CN101505110B - Inverter for instant voltage PID current PI digital control - Google Patents

Abstract

The invention discloses an instantaneous voltage PID current PI digital control inverted power supply. The output end of the current PI digital controller is connected with the control end of the inverter; the output end of the inverter is connected with the input end of a voltage transducer and a load; the output end of a voltage transducer is connected with the negative input end of a first substracter; the positive input end of the substracter receives reference quantity; the output end of the first substracter is connected with the input end of a voltage PID digital controller; the direct current end of the inverter is connected with a direct current power supply; the current led out from the inverter is connected with the input end of a current transducer; the output end of the current transducer is connected with the negative input end of a second substracter; the positive input end of the second substracter is connected with the output end of the voltage PID digital controller; and the output end of the second substracter is connected with the input end of the current PI digital controller. The inverted power supply is high in precision, quick and smooth in response, and small in total harmonic wave distortion of output voltage under nonlinear load condition. The inverted power supply can be widely applied to alternating stabilized power supplies, uninterruptible power supplies, active power filters and the like.

Description

The numerically controlled inverter of instant voltage PID current PI

Technical field

The present invention relates to a kind of power conversion circuit, the numerically controlled inverter of particularly a kind of instant voltage PID (proportion integration differentiation) current PI (proportional integral).

Background technology

Along with the development of very large scale integration technology, the performance of microprocessor rapidly improves, and cost price constantly descends, and makes that the full-digital control of inverter is increasing.Direct Digital control is compared with simulation control, and following advantage is arranged: from noise and drift effect, digitial controller is superior far beyond corresponding analog controller; Can carry out complicated calculations fast with constant accuracy, antijamming capability is strong; Can be easy to as required change control program (controller characteristic), versatility is extremely strong, and upgrading is convenient; Has stronger monitoring function, system maintenance easy; Digital piece volumes is little, and is in light weight, is easy to standardization.

Digital control have many superior parts with respect to simulation control, makes it to be subjected to extensive concern.The digitial controller of inverter adopts and repeats control and can suppress periodic disturbance well, improve the steady-state response of system, but dynamic response is unhappy, at least at one more than the primitive period; Adopt dead beat control to have dynamic responding speed faster, but control performance is strong to the system parameters dependence, and parameter is changed sensitivity, poor robustness might reduce the stability of a system or even unstable; The digital control dynamic response of PI of employing routine is slow, control precision is poor.Be suggested though as seen can bring into play several digital control methods of digital control advantage, exist not enough.

Invention Inner holds

The objective of the invention is to overcome above-mentioned the deficiencies in the prior art part, provide a kind of instant voltage PID current PI numerically controlled inverter, this inverter stable state accuracy height; Dynamic response fast, steadily; The total percent harmonic distortion of output voltage is low under the nonlinear load situation, surpasses under 3 the situation at specified nonlinear load, load current crest factor, and the total percent harmonic distortion of output voltage is also lower; The Control Robustness height, antijamming capability is strong, can export high-quality AC power.

The numerically controlled inverter of instant voltage PID current PI provided by the invention is characterized in that:

The control end and the microprocessor of inverter join, the output of inverter joins with the input of voltage sensor and load, the electric current of drawing in the inverter and the input of current sensor join, the inverter dc terminal links to each other with DC power supply, and the output of voltage sensor and the output of current sensor join with microprocessor respectively;

Described microprocessor comprises voltage pid number controller, current PI digitial controller and first, second subtracter, the output of current PI digitial controller and the control end of inverter join, the negative input end of the output of voltage sensor and first subtracter joins, and the positive input terminal of first subtracter receives reference quantity u _rThe input of the output of first subtracter and voltage pid number controller joins, the negative input end of the output of current sensor and second subtracter joins, the output of the positive input terminal of second subtracter and voltage pid number controller joins, and the input of the output of second subtracter and current PI digitial controller joins.

The present invention compared with prior art has the following advantages:

(1) under the various loading conditions from the zero load to the nominal load, all within 0.42%, steady-state error reduces the output voltage precision of voltage regulation greatly.

When (2) load changing reached 50% rated power, dynamic transition process was no more than 0.8ms, and the output voltage rate of change is no more than 7.07%, and workload-adaptability strengthens.

