CN104618504A - Large-scale hydraulic project crack and extension joint variation quantitative monitoring system and predication method thereof - Google Patents

Abstract

The invention discloses a large-scale hydraulic project crack and extension joint variation quantitative monitoring system and a predication method thereof. The monitoring system is formed by a monitoring node, a gateway node and a monitoring center electronic computer; the monitoring node is a ZigBee wireless sensor network node; the monitoring node transmits image information of a hydraulic project crack and an extension joint to the monitoring center electronic computer in real time through the gateway node, the GPRS network and the Internet network. The large-scale hydraulic project crack and extension joint variation quantitative monitoring system monitors the hydraulic engineering crack and the extension joint due to the camera sensor, is free of contact, can increase the monitoring system reliability and improves the monitoring accuracy; is particularly suitable for safety monitoring of a large-scale hydraulic project; does not need to pave a large amount of leads on the hydraulic project, saves costs, reduces construction difficulties and saves a large amount of manpower resources; allows 24-hour monitoring through the infrared camera sensor without being influenced by the weather.

Description

A kind of large hydraulic engineering crack and expansion joint change Quantitative Monitoring system and Forecasting Methodology thereof

Technical field

The invention belongs to the safety testing field of large hydraulic engineering (as dam, key water control project), particularly a kind of large hydraulic engineering crack and expansion joint change Quantitative Monitoring system and Forecasting Methodology thereof.

Background technology

Its status quo and development trend of periodic monitoring is answered to the crack in hydraulic engineering and expansion joint, to analyze the producing cause of its change and the impact on hydraulic engineering safety thereof, and effectively administers in time.The technology of current employing, earth dam internal fissure monitoring method as mixed in Chinese patent CN103512545A mono-kind and monitoring technology, be all different monitor of two ranges that is arranged in parallel in concrete dam, when in the scope that Size of Crack appears at instrument setting, Fracture Deformation value can be recorded.Medium and small reservoirs earth dam safety monitoring system and forecast analysis (Han Zhe, Hunan University, Master's thesis, 2011.4) by can only transducer, the integrated use of power supply automated management system and point-to-point real-time network Digital Transmission three, and making software system, carry out dynamic analysis and safety evaluation.

Based on the Dam safety Monitoring System (Li Peng etc. of ARM9, instrument and meter for automation, 2010,31 (11), P47-50) for the problem that the safety monitoring system automation of medium waterpower generator station and reservoir dam is not high, the network monitoring system based on ARM9 embedded system is devised.

In sum, the effect of guarantee property is played in the safety monitoring of prior art to hydraulic engineering, but also have some limitations, as, in earth dam, bury transducer underground, power supply all will regularly replace, the circuit of field monitoring is very long, be erected at field again, the transducer exposed is subject to the destruction etc. of external environment, brings certain hidden danger and disturbance factor to water conservancy engineering monitoring.

Summary of the invention

In order to overcome the defect of prior art, the present invention adopts radio sensing network and image processing techniques, and is aided with certain Forecasting Methodology, realizes the crack of large hydraulic engineering and expansion joint and changes and carry out Quantitative Monitoring and trend prediction thereof.Concrete technical scheme is as follows:

A kind of large hydraulic engineering crack and expansion joint change Quantitative Monitoring system, by monitoring node, gateway node, monitoring center's electronic computer composition, monitoring node is the ZigBee radio sensing network node with image sensor, the image information at hydraulic engineering crack and expansion joint is real-time transmitted in monitoring center's electronic computer through gateway node by GPRS network and Internet network by monitoring node, dedicated computing software is on computers installed, quantitatively calculate monitored hydraulic engineering crack and expansion joint changing value, when changing value exceeds defined threshold, computer shows early warning automatically, early warning is sent on operating personnel's mobile phone simultaneously, thus the development trend at hydraulic engineering crack and expansion joint is predicted.

Further, monitoring node forms by with lower component: image sensor, signal conditioning circuit, radiofrequency emitting module based on ZigBee communication agreement, wherein image sensor is connected with the radiofrequency emitting module based on ZigBee communication agreement by signal conditioning circuit, and be all connected with photoelectric source, photoelectric source is connected with power management chip, and unified control is powered; Radiofrequency emitting module based on ZigBee communication agreement is connected with liquid crystal display and USB interface.

Preferably, image sensor is SAT series of IR detector.

Preferably, the radiofrequency emitting module based on ZigBee communication agreement is CC2430 chip.

