CN101393247A - Sensing method based on multidimensional difference decoupling powerline running state parameter - Google Patents

Abstract

The invention discloses a multi-dimension interpolation decoupling based sensing method for running state parameters of a transmission line, and relates to a sensing method for running state parameters of an alternating current high voltage transmission line. The method comprises the following steps: firstly, a calibrated multi-dimension dissimilar sensor is adopted to detect the running status of a transmission line; secondly, a multi-dimension interpolation decoupling method which has a good convergence property is adopted to eliminate the influence of coupling information on the sensor so as to quickly obtain an accurate measurement value of the sensor information in a single-sensor state; sensing amount obtained after decoupling processing provides accurate sensor information for a back end status estimation module. The method gradually eliminates the influence of coupling information on measurement target parameters and improves the sensing accuracy through constant interpolation and dimensionality reduction of coupling sensor information.

Description

A kind of method for sensing of the transmission line of electricity running state parameter based on the multi-dimensional interpolation decoupling zero

Technical field

The present invention relates to the method for sensing of transmission line of electricity running state parameter, relate in particular to the method for sensing of ac high voltage transmission line of electricity operational reliability state parameter under the strong coupling environment.

Background technology

Along with the high speed development of modern economy, the pollution source of each department are on the increase, the discharging of flue dust, and waste gas, sandstorm are to the harm of environment, and these factors all produce the power system security reliability service and have a strong impact on.Insulator long-time running in power station and the transmission line of electricity is in rugged surroundings such as strong-electromagnetic field, strong mechanical stress, extraneous erosion, suddenly cold and hot, multiple potential safety hazards such as be prone to insulator surperficial contamination, internal crack, surface fracture, impedance reduce, aging, the safe operation of serious threat electric system.For example: the electric system pollution flashover is continuous in recent years, has caused the tripping operation of whole piece transmission line of electricity and the power failure of electric substation.According to relevant department's statistics, oneself has accounted for second (being only second to the damage to crops caused by thunder accident) of whole power system accident pollution flashover accident, and because accident involves a wide range of knowledge, and power off time is long, it causes damage is more than 10 times of damage to crops caused by thunder accident.Power equipment is monitored, and is the key that guarantees the device security reliability service.And traditional detection method must detect one by one, piecemeal to the scene, expends great amount of manpower and material resources, financial resources, but the generation that still can't stop accident like this.

Aspect the Maintenance and Repair of power equipment, by the trouble hunting of equipment, scheduled overhaul progressively carries out the transition to repair based on condition of component to the guilding principle of China at present, and the on-line monitoring of equipment is the key that realizes repair based on condition of component.The fast development of little electric quantity acquisition technology under computer technology, new energy technology, the communication technology, the strong-electromagnetic field environment for the research and the realization in on-line monitoring field provides technical guarantee, also makes transmission line of electricity operational reliability state-detection become possibility.

Prior art one related to the present invention

Domestic (patent of invention) number: 02115506 " failure point of power transmission line localization method and device " proposes a kind of failure point of power transmission line localization method, utilizes time that voltage that the trouble spot produces or current traveling wave signal arrive the transmission line of electricity two ends and speed and circuit length overall to calculate the position of trouble spot.

Domestic (patent of invention) number: 200310111211 " a kind of power system transmission line fault diagnosis and phase-selecting method " relates to a kind of power system transmission line fault diagnosis and phase-selecting method, may further comprise the steps: voltage, the current signal of data acquisition to obtain detected circuit multiterminal; Electric parameters before and after record and the exchange trouble is to obtain the localization of fault desired data; It is separate to differentiate fault.

More than two technology be from the fault detect angle, only when ultra-high-tension power transmission line breaks down, could judge the trouble spot of transmission line of electricity from the electric weight transducing signal.

Prior art two related to the present invention

" overhead transmission line long-distance monitoring system " of domestic applications (patent of invention) number: CN200410073227.5 proposes a kind of Long-Range Surveillance System that does not have out of doors under the external power supply condition.The user regularly receives monitor message or active request monitor message by information receiver, realizes the remote monitoring of transmission line of electricity.The information of being monitored can be image information and meteorological parameter.

Domestic applications (utility model patent) number: 200520079668 " insulator contamination on-Line Monitor Device " proposed the similar utility model of a cover, and increased the measurement of leakage current values, pulse current.