(3) under specified nonlinear load, load current crest factor reached 3.2 situation, the total percent harmonic distortion THD=1.316% of output voltage showed the wave distortion that nonlinear load is caused and has stronger inhibition ability.

(4) the present invention is in the design to inverter instant voltage PID current PI digitial controller Control Parameter, directly Control Parameter and inverter performance index are required to set up quantitative relationship in discrete domain, whole inverter has stronger robustness, under various different loads disturbance situations, all can obtain colory ac output voltage; Inverter changes insensitive to inverter parameter, digitial controller parameter, the system responses performance is stable; Can satisfy the high performance index requirement, have obvious superiority.

(5) circuit structure of the present invention is simple, and cost is low, is easy to realize.

Description of drawings

Fig. 1 is the structural representation of the numerically controlled inverter of instant voltage PID current PI;

Fig. 2 is the microprocessor main program flow chart;

Fig. 3 is the control algolithm program flow diagram one among Fig. 2;

Fig. 4 is the schematic circuit block diagram one of Fig. 1;

Fig. 5 is the schematic circuit block diagram two of Fig. 1;

Fig. 6 is the control algolithm program flow diagram two among Fig. 2;

Fig. 7 is the control algolithm program flow diagram three among Fig. 2;

Fig. 8 is the schematic circuit block diagram three of Fig. 1;

Fig. 9 is the schematic circuit block diagram four of Fig. 1.

Embodiment

Below in conjunction with accompanying drawing and example the present invention is described in further detail.

As shown in Figure 1, the structure of the numerically controlled inverter of instant voltage PID current PI of the present invention is:

The control end of the output of current PI digitial controller 8 and inverter 2 joins, the input of the output of inverter 2 and voltage sensor 5 and load 3 are joined, the negative input end of the output of voltage sensor 5 and first subtracter 9 joins, and the positive input terminal of first subtracter 9 receives reference quantity u _rThe input of the output of first subtracter 9 and voltage pid number controller 7 joins, the direct current termination DC power supply 4 of inverter 2, the input of electric current of drawing in the inverter 2 and current sensor 6 joins, the negative input end of the output of current sensor 6 and second subtracter 10 joins, the output of the positive input terminal of second subtracter 10 and voltage pid number controller 7 joins, and the input of the output of second subtracter 10 and current PI digitial controller 8 joins.

Inverter 2, voltage sensor 5 and current sensor 6 can be selected common inverter, voltage sensor and current sensor for use.

First, second subtracter 9,10 and voltage pid number controller 7, current PI digitial controller 8 constitute microprocessor 1.Wherein microprocessor can be single-chip microcomputer or digital signal processing chip.

Microprocessor 1 is gathered the voltage signal of voltage sensor 5 outputs and the current signal of current sensor 6 outputs, according to voltage, current signal and reference quantity, and the calculation control signal, and export inverter 2 to, control inverter 2 work.

Microprocessor 1 and inverter 2 constitute an instant voltage PID current PI numerical control system, current i in the inverter 2 and output voltage u ₀Send into microprocessor 1 through over-current sensor and voltage sensor respectively, microprocessor 1 is through producing control signal u behind the sequential operation ₁Inverter 2 is implemented control, and wherein the current signal i in the inverter 2 can be the filter inductance current i ₁, the filter capacitor current i _cWith load current i _o

The control method that microprocessor 1 during instant voltage PID current PI is digital control is adopted the steps include: as shown in Figure 2

(1) gathers the output voltage u of the current bat that voltage sensor obtains _o(k) and the current i (k) of the current bat that obtains of current sensor, a sampling period T is called a bat in numerical control system, discrete constantly the expression with kT, be abbreviated as k, represent k constantly discrete, its initial value is 0.

(2) calculate the control signal u of next bat ₁(k+1);

Difference according to current signal i in the inverter of gathering 2 comprises the filter inductance current i ₁, the filter capacitor current i _cWith load current i _o, adopt the control signal u of different next bats of algorithm computation ₁(k+1), illustrated respectively below.