Further, webmaster node is by with lower component: the radio frequency based on ZigBee communication agreement accepts module, microprocessor, SIM card holder, GPRS module are formed by connecting in turn, and they are powered by wind-solar power supply, and to be sought unity of action power supply management by power management chip.

Preferably, AT91RM9200 chip selected by microprocessor.

Large hydraulic engineering crack and expansion joint change a Forecasting Methodology for Quantitative Monitoring system, use gray prediction method, open up the geomery parameter X that grey forecasting model makes full use of crack and expansion joint ₁, X ₂, X ₃cumulative for time dependent size summation, generate new sequence, approach with a curve, be after approximating curve reduction and open up grey forecasting model.

Beneficial effect

The present invention compared with prior art has following good effect: (1) adopts camera sensing device, monitors hydraulic engineering crack and expansion joint, is a kind of contactless, can increases the reliability of monitoring system, improve monitoring accuracy.(2) adopt ZigBee-network agreement, due to network configuration flexibly, the network capacity of super large, is particularly suitable for the safety monitoring of large hydraulic engineering.(3) radio sensing network is adopted, without the need to laying a large amount of wire in hydraulic engineering, cost-saving, reduce difficulty of construction.(4) without the need to the regular tours of inspection monitoring system of staff, a large amount of human cost is saved.(5) ZigBee-network, GPRS network and Internet network are combined, be applicable to safety monitoring that is large-scale and ultra-large type hydraulic engineering.(6) infrared photography transducer is adopted can not to implement monitoring in 24 hours by weather effect.

Accompanying drawing explanation

Fig. 1 is that a kind of large hydraulic engineering crack and expansion joint change Quantitative Monitoring system;

Fig. 2 is ZigBee-network monitoring node schematic diagram;

Fig. 3 is ZigBee-network gateway node schematic diagram.

Wherein, 1. monitoring node; 2. gateway node; 3.GPRS network; 4.Internet network; 5. monitoring center's computer; 6. image sensor; 7. signal conditioning circuit; 8. based on radio-frequency transmissions (reception) module of ZigBee communication agreement; 9 liquid crystal display; 10.USB interface; 11. photoelectric sources; 12. power management chips; 13. microprocessors; 14. wind-solar power supply; 15.SIM deck; 16.GPRS module.

Embodiment

For making object of the present invention and technical scheme clearly, be clearly and completely described below in conjunction with the technical scheme of accompanying drawing to the application.Described embodiment is a part of embodiment of the application; instead of whole embodiments; based on the embodiment of described the application, the every other embodiment that those of ordinary skill in the art obtain under without the need to the prerequisite of creative work, all belongs to the scope of the application's protection.

Embodiment 1: a kind of large hydraulic engineering crack and expansion joint change Quantitative Monitoring system, by monitoring node 1, gateway node 2, monitoring center's electronic computer 5 forms, it is characterized in that, monitoring node 1 is the ZigBee radio sensing network node with image sensor 6, the image information at hydraulic engineering crack and expansion joint is real-time transmitted in monitoring center's electronic computer 5 through gateway node 2 by GPRS network 3 and Internet network 4 by monitoring node 1, automatically each concrete the hydraulic engineering crack of position and the changing value at expansion joint really of monitoring is shown on the computer screen, when changing value exceeds defined threshold, computer shows early warning automatically, early warning is sent on operating personnel's mobile phone simultaneously, calculate the development trend at function prediction hydraulic engineering crack and expansion joint.

(1) monitoring node

Monitoring node 1 as shown in Figure 2, infrared photography transducer 6 will aim at the crack and expansion joint that will monitor in hydraulic engineering, by the image of production after signal conditioning circuit 7 processes, launch image information by the radio-frequency module 8 based on ZigBee communication agreement.Monitoring node 1 photo-voltaic power supply 11.Each chip model of monitoring node 1 can do following selection: CC2430 is the chip system being used for realizing embedded ZigBee application that Chipcon company produces, and its supports 2.4GHzIEEE802.15.4/ZigBee agreement.CC2430 chip is using powerful Integrated Development Environment as support, the interactive mode debugging of internal wiring is to defer to the IAR industrial standard of IDE for supporting, obtain the height accreditation of embedded mechanism, CC2430 chip system module integration CC2420RF transceiver, strengthen the 8051MCU.32/64/128KB flash memory of industrial standard, the contour performance module of 8KB SRAM, and built-in Zigbee protocol, add super low energy consumption, make it to use very low expenditure pattern ZigBee node.