More than two technology from the state-detection angle, by detecting the leakage current and the meteorological environment parameter of transmission line of electricity, predict the transmission line of electricity reliability of operation in view of the above.For realizing to the effective state-detection of transmission line operational reliability, prerequisite is that the heat transfer agent of transmission line of electricity running status accurately can be provided, above-mentioned two technology have all been ignored the coupled relation between each heat transfer agent under strong electromagnetic, the strong coupling state, the measured value that directly will not carry out the heat transfer agent decoupling zero is as final sensed values, can cause resulting heat transfer agent and actual information that deviation is arranged like this, thereby influence is to the judgement of transmission line operational reliability.

Summary of the invention

For solving above-mentioned middle problem and the defective that exists, the invention provides a kind of method for sensing of the transmission line of electricity running state parameter based on the multi-dimensional interpolation decoupling zero, handle by constantly carrying out the interpolation decoupling zero, finally obtain under the single condition sensing each sensing amount measured value accurately.

The present invention is achieved by the following technical solutions:

The method for sensing of a kind of transmission line of electricity running state parameter based on the multi-dimensional interpolation decoupling zero involved in the present invention, technological core is: the heterogeneous heat transfer agent of multidimensional of the transmission line of electricity running status of strong coupling is carried out decoupling zero handle, progressively eliminate the influence of coupling information, to obtain under the single condition sensing each heat transfer agent measured value accurately rapidly and accurately to the measurement target parameter; Described method mainly may further comprise the steps:

A sets up the heterogeneous sensing mathematical model of multidimensional under the couple state, and all the sensors of transmission line of electricity monitoring running state is demarcated in range;

B carries out disappear first decoupling zero of dimensionality reduction to the multidimensional heat transfer agent that has coupled relation to be handled, and progressively eliminates the influence of coupling information to sensor, obtains under the single condition sensing measurement target parameter measured value accurately.

Described steps A specifically comprises: after setting up the heterogeneous sensing mathematical model of multidimensional under the couple state, obtain the sensing relation function:

Y _i＝f(x ₁，x ₂，...，x _i，...，x _n)(1≤i≤n)

Wherein, Y _iBe the output quantity of particular sensor, x ₁, x ₂..., x _i..., x _n(1≤i≤n) is the heat transfer agent input quantity under the couple state;

By progressively dwindling the degree of freedom of relation function, this relation function is converted to the form Y of single input-list output at last _i=f (x _i).

Described step B specifically comprises: to all coupling information x _j(1≤j≤n, j ≠ i) carry out the linear interpolation unit that disappears, and final decoupling zero obtains $x_1=x_1^{\′},$ $x_2=x_2^{\′},...,$ $x_{i-1}=x_{i-1}^{\′},$ $x_{i+1}=x_{i+1}^{\′},...,$ $x_n=x_n^{\′}$ The family curve of condition lower sensor i:

$f(x_1^{\′},x_2^{\′},...,x_{i-1}^{\′},x_i,x_{i+1}^{\′},...,x_n^{\′})=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{l}_k\left(x_i\right)f(x_1^{\′},x_2^{\′},...,x_{i-1}^{\′},x_i,x_{i+1}^{\′},...,x_n^{\′})$

Interpolation basis function wherein

$l_0\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i1}}{x_{i0}-x_{i1}}& x_i\∈[x_{i0},x_{i1}]\\ 0& x_i\∉[x_{i0},x_{i1}]\end{array}\right.$

$l_k\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i,k-1}}{x_{i,k}-x_{i,k-1}}& x_i\∈[x_{i,k-1},x_{i,k}]\\ \frac{x_i-x_{i,k+1}}{x_{i,k}-x_{i,k+1}}& x_i\∈[x_{i,k},x_{i,k+1}]\\ 0& x_i\∉[x_{i,k-1},x_{i,k+1}]\end{array}\right.(k=\mathrm{1,2},...,n-1)$

$l_n\left(x_i\right)=\left\{\begin{array}{cc}0& x_i\∉[x_{i,n-1},x_{\mathrm{in}}]\\ \frac{x_i-x_{i,n-1}}{x_{i,n}-x_{i,n-1}}& x_i\∈[x_{i,n-1},x_{\mathrm{in}}]\end{array}\right.$

This characteristic expression formula is the input-input form of the two dimension after the decoupling zero.

The beneficial effect of technical scheme provided by the invention is:

Adopted the linear interpolation decoupling method that computer memory and time complexity are lower, convergence is good, the heterogeneous sensing mathematical model of multidimensional of transmission line of electricity under strong electromagnetic, the strong coupling state has finally been changed into simple two-dimensional input-output form; By obtaining sensor family curve under the combination of specific coupling factor in measurement range, can obtain the heat transfer agent of accurate transmission line running status more quickly, for the status predication of transmission line of electricity operational reliability provides heat transfer agent support accurately.