(2A) the current signal i when collection is the filter inductance current i ₁Or filter capacitor current i _cThe time, as shown in Figure 3, (2A1)～(2A4) calculates the control signal u of next bat according to step ₁(k+1):

(2A1) utilize formula (A1) to calculate the voltage error signal e of current bat ₁(k), u wherein _r(k) be the reference quantity of current bat:

e ₁(k)＝u _r(k)-u _o(k) (A1)

(2A2) utilize formula (A2) to calculate the given signal u of electric current of current bat _Ir(k):

u _ir(k)＝[k _vp+k _viTz/(z-1)+k _vd(z-1)/(Tz)]e ₁(k) (A2)

K wherein _Vp, k _Vi, k _VdBe respectively ratio, integration, the differential coefficient of voltage pid number controller, z represents the discrete domain operator.

(2A3) utilize formula (A3) to calculate the current error signal e of current bat ₂(k), the current signal i when collection is the filter inductance current i ₁The time, i (k) is the filter inductance current i of current bat ₁(k); When the current signal i that gathers is the filter capacitor current i _cThe time, i (k) is the filter capacitor current i of current bat _c(k):

e ₂(k)＝u _ir(k)-i(k) (A3)

(2A4) utilize formula (A4) to calculate the control signal u of next bat ₁(k+1):

u ₁(k+1)＝[k _ip+k _iiTz/(z-1)]e ₂(k) (A4)

K wherein _Ip, k _IiBe respectively ratio, the integral coefficient of current PI digitial controller.

Fig. 4 be with the current signal i that gathers be the filter inductance current i ₁Corresponding schematic circuit block diagram.As shown in Figure 4, output voltage u _oWith reference quantity u _rThe voltage error signal e that produces relatively ₁Produce the given signal u of electric current through overvoltage pid number controller 7 _Ir, the given signal u of electric current _IrDeduct the filter inductance current i ₁Produce current error signal e ₂, current error signal e ₂Through current PI digitial controller 8 last controlled signal u ₁Inverter 2 is controlled.

Fig. 5 be with the current signal i that gathers be the filter capacitor current i _cCorresponding schematic circuit block diagram.As shown in Figure 5, its structure is similar to Fig. 4, and difference is that the electric current of inverter 2 among Fig. 4 is filter inductance current i ₁, and the electric current of inverter 2 is filter capacitor current i among Fig. 5 _c

(2B) the current signal i when collection is load current i _o, (2B1)～(2B6) calculates the control signal u of next bat according to step ₁(k+1):

(2B1) utilize formula (B1) to calculate the output voltage measured value of next bat

Filter inductance electric current measured value with next bat U wherein ₁(k) be the control signal of current bat, i ₀(k) be the load current of current bat:

$\left[\begin{array}{c}{\hat{u}}_0(k+1)\\ {\hat{i}}_1(k+1)\end{array}\right]=(A_s-H_s{C}_s)\left[\begin{array}{c}{\hat{u}}_0\left(k\right)\\ {\hat{i}}_1\left(k\right)\end{array}\right]+B_s\left[\begin{array}{c}{u}_1\left(k\right)\\ i_0\left(k\right)\end{array}\right]+H_s{C}_s\left[\begin{array}{c}{u}_0\left(k\right)\\ i_1\left(k\right)\end{array}\right]---\left(B1\right)$

$A_s=\left[\begin{array}{cc}{e}^{-\frac{r}{2L}T}\cos {\mathrm{\ω}}_{d}T+\frac{r}{2L{\mathrm{\ω}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\ω}}_{d}T& \frac{1}{C{\mathrm{\ω}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\ω}}_{d}T\\ -\frac{1}{L{\mathrm{\ω}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\ω}}_{d}T& e^{-\frac{r}{2L}T}\cos {\mathrm{\ω}}_{d}T-\frac{r}{2L{\mathrm{\ω}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\ω}}_{d}T\end{array}\right]=\left[\begin{array}{cc}{\mathrm{\φ}}_{11}& {\mathrm{\φ}}_{12}\\ {\mathrm{\φ}}_{21}& {\mathrm{\φ}}_{22}\end{array}\right]$

B _s＝[B ₁ B ₂]

${\mathrm{<figure-callout\; id="1"\; label="B"\; filenames="H2009100608853E00011.png,H2009100608853E00012.png"\; state="\{\{state\}\}">B</figure-callout>}}_1=\left[\begin{array}{c}{e}^{-\frac{r}{2L}T}(-\cos {\mathrm{\&omega;}}_{d}T-\frac{r}{2L{\mathrm{\&omega;}}_d}\sin {\mathrm{\&omega;}}_{d}T)+1\\ \frac{1}{L{\mathrm{\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T\end{array}\right]=\left[\begin{array}{c}{b}_{11}\\ b_{21}\end{array}\right]$