Infrared photography transducer 6 adopts SAT series of IR video camera, it all adopts non-brake method Jiaozhuo plane Infrared Detectors, the one chip resistor-type microbolometer technology that this Infrared Detectors adopts polycrystalline silicon material to prepare, detector column outmoded conventions mould 320 × 240, pixel centre-to-centre spacing 45 μMs, fill factor, curve factor is greater than 80%, and noise equivalent temperature difference (NETD) reaches 100mk (representative value).

In order to adapt to the actual needs of hydraulic engineering safety monitoring, energy savings, monitoring node 1 applies photo-voltaic power supply, and the photo-voltaic power supply model selected is SAS2.5-WED.

(2) gateway node

Because Zigbee protocol is according to IEEE802.15.4 standard, mutually coordinate to realize communication between thousands of small monitoring node 1, the energy that these monitoring nodes 1 need are little, by radio wave, data are delivered to another monitoring node 1 from a monitoring node 1 in the mode of relay, their communication efficiencies are very high, but also illustrate that the low data rate of ZigBee technology and the less feature of communication range simultaneously.Therefore, when large hydraulic engineering safety monitoring system monitoring node information is transmitted, gateway node 2 must be set, monitoring node 1 information can be made to transfer on the electronic computer 5 of monitoring center smoothly, the schematic diagram of gateway node 2 as shown in Figure 3, its course of work is as follows: by receiving the image information that monitoring node is sent based on the Receiver Module 8 of ZigBee communication agreement, denoising is carried out by microprocessor 13 pairs of image files, after compression process, launched by GPRS module 16 again, be coupled with Internet network 4 through GPRS network 3, monitoring center's electronic computer 5 can receive the image that monitoring node 1 is produced.

Microprocessor 13 selects the high speed arm processor AT91RM9200 of a embedded 32 ARM920T cores of Atmel company as center processor, there is high-performance, low-power consumption, low cost feature, its instruction process speed can reach 200MI/s (million VAX Instructions Per Second VAXs), the high-speed transfer requirement of monitoring net artis 2 can be met, it is again a technical grade microprocessor simultaneously, gateway node 2 can be applicable to work the severe requirement in looped network border, ensure the stability that gateway node 2 works.AT91RM9200 can transplant the (SuSE) Linux OS of standard simultaneously, decrease the development difficulty of webmaster node 2 software, and enhance its portability, be conducive to the secondary development of software.

Wireless launcher comprises sim card socket 15, GPRS module 16 and antenna, and the input of sim card socket 15 connects microprocessor 13, and output connects GPRS module 16.GPRS module 16 selects Mc55 module, and the built-in ICP/IP protocol of Mc55 module controls service routine by AT instruction and is easy to access network.

Wind-solar power supply 14 selects wind and light complementary power supply, and model is SDC-DMI150.

Carry out in microprocessor 13 and carry out preliminary treatment and compress transmitting to image file.

Image denoising application self-adapting medium filtering (AMF), principle is as follows: establish S _xyrepresent and make Z by the mask window that central pixel point (x, y) is corresponding when filtering _minfor S _xymiddle gray scale minimum value, Z _maxfor S _xymiddle gray scale maximum, Z _medfor S _xymiddle gray scale intermediate value, Z _xyfor the gray scale on coordinate (x, y), S _maxfor S _xythe full-size allowed.Adaptive median filter algorithm is operated in two levels, is defined as A layer and B layer, A layer: A ₁=Z _med-Z _min, A ₂=Z _med-Z _maxif, A ₁>0, A ₂<0, forwards B layer to, otherwise increases window size, if window size≤S _max, repeat A layer, otherwise export Z _xy; B layer: B ₁=Z _xy-Z _min, B ₂=Z _xy-Z _maxif, B ₁>0, B ₂<0, exports Z _xy, otherwise export Z _med.

