor. The method of claim 15, Display device for outputting the prognostic diagnosis results Prognostic diagnosis system using a MEMS sensor driven by the energy harvester further comprising. The method of claim 18, The display device is a prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that it comprises an alarm device corresponding to the prognostic result. The method of claim 15, The sensor module includes a mux, a high pass filter, a high pass filter, an amplifier, and a low pass filter. A prognostic diagnosis system using a MEMS sensor driven by an energy harvester. The method of claim 15, The interface module includes an A / D converter (Analog / Digital Converter), a memory, a wireless LAN bridge, a central processing unit (CPU), and an auxiliary memory device. Prognostic diagnosis system using sensors. The method of claim 15, The driving energy is a prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that the electrical energy. The method of claim 15, The energy harvester is A piezoelectric element in which displacement occurs in response to vibration of the object; And Magnetic generator for generating the driving energy in response to the displacement of the piezoelectric element Prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that it comprises a. The method of claim 23, wherein The energy harvester is A length adjusting plate for adjusting a resonance region of the piezoelectric element; A micrometer for moving the length control plate; And Magnet as an added mass for raising the resonance frequency of the piezoelectric element Prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that it further comprises. The method of claim 24, The energy harvester is A battery for storing the driving energy Prognostic diagnosis system using a MEMS sensor driven by the energy harvester, characterized in that it further comprises. The method of claim 15, The prognostic diagnosis algorithm is a prognostic diagnosis system using a MEMS sensor driven by an energy harvester, characterized in that the support vector regression (SVR) algorithm. The method of claim 15, The prognosis diagnosis result is a prognostic diagnosis system using a MEMS sensor driven by an energy harvester, characterized in that the life prediction corresponding to the object.