Be coupled heterogeneous sensing mathematical model and heat transfer agent decoupling zero process of the multidimensional of setting up according to this method has generality, and in the fusion of the drift of the multidimensional of integrated sensor, Heterogeneous Sensor, also exist and problem that the coupling of multidimensional sensing is similar, therefore this method not only can be applicable to the sensory field of transmission line of electricity running state parameter, can also be generalized to other N dimension coupling sensory field.

Description of drawings

Fig. 1 is based on the method for sensing process flow diagram of the transmission line of electricity running state parameter of multi-dimensional interpolation decoupling zero;

Fig. 2 is based on the method for sensing of the transmission line of electricity running state parameter of multi-dimensional interpolation decoupling zero and implements illustration;

Fig. 3 is the heterogeneous heat transfer agent mathematical model coupling of a multidimensional of the present invention synoptic diagram.

Embodiment

For making the purpose, technical solutions and advantages of the present invention clearer, embodiment of the present invention is described further in detail below in conjunction with accompanying drawing:

Comprehensively transmission line of electricity running status heat transfer agent comprises the electric weight heat transfer agent, as amount of leakage current, mutual voltage amount, the mutual inductance magnitude of current; Comprise meteorological heat transfer agent, as temperature amount, humidity amount, amount of gas pressure, sunshine amount, rainfall amount, wind speed and direction amount etc.; Also comprise image sensing information, as real-time or image volume regularly etc.Above listed heat transfer agent can be measured respectively by a plurality of Heterogeneous Sensor, but because the relatively poor sensor (as humidity sensor, leakage current sensor etc.) of some sensing unicity is except being subjected to the influence of measurement target parameter, also can be subjected to of the influence of other non-survey aim parameters, cause to produce than mistake between measurement result and the true value its coupling.This time, if adopt other and its to have coupled relation but the test value of the relative sensor preferably of job stability as the benchmark of multi-dimensional interpolation, by the disappear method of first decoupling zero of interpolation, can progressively revise of the influence of non-survey aim parameter, finally obtain measurement target parameter measured value accurately sensor.

Present embodiment provides a kind of method for sensing of the transmission line of electricity running state parameter based on the multi-dimensional interpolation decoupling zero based on above-mentioned thought, and as shown in Figure 1, this method mainly comprises:

Step 10 is set up the heterogeneous multidimensional sensing mathematical model under the couple state, and all the sensors is demarcated in range;

Step 20 pair multidimensional coupling heat transfer agent is carried out the first decoupling zero that disappears of interpolation dimensionality reduction and is handled.

The foundation of described multidimensional sensing model, the demarcation of sensor and the decoupling zero calculating of multi-dimensional interpolation are described in detail:

1, the foundation of multidimensional sensing model

According to Fig. 3, multidimensional heat transfer agent coupling model is:

$\left\{\begin{array}{c}{Y}_i=Y_{i1}+Y_{i2}\\ Y_{i1}=X_i{G}_{\mathrm{ii}}\\ Y_{i2}=\underset{j=1,j\≠i}{\overset{n}{\mathrm{\Σ}}}{X}_j{G}_{\mathrm{ji}}\end{array}\right.(1\≤i\≤n)---\left(1\right)$

The method that the coupled characteristic equation of sensor is demarcated is by experiment obtained by data pre-service and homing method.

According to testing goal, the target that decoupling zero is calculated is to obtain Y _iWith X _iFundamental function one to one.Therefore the essence of decoupling zero calculating is: to relation function Y _i=f (x ₁, x ₂..., x _i..., x _n) (1≤i≤n), progressively dwindle the degree of freedom of function variable is converted to Y at last _i=f (x _i) form.

If Function Y _i=f (x ₁, x ₂..., x _i..., x _n) (1≤i≤n) satisfies monotonicity, then output quantity Y _iWith input quantity X _iAnd coupling amount X _j(1≤j≤n, j ≠ i) just existence concerns one to one.The multi-dimensional interpolation decoupling zero is exactly according to Y _i=f (x ₁, x ₂..., x _i..., x _n) corresponding relation, utilize repeatedly the way of interpolation, progressively dwindle degree of freedom, obtain at last at input quantity x _iWith other coupling amounts x _j(1≤j≤n, the output Y under the common influence of j ≠ i) _iThereby, simulate Y under complex environment _i=f (x _i) family curve.

2, the demarcation of sensor

The decoupling zero of utilization method of interpolation needs on given interpolation point each sensor to be demarcated in advance.In particular sensor i working range,, select one group of calibration point x according to actual conditions _Io, x _Il..., x _Ik(1≤i≤n, k〉1), make x _Io＜x _Il＜...＜x _Ik, and x _IoAnd x _IkTwo end points for sensor i working range.