Image compression encoding method is as follows: the average number of bits of image entropy presentation video gray scale set, unit is bits/pixel, describes the average information of image information source.Entropy code algorithm has multiple, image compression application Huffman encoding of the present invention, and its principle is as follows: Huffman encoding determines code length in strict accordance with probability match method, and the gray value that probability is large corresponds to short code, and the gray value that probability is little corresponds to long code.Huffman encoding step is as follows: (1) counts the probability that in image, each gray value occurs, and according to order arrangement from big to small; (2) select two values that probability is minimum each time, they be added, the new frequency values of formation and other frequency values form a new frequency sets; (3) repeat (2) step, to the last obtain frequency and be 1; (4) distribution codeword, progressively encodes forward conversely to above-mentioned steps, and each Bu Youliangge branch respectively gives a binary code, the imparting code element 0 large to probability, the imparting code element 1 (or contrary) little to probability.

(3) electronic computer of monitoring center

The monitoring image file compressed through gateway node 2 passes in the electronic computer (5) of monitoring center through GPRS network and Internet network.First carry out image decompression, application inverse discrete cosine transform fast algorithm (IDCT fast algorithm), its basic thought can be described below: to calculate two dimension 8 × 8IDCT:

$\left\{\begin{array}{c}F(u,v)=\frac{1}{4}c\left(u\right)c\left(v\right)\underset{m=0}{\overset{7}{\mathrm{\&Sigma;}}}\underset{n=0}{\overset{7}{\mathrm{\&Sigma;}}}[f(m,n)\&CenterDot;\cos \frac{(2m+1)\mathrm{\&pi;u}}{16}\&CenterDot;\cos \frac{(2n+1)\mathrm{\&pi;v}}{16}]\\ f(m,n)=\frac{1}{4}\underset{u=0}{\overset{7}{\mathrm{\&Sigma;}}}\underset{v=0}{\overset{7}{\mathrm{\&Sigma;}}}[c\left(u\right)\&CenterDot;c\left(v\right)\&CenterDot;F(u,v)\&CenterDot;\cos \frac{(2m+1)\mathrm{\&pi;u}}{16}\&CenterDot;\cos \frac{(2n+1)\mathrm{\&pi;v}}{16}]\end{array}\right.---\left(1\right)$

Wherein m, n, u, v=0,1 ..., 7.

Formula (1) matrix notation: $\begin{array}{c}\left[F\right]=\left[G\right]\&CenterDot;\left[f\right]\&CenterDot;{\left[G\right]}^T\\ \left[f\right]={\left[G\right]}^T\&CenterDot;\left[F\right]\&CenterDot;\left[G\right]\end{array}$

$\left[F\right]=\left[\begin{array}{cccc}F\left(\mathrm{0,0}\right)& F\left(\mathrm{0,0}\right)& ...& F\left(\mathrm{0,0}\right)\\ F\left(\mathrm{1,0}\right)& F\left(\mathrm{1,1}\right)& ...& F\left(\mathrm{1,7}\right)\\ ...& & & \\ F\left(\mathrm{7,0}\right)& F\left(\mathrm{7,1}\right)& ...& F\left(\mathrm{7,7}\right)\end{array}\right]$

Wherein: $\left[f\right]=\left[\begin{array}{cccc}f\left(\mathrm{0,0}\right)& f\left(\mathrm{0,0}\right)& ...& f\left(\mathrm{0,0}\right)\\ f\left(\mathrm{1,0}\right)& f\left(\mathrm{1,1}\right)& ...& f\left(\mathrm{1,7}\right)\\ ...& & & \\ f\left(\mathrm{7,0}\right)& f\left(\mathrm{7,1}\right)& ...& f\left(\mathrm{7,7}\right)\end{array}\right]$

$\left[G\right]=\left[\begin{array}{cccc}\frac{1}{2\sqrt{2}}& \frac{1}{2\sqrt{2}}& ...& \frac{1}{2\sqrt{2}}\\ \frac{1}{2}\cos \frac{\mathrm{\&pi;}}{16}& \frac{1}{2}\cos \frac{3\mathrm{\&pi;}}{16}& ...& \frac{1}{2}\cos \frac{15\mathrm{\&pi;}}{16}\\ ...& & & \\ \frac{1}{2}\cos \frac{7\mathrm{\&pi;}}{16}& \frac{1}{2}\cos \frac{21\mathrm{\&pi;}}{16}& ...& \frac{1}{2}\cos \frac{105\mathrm{\&pi;}}{16}\end{array}\right]=\left[\begin{array}{c}{G}_0\\ G1\\ .\\ .\\ .\\ G_7\end{array}\right]$