$B_2=\left[\begin{array}{c}r(e^{-\frac{r}{2L}T}\cos {\mathrm{\&omega;}}_{d}T+\frac{r}{2L{\mathrm{\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T-1)-\frac{1}{C{\mathrm{\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T\\ -e^{-\frac{r}{2L}T}\cos {\mathrm{\&omega;}}_{d}T-\frac{r}{2L{\mathrm{\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T+1\end{array}\right]=\left[\begin{array}{c}{b}_{12}\\ b_{22}\end{array}\right]$

C _s＝[1 0]

${\mathrm{\&omega;}}_n=\frac{1}{\sqrt{\mathrm{LC}}},$ Natural frequency of oscillation for inverter 2

${\mathrm{\&omega;}}_d=\sqrt{\frac{1}{\mathrm{LC}}-\frac{r^2}{{4L}^2}},$ Damped oscillation frequency for inverter 2

Wherein L is total filter inductance of inverter 2 outputs, and C is total filter capacitor of inverter 2 outputs, and r is the equivalent damping resistance of inverter 2.

H _sFor the feedback gain matrix of state observer 11 among Fig. 8 and Fig. 9, according to (A _s-H _sC _s) characteristic value select feedback gain matrix H than the fast principle more than 3 times of the closed loop characteristic value of inverter 2 _sGet final product.

(2B2) utilize formula (B2) to calculate the load current measured value of next bat

$\left[\begin{array}{c}{\widehat{\stackrel{.}{i}}}_o(k+1)\\ {\hat{i}}_o(k+1)\end{array}\right]=(A_d-H_d{C}_d)\left[\begin{array}{c}{\widehat{\stackrel{.}{i}}}_o\left(k\right)\\ {\hat{i}}_o\left(k\right)\end{array}\right]+H_d{C}_d\left[\begin{array}{c}{\stackrel{.}{i}}_o\left(k\right)\\ i_o\left(k\right)\end{array}\right]---\left(B2\right)$

$A_d=\left[\begin{array}{cc}1& 0\\ \mathrm{<figure-callout\; id="1"\; label="T"\; filenames="H2009100608853E00011.png,H2009100608853E00012.png"\; state="\{\{state\}\}">T</figure-callout>}& 1\end{array}\right]$

C _d＝[0 1]

H _dFor the feedback gain matrix of disturbance observer 12 among Fig. 8 and Fig. 9, according to (A _d-H _dC _d) characteristic value select feedback gain matrix H than the fast principle more than 5 times of the closed loop characteristic value of inverter 2 _dGet final product.

(2B3) utilize formula (B3) to calculate the voltage error signal measured value ê of next bat ₁(k+1), u wherein _r(k+1) be the reference quantity of next bat:

${\hat{e}}_1(k+1)=u_r(k+1)-{\hat{u}}_0(k+1)---\left(B3\right)$

(2B4) calculate the given signal u of electric current of next bat _Ir(k+1);

According to having or not load-current feedforward, adopt the given signal u of electric current of different next bats of algorithm computation _Ir(k+1), illustrated respectively below.

When no-load current feedovers, utilize formula (B4) to calculate the given signal u of electric current of next bat _Ir(k+1):

u _ir(k+1)＝[k _vp+k _viTz/(z-1)+k _vd(z-1)/(Tz)]ê ₁(k+1) (B4)

When load-current feedforward, as shown in Figure 7, utilize formula (B5) to calculate the predetermined signal u of electric current of next bat _Ir1(k+1):

$u_{\mathrm{ir}1}(k+1)=[k_{\mathrm{vp}}+k_{\mathrm{vi}}\mathrm{Tz}/(z-1)+k_{\mathrm{vd}}(z-1)/\left(\mathrm{Tz}\right)]{\hat{e}}_1(k+1)+{\hat{i}}_o(k+1)---\left(B5\right)$

The predetermined signal u of the electric current of next bat _Ir1(k+1) be the given signal u of electric current of next bat behind the process amplitude limit _Ir(k+1).