In like manner, by test, measure the standard point x that there is the coupled relation sensor in all _Jo, x _Jl..., x _Ik(1≤j≤n, j ≠ i; K〉1), and can construct the family curve f (x under the coupling of multidimensional heat transfer agent in the ultra-high-tension power transmission line according to this ₁₀, x ₂₀..., x _I-1,0, x _i, x _I+1,0.., x _N0), f (x ₁₁, x ₂₁..., x _I-1,1, x _i, x _I+1,1.., x _N1) ..., f (x _1k, x _2k..., x _{I-1, k}, x _i, x _{I+1, k}.., x _Nk).

3, the multi-dimensional interpolation decoupling zero is calculated

Multi-dimensional interpolation decoupling zero computation process is finished by the function of a single variable interpolation of repeated multiple times.When requiring $x_1=x_1^{\′},$ $x_2=x_2^{\′},...,$ $x_{i-1}=x_{i-1}^{\′},$ $x_{i+1}=x_{i+1}^{\′},...,$ $x_n=x_n^{\′}$ Sensing characteristics curve Y under the condition _i=f (x _i), concrete steps are as follows:

(1) to coupling information x ₁Carry out the interpolation unit that disappears

Work as x ₂=x ₂₀, x ₃=x ₃₀... x _I-1=x _I-1,0, x _I+1=x _I+1,0... x _n=x _N0, get x _i=x _I0With f (x ₁₀, x ₂₀..., x _I-1,0, x _i, x _I+1,0.., x _N0), f (x ₁₁, x ₂₀..., x _I-1,0, x _i, x _I+1,0.., x _N0) ..., f (x _1n, x ₂₀..., x _I-1,0, x _i, x _I+1,0.., x _N0) a series of joining f (x ₁₀, x ₂₀..., x _I-1,0, x _I0, x _I+1,0.., x _N0), f (x ₁₁, x ₂₀..., x _I-1,0, x _I0, x _I+1,0.., x _N0) ..., f (x _1n, x ₂₀..., x _I-1,0, x _I0, x _I+1,0.., x _N0).In like manner, get x _i=x _Ik(during 1≤k≤n), obtain each serial f (x ₁₀, x ₂₀..., x _I-1,0, x _{I, k}, x _I+1,0.., x _N0), f (x ₁₁, x ₂₀..., x _I-1,0, x _{I, k}, x _I+1,0.., x _N0) ..., f (x _1n, x ₂₀..., x _I-1,0, x _{I, k}, x _I+1,0.., x _N0) (value of 2≤k≤n) is as the basic point of interpolation.

According to known point (x _I0, f (x _1k, x ₂₀..., x _N0)) (0≤k≤n), utilize the linear interpolation formula, structure x _i=x _I0The time coupling variable x ₁Interpolating function:

$f(x_1,x_{20},x_{30},...,x_{n0})=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{l}_k\left(x_1\right)f(x_{1k},x_{20},x_{30},...,x_{n0})---\left(2\right)$

Interpolation basis function wherein:

$l_0\left(x_1\right)=\left\{\begin{array}{cc}\frac{x_1-x_{11}}{x_{10}-x_{11}}& x_1\∈[x_{10},x_{11}]\\ 0& x_1\∉[x_{10},x_{11}]\end{array}\right.$

$l_k\left(x_1\right)=\left\{\begin{array}{cc}\frac{x_1-x_{1,k-1}}{x_{1,k}-x_{1,k-1}}& x_1\∈[x_{1,k-1},x_{1,k}]\\ \frac{x_1-x_{1,k+1}}{x_{1,k}-x_{1,k+1}}& x_1\∈[x_{1,k},x_{1,k+1}]\\ 0& x_1\∉[x_{1,k-1},x_{1,k+1}]\end{array}\right.(k=\mathrm{1,2},...,n-1)---\left(3\right)$

$l_n\left(x_1\right)=\left\{\begin{array}{cc}0& x_1\∉[x_{1,n-1},x_{1n}]\\ \frac{x_1-x_{1,n-1}}{x_{1,n}-x_{1,n-1}}& x_1\∈[x_{1,n-1},x_{1n}]\end{array}\right.$

Can obtain according to this $x_1=x_1^{\′}$ The time

Value, in like manner, can be in the hope of a series of

(value of 1≤k≤n), structure in view of the above $x_1=x_1^{\′},$ x ₂=x ₂₀..., x _I-1=x _I-1,0, x _I+1=x _I+1,0..., x _n=x _N0Sensing characteristics curve under the condition:

$f(x_1^{\′},x_{20},...,x_{i-\mathrm{1,0}},x_i,x_{i+\mathrm{1,0}},...,x_{n0})=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{l}_k\left(x_i\right)f(x_1^{\′},x_{20},...,x_{i-\mathrm{1,0}},x_{i,k},x_{i+\mathrm{1,0}},...,x_{n0})---\left(4\right)$

Interpolation basis function wherein

$l_0\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i1}}{x_{i0}-x_{i1}}& x_i\∈[x_{i0},x_{i1}]\\ 0& x_i\∉[x_{i0},x_{i1}]\end{array}\right.$

$l_k\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i,k-1}}{x_{i,k}-x_{i,k-1}}& x_i\∈[x_{i,k-1},x_{i,k}]\\ \frac{x_i-x_{i,k+1}}{x_{i,k}-x_{i,k+1}}& x_i\∈[x_{i,k},x_{i,k+1}]\\ 0& x_i\∉[x_{i,k-1},x_{i,k+1}]\end{array}(k=\mathrm{1,2},...,n-1)\right.---\left(5\right)$

$l_n\left(x_i\right)=\left\{\begin{array}{cc}0& x_i\∉[x_{i,n-1},x_{\mathrm{in}}]\\ \frac{x_i-x_{i,n-1}}{x_{i,n}-x_{i,n-1}}& x_i\∈[x_{i,n-1},x_{\mathrm{in}}]\end{array}\right.$

(2) to other coupling informations x _j(1≤j≤n, j ≠ i) carry out the interpolation unit that disappears

The above method of interpolation of utilization can obtain $x_1=x_1^{\′},$ $x_2=x_2^{\′},...,$ $x_{j-1}=x_{j-1}^{\′},$ x _j=x _J0..., x _n=x _N0Family curve under the condition can see that for the target property curve, the function degree of freedom reduces one by one, and so far, degree of freedom has been n-(j-1).In like manner, change coupling heat transfer agent x _jValue, make x _jEqual x successively _J0, x _J1..., x _JkThen can obtain $x_1=x_1^{\′},$ $x_2=x_2^{\′},...,$ $x_{j-1}=x_{j-1}^{\′},$ x _j=x _J0..., x _n=x _N0N+1 bar family curve under the condition.Get x _i=x _I0With each characteristic intersection point, can get equally $f(x_1^{\′},x_2^{\′},...,x_{j-1}^{\′},x_{j,0},x_{j+\mathrm{1,0}},..,x_{n0}),$ $f(x_1^{\′},x_2^{\′},...,x_{j-1}^{\′},x_{j,1},x_{j+\mathrm{1,0}},..,x_{n0}),...,$ $f(x_1^{\′},x_2^{\′},...,x_{j-1}^{\′},x_{j,k},x_{j+\mathrm{1,0}},..,x_{n0})$ Value, utilize the piecewise linear interpolation formula, the structure x _i=x _I0Time coupling sensing variable x _jInterpolating function:

$f(x_1^{\′},x_2^{\′},...,x_{j-1}^{\′},x_j,..,x_{n0})=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{l}_k\left(x_j\right)f(x_1^{\′},x_2^{\′},...,x_{j-1}^{\′},x_j,..,x_{n0})---\left(6\right)$

Interpolation basis function wherein

$l_0\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i1}}{x_{i0}-x_{i1}}& x_i\∈[x_{i0},x_{i1}]\\ 0& x_i\∉[x_{i0},x_{i1}]\end{array}\right.$

$l_k\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i,k-1}}{x_{i,k}-x_{i,k-1}}& x_i\∈[x_{i,k-1},x_{i,k}]\\ \frac{x_i-x_{i,k+1}}{x_{i,k}-x_{i,k+1}}& x_i\∈[x_{i,k},x_{i,k+1}]\\ 0& x_i\∉[x_{i,k-1},x_{i,k+1}](k=\mathrm{1,2},...,n-1)\end{array}---\left(7\right)\right.$

$l_n\left(x_i\right)=\left\{\begin{array}{cc}0& x_i\∉[x_{i,n-1},x_{\mathrm{in}}]\\ \frac{x_i-x_{i,n-1}}{x_{i,n}-x_{i,n-1}}& x_i\∈[x_{i,n-1},x_{\mathrm{in}}]\end{array}\right.$