Note for the vector that u in matrix [G] is capable formed, then

$\left[f\right]=\left[\begin{array}{cccc}{G}_0^T& G_1^T& ...& G_7^T\end{array}\right]\&CenterDot;\left[F\right]\&CenterDot;\begin{array}{}\end{array}\left[\begin{array}{c}{G}_0\\ G_1\\ .\\ .\\ .\\ G_7\end{array}\right]$

Matrix F is write as following summation form:

Then have $\begin{array}{c}f=\underset{u=0}{\overset{7}{\mathrm{\&Sigma;}}}\underset{v=0}{\overset{7}{\mathrm{\&Sigma;}}}\left\{\right[F(u,n)]\&CenterDot;{\left[G_u\right]}^T\&CenterDot;[G_v\left]\right\}\\ =\underset{u=0}{\overset{7}{\mathrm{\&Sigma;}}}\underset{v=0}{\overset{7}{\mathrm{\&Sigma;}}}\left\{\right[F(u,v)]\&CenterDot;[T_{\mathrm{uv}}\left]\right\}\end{array}---\left(2\right)$

Wherein, [T _uv]=[G _u] ^t× [G _v] (u, v=0,1 ..., 7) and be called as primary image corresponding to conversion coefficient F (u, v).The physical significance of formula (2) is: using coefficient in transform domain F (u, v) as weight coefficient when, can obtain original image matrix [f] by the linear combination of all primary images.

At [F (the u that formula (2) calculates, v)] coefficient is all through quantification treatment, after general quantification 8 × 8 sub-blocks DCT coefficient in, the low frequency coefficient of minority is only had to be nonzero value, the conversion coefficient of all the other most of upper frequencies is all null value, and have the value of quite a few nonzero-value coefficient to be ± 1, there is very large symmetry between the element simultaneously in IDCT in each primary image matrix.Make full use of the amount of calculation that above-mentioned feature can greatly reduce formula (2).

Through the image of decompress(ion), be presented on the screen of computer, for the surveillance map picture be sent on screen, what calculate crack that function automatic recognition image absorbs and expansion joint specifically determines position, and specific algorithm is as follows:

In monitoring system, the position of description node is come in embedded space, measures the coordinate that estimated value locates it linear system from a monitoring node 1 to gateway node 2.Suppose to there is M gateway node 2, monitoring node 1S _icoordinate in the embedded space of M dimension represents with estimated value vector value: [P _i]=[P _i1, P _i2, P _i3..., P _iM] ^t, [P _ij] represent the measured value of node i to node j, [P _ii]=0.Whole embedded space can be expressed as [P]=[P with the estimated matrix of a M × M ₁, P ₂..., P _m].

In like manner, geographic distance vector representation is [L _i]=[L _i1, L _i2..., L _iM] ^t, [L _ij] represent the geographic distance of node i to node j.Then geographic distance matrix notation is [L]=[L ₁, L ₂..., L _m].Location algorithm main thought of the present invention is structure optimum linearity conversion [T], provides one from estimated matrix [P] to the mapping relations of distance matrix [L].Unknown node, after obtaining an estimated vector, just can utilize this mapping relations to calculate its vector distance, thus the position coordinates of computing node.[T] is the matrix of a M × M, and each row of [T] is by minimizing variance to obtain.

$e_i=\underset{k=1}{\overset{M}{\mathrm{\&Sigma;}}}{(l_{\mathrm{ik}}-t_i{p}_k)}^2={\left|\right|l_i^T-t_{i}p\left|\right|}^2$

[L ₁,L ₂,…,L _M]＝[T][P ₁,P ₂,…,P _M]

Row vector t is obtained by minimum variance ₁: t _i=l _iT[P] [T] { [P] [P] ^t} ^-1, [T]=[L] [P] [T] { [P] [P] ^t} ^-1.

Image scale size marking is carried out to the crack in the hydraulic engineering of monitoring and expansion joint, and in image processing software, application ruler measure crack and expansion joint the widest part is of a size of X1, the narrowest place is of a size of X2, the distance of two end points in crack is X3.To often open transmitting image (X1, X2, X3), consecutive variations situation is by computer automatic drafting curve, if the value of X1, X2, X3 exceedes defined threshold, automatically show early warning information by computer, and early warning information is sent on the mobile phone of staff.Open up grey forecasting model (EGM (1,1)) according to the numerical applications of (X1, X2, X3) to predict the crack in hydraulic engineering and expansion joint development trend simultaneously.