(2B5) utilize formula (B6) to calculate the current error signal e of next bat ₂(k+1), wherein, when no-load current feedovers, the electric current measured value of next bat

Filter inductance electric current measured value for next bat

Or the filter capacitor electric current measured value of next bat ${\hat{i}}_c(k+1)={\hat{i}}_1(k+1)-{\hat{i}}_o(k+1);$ When load-current feedforward, the electric current measured value of next bat

Filter inductance electric current measured value for next bat

$e_2(k+1)=u_{\mathrm{ir}}(k+1)-\hat{i}(k+1)---\left(B6\right)$

(2B6) utilize formula (B7) to calculate the control signal u of next bat ₁(k+1):

u ₁(k+1)＝[k _ip+k _iiTz/(z-1)]e ₂(k+1) (B7)

The handling process of the controller in no-load current when feedforward as shown in Figure 6, the handling process of the controller when load-current feedforward is arranged is as shown in Figure 7.

Fig. 8 is the schematic circuit block diagram corresponding with Fig. 6.As shown in Figure 8, state observer 11 is according to the output voltage u of current bat ₀(k) and the load current i of current bat ₀(k) observe the output voltage measured value of next bat

Filter inductance electric current measured value with next bat

Its computing formula is formula (B1).Disturbance observer 12 is according to the load current i of current bat ₀(k) observe the load current measured value of next bat

Its computing formula is formula (B2).The output voltage measured value of next bat

With reference quantity u _r(k+1) the voltage error signal measured value ê of relatively back next bat that produces ₁(k+1) produce the given signal u of electric current of next bat through overvoltage pid number controller 7 _Ir(k+1), the given signal u of the electric current of next bat _Ir(k+1) deduct the electric current measured value of next bat

Produce the current error signal e of next bat ₂(k+1), the electric current measured value of next bat wherein

Filter inductance electric current measured value for next bat

Or the filter capacitor electric current measured value of next bat

The current error signal e of next bat ₂(k+1) process current PI digitial controller 8 obtains the control signal u of next bat at last ₁(k+1) inverter 2 is controlled.

Fig. 9 is the schematic circuit block diagram corresponding with Fig. 7.As shown in Figure 9, state observer 11 is according to the output voltage u of current bat ₀(k) and the load current i of current bat _o(k) observe the output voltage measured value of next bat

Filter inductance electric current measured value with next bat

Its computing formula is formula (B1).Disturbance observer 12 is according to the load current i of current bat _o(k) observe the load current measured value of next bat Its computing formula is formula (B2).The output voltage measured value of next bat With reference quantity u _r(k+1) the voltage error signal measured value ê of relatively back next bat that produces ₁(k+1) give voltage pid number controller 7, the output of voltage pid number controller 7 adds the load current measured value of next bat

Obtain the predetermined signal u of electric current of next bat _Ir1(k+1), the predetermined signal u of the electric current of next bat _Ir1(k+1) be the given signal u of electric current of next bat behind amplitude limit _Ir(k+1), the given signal u of the electric current of next bat _Ir(k+1) deduct the filter inductance electric current measured value of next bat

Produce the current error signal e of next bat ₂(k+1), the current error signal e of next bat ₂(k+1) process current PI digitial controller 8 obtains the control signal u of next bat at last ₁(k+1) inverter 2 is controlled.

For the numerically controlled inverter of instant voltage PID current PI, key is to determine the Control Parameter of voltage pid number controller, current PI digitial controller.On the state space theory basis, use the method for and POLE PLACEMENT USING directly discrete to controlling object, determine voltage pid number controller, current PI digitial controller parameter, now be described as follows:

When adopting the instant voltage PID current PI digital control approach, order

$A=\left[\begin{array}{ccccc}{\mathrm{\&phi;}}_{11}& 0& 0& 0& {\mathrm{\&phi;}}_{12}\\ T{\mathrm{\&phi;}}_{11}& 1& 0& 0& T{\mathrm{\&phi;}}_{12}\\ T^2{\mathrm{\&phi;}}_{11}& \mathrm{<figure-callout\; id="1"\; label="T"\; filenames="H2009100608853E00011.png,H2009100608853E00012.png"\; state="\{\{state\}\}">T</figure-callout>}& 1& 0& T^2{\mathrm{\&phi;}}_{12}\\ 1& 0& 0& 0& 0\\ {\mathrm{\&phi;}}_{21}& 0& 0& 0& {\mathrm{\&phi;}}_{22}\end{array}\right],$ $B=\left[\begin{array}{c}{b}_{11}\\ T{b}_{11}\\ T^2{\mathrm{<figure-callout\; id="11"\; label="b"\; filenames="H2009100608853E00041.png,H2009100608853E00042.png"\; state="\{\{state\}\}">b</figure-callout>}}_{11}\\ 0\\ b_{21}\end{array}\right],$ $P=\left[\begin{array}{c}{z}_1\\ z_2\\ z_3\\ z_4\\ z_5\end{array}\right],$ $K=\left[\begin{array}{c}{k}_1\\ k_2\\ k_3\\ k_4\\ k_5\end{array}\right]$