Can obtain according to this $x_j=x_j^{\′},$ x _i=x _I0The time $f(x_1^{\′},x_2^{\′},...,x_j^{\′},x_{j+\mathrm{1,0}},..,x_{i-\mathrm{1,0}},x_{i0},x_{i+\mathrm{1,0}},...,x_{n0})$ Value, in like manner can obtain x _i=x _Ik(during 1≤k≤n) $f(x_1^{\′},x_2^{\′},...,x_j^{\′},x_{j+\mathrm{1,0}},..,x_{i-\mathrm{1,0}},x_{\mathrm{ik}},x_{i+\mathrm{1,0}},...,x_{n0})$ (value of obtaining of 1≤k≤n), structure in view of the above $x_1=x_1^{\′},...,$ $x_j=x_j^{\′},$ x _J+1=x _J+1,0..., x _n=x _N0(1≤j≤n, the family curve under the condition of j ≠ i)

$f(x_1^{\′},x_2^{\′},...,x_j^{\′},x_{j+\mathrm{1,0}},..,x_{i-\mathrm{1,0}},x_i,x_{i+\mathrm{1,0}},...,x_{n0})$

$=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{l}_k\left(x_i\right)f(x_1^{\′},x_2^{\′},...,x_j^{\′},x_{j+\mathrm{1,0}},..,x_{i-\mathrm{1,0}},x_i,x_{i+\mathrm{1,0}},...,x_{n0})---\left(8\right)$

Interpolation basis function wherein

$l_0\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i1}}{x_{i0}-x_{i1}}& x_i\∈[x_{i0},x_{i1}]\\ 0& x_i\∉[x_{i0},x_{i1}]\end{array}\right.$

$l_k\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i,k-1}}{x_{i,k}-x_{i,k-1}}& x_i\∈[x_{i,k-1},x_{i,k}]\\ \frac{x_i-x_{i,k+1}}{x_{i,k}-x_{i,k+1}}& x_i\∈[x_{i,k},x_{i,k+1}]\\ 0& x_i\∉[x_{i,k-1},x_{i,k+1}]\end{array}\right.(k=\mathrm{1,2},...,n-1)---\left(9\right)$

$l_n\left(x_i\right)=\left\{\begin{array}{cc}0& x_i\∉[x_{i,n-1},x_{\mathrm{in}}]\\ \frac{x_i-x_{i,n-1}}{x_{i,n}-x_{i,n-1}}& x_i\∈[x_{i,n-1},x_{\mathrm{in}}]\end{array}\right.$

Using said method repeatedly can obtain at last $x_1=x_1^{\′},$ $x_2=x_2^{\′},...,$ $x_{i-1}=x_{i-1}^{\′},$ $x_{i+1}=x_{i+1}^{\′},...,$ $x_n=x_n^{\′}$ The family curve of condition lower sensor i

$f(x_1^{\′},x_2^{\′},...,x_{i-1}^{\′},x_i,x_{i+1}^{\′},...,x_n^{\′})=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{l}_k\left(x_i\right)f(x_1^{\′},x_2^{\′},...,x_{i-1}^{\′},x_i,x_{i+1}^{\′},...,x_n^{\′})---\left(10\right)$

This characteristic expression formula has been converted into general two dimension input-input model.

Described step 10 specifically comprises: as Fig. 2, shown in Figure 3, at first set up the heterogeneous sensing mathematical model of multidimensional of reflection transmission line of electricity running status, the parameter of transmission line of electricity running status through a plurality of sensors (in Fig. 2, sensor 1, sensor 2 ..., sensor n) and by the sensing detection module be transferred to output sensing amount in the interpolation decoupling zero module (sensing amount 1, sensing amount 2 ..., sensing amount n), obtain the sensing relation Function Y _i=f (x ₁, x ₂..., x _i..., x _n) (1≤i≤n), wherein, Y _iBe the output quantity of particular sensor, x ₁, x ₂..., x _i..., x _n(1≤i≤n) is the heat transfer agent input quantity under the couple state; Then all Heterogeneous Sensor of transmission line of electricity monitoring running state are demarcated in range,, measure the calibration point x of all Heterogeneous Sensor that have coupled relation by test _Jo, x _Jl..., x _Jk(1≤j≤n; K〉1), and construct the family curve f (x under the heterogeneous heat transfer agent coupling of multidimensional in the ultra-high-tension power transmission line according to this ₁₀, x ₂₀..., x _I-1,0, x _i, x _I+1,0.., x _N0), f (x ₁₁, x ₂₁..., x _I-1,1, x _i, x _I+1,1.., x _N1) ..., f (x _1k, x _2k..., x _{I-1, k}, x _i, x _{I+1, k}.., x _Nk).