Open up the geomery parameter X that grey forecasting model (EGM (1,1)) makes full use of crack and expansion joint ₁, X ₂, X ₃cumulative for time dependent size summation, generate new ordered series of numbers, approach with a suitable curve, be after approximating curve reduction and open up grey forecasting model.Be provided with the discrete data of one group of unequal time-interval:

X ⁽⁰⁾＝x ⁽⁰⁾(t ₁),x ⁽⁰⁾(t ₂),...,x ⁽⁰⁾(t _n)

T in formula _i(i=1,2 ..., n) be time that discrete data is corresponding.

For solving data rows difference and the non-linearity of time difference, on GM (1,1) the model basis of even time interval, requirement forecast value and original value approximately equal, propose following equation group:

$\left\{\begin{array}{c}{x}^{\left(0\right)}\left(t_i\right)=c(1-e^a)e^{-{\mathrm{at}}_i}\\ x^{\left(0\right)}\left(t_j\right)=c(1-e^a)e^{{-\mathrm{at}}_j}\end{array}\right.,i=\mathrm{2,3},...,n-1;j=i+1,...,n$

Solution above formula obtains:

$a_{\mathrm{ij}}=\frac{1}{t_i-t_j}\ln \frac{{\text{x}}^{\left(0\right)}\left(t_j\right)}{x^{\left(0\right)}\left(t_i\right)}$

The a drawn _ijaverage is got in summation: $\hat{a}=\frac{1}{{(n-1)}^2}\underset{i=2}{\overset{m-1}{\mathrm{\&Sigma;}}}\underset{j=i+1}{\overset{m}{\mathrm{\&Sigma;}}}{a}_{\mathrm{ij}}$

Solve: $x^{\left(0\right)}\left(t_i\right)=c_i(1-e^{\hat{a}})e^{-\hat{a}{t}_i},i=\mathrm{1,2},...,n$

According to above formula, try to achieve c _i, cumulative summation is averaged:

$\hat{c}=\frac{1}{n-1}{c}_i$

Finally solve and open up grey forecasting model: ${\hat{x}}^{\left(0\right)}\left(t_i\right)={\hat{c}}_i(1-e^{\hat{a}})e^{-{\hat{a}t}_i}.$

Claims (4)

1. a large hydraulic engineering crack and expansion joint change Quantitative Monitoring system, by monitoring node (1), gateway node (2), monitoring center's electronic computer (5) forms, it is characterized in that, monitoring node (1) distribution is arranged in hydraulic engineering, particularly crack is or/and expansion joint place, monitoring node (1) comprises the ZigBee radio sensing network node with image sensor (6), monitoring node (1) is connected monitoring center electronic computer (5) by GPRS network (3) with Internet network (4) through gateway node (2), the image information at hydraulic engineering crack and expansion joint is real-time transmitted in monitoring center's electronic computer (5), automatically each concrete the hydraulic engineering crack of position and the changing value at expansion joint really of monitoring is shown on the computer screen, changing value exceeds defined threshold, computer automatic early-warning, early warning is sent on operating personnel's mobile phone simultaneously, utilize the development trend at computer forecast hydraulic engineering crack and expansion joint.

2. a kind of large hydraulic engineering crack and expansion joint change Quantitative Monitoring system according to claim 1, it is characterized in that, monitoring node (1) forms by with lower component: image sensor (6), signal conditioning circuit (7), based on the radiofrequency emitting module (8) of ZigBee communication agreement, wherein image sensor (6) is connected with the radiofrequency emitting module (8) based on ZigBee communication agreement by signal conditioning circuit (7), and be all connected with photoelectric source (11), photoelectric source (11) is connected with power management chip (12), unified control is powered.

3. a kind of large hydraulic engineering crack and expansion joint change Quantitative Monitoring system according to claim 1, it is characterized in that, webmaster node (2) is by with lower component: the radio frequency based on ZigBee communication agreement accepts module (8), microprocessor (13), SIM card holder (15), GPRS module (16) are formed by connecting in turn, they by wind-solar power supply (14) power supply, and to be sought unity of action power supply management by power management chip (12).

4. described in claim 1,2 or 3, a kind of large hydraulic engineering crack and expansion joint change the Forecasting Methodology of Quantitative Monitoring system, use gray prediction method, it is characterized in that, open up that grey forecasting model makes full use of the geomery parameter X1 at crack and expansion joint, X2, X3 sue for peace cumulative for time dependent size, generate new sequence, approach with a curve, be after approximating curve reduction and open up grey forecasting model.