Described step 20 specifically comprises: progressively to all coupling information x _j(1≤j≤n, j ≠ i) carry out the linear interpolation unit that disappears, and multi-dimensional interpolation decoupling zero computation process is finished by the function of a single variable interpolation of repeated multiple times.

Can decoupling zero obtain at last $x_1=x_1^{\′},$ $x_2=x_2^{\′},...,$ $x_{i-1}=x_{i-1}^{\′},$ $x_{i+1}=x_{i+1}^{\′},...,$ $x_n=x_n^{\′}$ The family curve of condition lower sensor i:

$f(x_1^{\′},x_2^{\′},...,x_{i-1}^{\′},x_i,x_{i+1}^{\′},...,x_n^{\′})=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{l}_k\left(x_i\right)f(x_1^{\′},x_2^{\′},...,x_{i-1}^{\′},x_i,x_{i+1}^{\′},...,x_n^{\′})$

Interpolation basis function wherein

$l_0\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i1}}{x_{i0}-x_{i1}}& x_i\∈[x_{i0},x_{i1}]\\ 0& x_i\∉[x_{i0},x_{i1}]\end{array}\right.$

$l_k\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i,k-1}}{x_{i,k}-x_{i,k-1}}& x_i\∈[x_{i,k-1},x_{i,k}]\\ \frac{x_i-x_{i,k+1}}{x_{i,k}-x_{i,k+1}}& x_i\∈[x_{i,k},x_{i,k+1}]\\ 0& x_i\∉[x_{i,k-1},x_{i,k+1}]\end{array}\right.(k=\mathrm{1,2},...,n-1)$

$l_n\left(x_i\right)=\left\{\begin{array}{cc}0& x_i\∉[x_{i,n-1},x_{\mathrm{in}}]\\ \frac{x_i-x_{i,n-1}}{x_{i,n}-x_{i,n-1}}& x_i\∈[x_{i,n-1},x_{\mathrm{in}}]\end{array}\right.$

This characteristic expression formula decoupling zero is the input-input form of two dimension.

By the value of conversion i, the decoupling method that utilizes step 20 to provide needing can to obtain the measured value accurately of the specific heat transfer agent of decoupling zero fast.

Embodiment 1

Humidity sensor during the transmission line of electricity running status detects is the relatively poor relatively class sensor of sensing unicity.In the present embodiment, the humidity sensor output quantity is except mainly being subjected to the ambient humidity influence, also be subjected to the coupling influence of environment temperature factor, therefore, for obtaining humidity sensor amount accurately, present embodiment adopt coupling that humidity sensor records the temperature factor influence humidity value and compare stable measured temperature and carry out the interpolation compensation.Concrete operations are as follows:

Can obtain relation function Y by step 10 _s=f (x _s, x _t), Y wherein _sBe the output quantity of humidity sensor, x _s, x _tBe respectively humidity sensor information input quantity and temperature sensing information input quantity.

Can utilize mathematical relation to temperature coupling information x according to step 20 _tCarry out the disappear first decoupling zero of convergent interpolation and handle, can obtain temperature $x_t=x_t^{\′}$ Humidity characteristic curve under the condition This moment, the degree of freedom of function reduced a dimension; Utilize this family curve can be by the measured value of humidity sensor

Humidity amount x accurately when interpolation obtains the transmission line of electricity running status _s

Embodiment 2

In addition, the leakage current sensor during the transmission line of electricity running status detects also is the relatively poor relatively class sensor of sensing unicity.In the present embodiment, the leakage current sensor output quantity is except mainly being subjected to the leakage current informational influence, also be subjected to the coupling influence of numerous factors such as environment temperature, ambient humidity, Environment Day photograph, wind-force, therefore, for obtaining leakage current sensing amount accurately, present embodiment adopts coupling sensed values that leakage current sensor records and has coupled relation with it but compare stable temperature sensing value, light intensity sensed values, wind speed sensed values and carry out the interpolation compensation through the humidity sensor value that decoupling zero obtains.Concrete operations are as follows:

By step 10, obtain relation function Y _i=f (x _i, x _t, x _l, x _f, x _s), Y wherein _iBe the output quantity of leakage current sensor, x _i, x _t, x _l, x _f, x _sBe respectively the humidity sensor amount after leakage current heat transfer agent input quantity, temperature sensing information input quantity, light intensity heat transfer agent input quantity, wind speed heat transfer agent input quantity and the decoupling zero.

According to step 20, at first utilize mathematical relation to temperature coupling information x _tCarry out the disappear first decoupling zero of convergent interpolation and handle, obtain temperature $x_t=x_t^{\′}$ Leakage current sensing characteristics curve under the condition $Y_i=f(x_i,x_t^{\′},x_l,x_f,x_s),$ This moment, the degree of freedom of function reduced a dimension; In like manner, more respectively to light intensity heat transfer agent x _l, wind speed heat transfer agent x _fWith the humidity sensor information x after the decoupling zero _sCarry out the disappear first decoupling zero of convergent interpolation and handle, finally obtain temperature $x_t=x_t^{\′},$ Intensity of sunshine $x_l=x_l^{\′},$ Wind speed $x_f=x_f^{\′}$ , humidity $x_s=x_s^{\′}$ Transmission line of electricity leakage current sensing characteristics curve under the condition $Y_i=f(x_i,x_t^{\′},x_l^{\′},x_f^{\′},x_s^{\′}),$ Utilize this family curve can be by the measured value of leakage current sensor

Amount of leakage current x accurately when interpolation obtains the transmission line of electricity running status _i

The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claim.

Claims (3)

1, a kind of method for sensing of the transmission line of electricity running state parameter based on the multi-dimensional interpolation decoupling zero, it is characterized in that, the heterogeneous heat transfer agent of multidimensional under the transmission line of electricity running status of strong coupling is carried out decoupling zero to be handled, progressively eliminate the influence of coupling information, to obtain under the single condition sensing each heat transfer agent measured value accurately rapidly and accurately to the measurement target parameter; Described method mainly may further comprise the steps:

A sets up the heterogeneous sensing mathematical model of multidimensional under the couple state, and all the sensors of transmission line of electricity monitoring running state is demarcated in range;

B carries out disappear first decoupling zero of dimensionality reduction to the multidimensional heat transfer agent that has coupled relation to be handled, and progressively eliminates the influence of coupling information to sensor, obtains under the single condition sensing measurement target parameter measured value accurately.

2, the method for sensing of the transmission line of electricity running state parameter based on the multi-dimensional interpolation decoupling zero according to claim 1 is characterized in that described steps A specifically comprises: after setting up the heterogeneous sensing mathematical model of multidimensional under the couple state, obtain the sensing relation function:

Y _i＝f(x ₁，x ₂，...，x _i，...，x _n)(1≤i≤n)

Wherein, Y _iBe the output quantity of particular sensor, x ₁, x ₂..., x _i..., x _n(1≤i≤n) is the heat transfer agent input quantity under the couple state;

By progressively dwindling the degree of freedom of relation function, this relation function is converted to the form Y of single input-list output at last _i=f (x _i).

3, the method for sensing of the transmission line of electricity running state parameter based on the multi-dimensional interpolation decoupling zero according to claim 1 is characterized in that described step B specifically comprises: to all coupling information x _j(1≤j≤n, j ≠ i) carry out the linear interpolation unit that disappears, and final decoupling zero obtains $x_1=x_1^{\′},$ $x_2=x_2^{\′},$ $x_{i-1}=x_{i-1}^{\′},$ $x_{i+1}=x_{i+1}^{\′},$ $x_n=x_n^{\′}$ The family curve of condition lower sensor i:

$f(x_1^{\′},x_2^{\′},...,x_{i-1}^{\′},x_i,x_{i+1}^{\′},...,x_n^{\′})=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{l}_k\left(x_i\right)f(x_1^{\′},x_2^{\′},...,x_{i-1}^{\′},x_i,x_{i+1}^{\′},...,x_n^{\′})$

Interpolation basis function wherein

$l_0\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i1}}{x_{i0}-x_{i1}}& x_i\∈[x_{i0},x_{i1}]\\ 0& x_i\∉[x_{i0},x_{i1}]\end{array}\right.$

$l_k\left(x_i\right)=\left\{\begin{array}{cc}\frac{x_i-x_{i,k-1}}{x_{i,k}-x_{i,k-1}}& x_i\∈[x_{i,k-1},x_{i,k}]\\ \frac{x_i-x_{i,k+1}}{x_{i,k}-x_{i,k+1}}& x_i\∈[x_{i,k},x_{i,k+1}]\\ 0& x_i\∉[x_{i,k-1},x_{i,k+1}]\end{array}\right.(k=\mathrm{1,2},...,n-1)$

$l_n\left(x_i\right)=\left\{\begin{array}{cc}0& x_i\∉[x_{i,n-1},x_{\mathrm{in}}]\\ \frac{x_i-x_{i,n-1}}{x_{i,n}-x_{i,n-1}}& x_i\∈[x_{i,n-1},x_{\mathrm{in}}]\end{array}\right.$

This characteristic expression formula is the input-input form of the two dimension after the decoupling zero.

Images