Wherein, A and B are respectively state matrix, the input matrix after the inverter state equation discretization, and P is a discrete domain expectation closed-loop pole matrix, and K is a gain matrix.Utilization Ackermamn formula is obtained each element of K matrix, can try to achieve voltage pid number controller, each Control Parameter of current PI digitial controller more as follows:

k _ip＝k ₅，

k _vd＝-k ₄·T/k ₅，

Ask equation $(k_4\&CenterDot;T/k_5-C)k_{\mathrm{ii}}^3+({\mathrm{<figure-callout\; id="1"\; label="k"\; filenames="H2009100608853E00011.png,H2009100608853E00012.png"\; state="\{\{state\}\}">k</figure-callout>}}_1+k_4)k_{\mathrm{ii}}^2-k_2\&CenterDot;k_5\&CenterDot;k_{\mathrm{ii}}+k_5^2\&CenterDot;k_3=0$ Real root be k _Ii, that is:

$k_{\mathrm{ii}}=\sqrt[3]{-b/2+\sqrt{b^2/4+a^3/27}}-\sqrt[3]{b/2+\sqrt{b^2/4+a^3/27}},$

Wherein, a=[-3k ₂K ₅/ (k ₄T/k ₅-C)-(k ₁+ k ₄) ²/ (k ₄T/k ₅-C) ²]/3,

$b=[2{({\mathrm{<figure-callout\; id="1"\; label="k"\; filenames="H2009100608853E00011.png,H2009100608853E00012.png"\; state="\{\{state\}\}">k</figure-callout>}}_1+k_4)}^3/{(k_4\&CenterDot;T/k_5-C)}^3+9k_2\&CenterDot;k_5\&CenterDot;({\mathrm{<figure-callout\; id="1"\; label="k"\; filenames="H2009100608853E00011.png,H2009100608853E00012.png"\; state="\{\{state\}\}">k</figure-callout>}}_1+k_4)/{(k_4\&CenterDot;T/k_5-C)}^2+27k_5^2\&CenterDot;k_3/(k_4\&CenterDot;T/k_5-C)]/27,$

k _vp＝[k ₁+k ₄+(k ₄·T/k ₅-C)k _ii]/k ₅，

k _vi＝k ₃/k _ii。

(3) utilize the control signal u of next bat ₁(k+1) inverter 2 is controlled;

(4) make k=k+1, repeating step (1)～(3) are until end-of-job.

The above is preferred embodiment of the present invention, but the present invention should not be confined to the disclosed content of this embodiment and accompanying drawing.So everyly do not break away from the equivalence of finishing under the spirit disclosed in this invention or revise, all fall into the scope of protection of the invention.

Claims (2)

1. numerically controlled inverter of instant voltage PID current PI is characterized in that:

The control end of inverter (2) and microprocessor (1) join, the input of the output of inverter (2) and voltage sensor (5) and load (3) are joined, the input of electric current of drawing in the inverter (2) and current sensor (6) joins, inverter (2) dc terminal links to each other with DC power supply (4), and the output of the output of voltage sensor (5) and current sensor (6) joins with microprocessor (1) respectively;

Described microprocessor (1) comprises voltage pid number controller (7), current PI digitial controller (8) and first, second subtracter (9,10), the control end of the output of current PI digitial controller (8) and inverter (2) joins, the negative input end of the output of voltage sensor (5) and first subtracter (9) joins, and the positive input terminal of first subtracter (9) receives reference quantity u _rThe input of the output of first subtracter (9) and voltage pid number controller (7) joins, the negative input end of the output of current sensor (6) and second subtracter (10) joins, the output of the positive input terminal of second subtracter (10) and voltage pid number controller (7) joins, and the input of the output of second subtracter (10) and current PI digitial controller (8) joins: microprocessor (1) is controlled according to following step:

The output voltage u of the current bat that the 1st step collection voltage sensor obtains _oAnd current bat current i (k) is the filter inductance current i (k) and the current i (k) of the current bat that obtains of current sensor, ₁Or filter capacitor current i _c, wherein, sampling period T is called a bat, and k represents k the discrete moment, and its initial value is 0;

The 2nd step was calculated the control signal u of next bat according to the 2.1st step to the process in the 2.4th step ₁(k+1):

The 2.1st step utilized the formula I to calculate the voltage error signal e of current bat ₁(k), u wherein _r(k) be the reference quantity of current bat:

e ₁(k)＝u _r(k)-u _o(k) Ⅰ

The 2.2nd step utilized the formula II to calculate the given signal u of electric current of current bat _Ir(k):

u _ir(k)＝[k _vp+k _viTz/(z-1)+k _vd(z-1)/(Tz)]e ₁(k)?Ⅱ

K wherein _Vp, k _Vi, k _VdBe respectively ratio, integration, the differential coefficient of voltage pid number controller; Z is the discrete domain operator;

The 2.3rd step utilized the formula III to calculate the current error signal e of current bat ₂(k), the current signal i when collection is the filter inductance current i ₁The time, i (k) is the filter inductance current i of current bat ₁(k); When the current signal i that gathers is the filter capacitor current i _cThe time, i (k) is the filter capacitor current i of current bat _c(k):

e ₂(k)＝u _ir(k)-i(k) Ⅲ

The 2.4th step utilized the formula IV to calculate the control signal u of next bat ₁(k+1), changed for the 3rd step then over to:

u ₁(k+1)＝[k _ip+k _iiTz/(z-1)]e ₂(k)Ⅳ

Wherein, k _Ip, k _IiBe respectively ratio, the integral coefficient of current PI digitial controller, z is the discrete domain operator;

k _Ip, k _Vd, k _Ii, k _Vp, k _ViCalculate according to the following equation and obtain:

k _ip＝k ₅

k _vd＝-k ₄·T/k ₅

k _IiBe equation

Real root;

k _vp＝[k ₁+k ₄+(k ₄·T/k ₅-C)k _ii]/k ₅，

k _vi＝k ₃/k _ii

Wherein, C is total filter capacitor of inverter output, k ₁, k ₂, k ₃, k ₄, k ₅Utilize state matrix and input matrix after the inverter state equation discretization, and discrete domain expectation closed-loop pole matrix computations obtains;

The 3rd step was utilized the control signal u of next bat ₁(k+1) inverter is controlled;

The 4th step made k=k+1, repeated～the 3 step of the 1st step, until shutdown.

2. the numerically controlled inverter of instant voltage PID current PI is characterized in that:

The control end of inverter (2) and microprocessor (1) join, the input of the output of inverter (2) and voltage sensor (5) and load (3) are joined, the input of electric current of drawing in the inverter (2) and current sensor (6) joins, inverter (2) dc terminal links to each other with DC power supply (4), and the output of the output of voltage sensor (5) and current sensor (6) joins with microprocessor (1) respectively;

Described microprocessor (1) comprises voltage pid number controller (7), current PI digitial controller (8) and first, second subtracter (9,10), the control end of the output of current PI digitial controller (8) and inverter (2) joins, the negative input end of the output of voltage sensor (5) and first subtracter (9) joins, and the positive input terminal of first subtracter (9) receives reference quantity u _rThe input of the output of first subtracter (9) and voltage pid number controller (7) joins, the negative input end of the output of current sensor (6) and second subtracter (10) joins, the output of the positive input terminal of second subtracter (10) and voltage pid number controller (7) joins, and the input of the output of second subtracter (10) and current PI digitial controller (8) joins;

Microprocessor (1) also comprises state observer (11) and disturbance observer (12), the input of state observer (11) links to each other with the output of current sensor (6) with voltage sensor (5) respectively, and the input of disturbance observer (12) links to each other with the output of current sensor (6): microprocessor (1) is controlled according to following step:

The output voltage u of the current bat that the 1st step collection voltage sensor obtains _oAnd current bat current i (k) is load current i (k) and the current i (k) of the current bat that obtains of current sensor, _o, wherein, sampling period T is called a bat, and k represents k the discrete moment, and its initial value is 0;

The 2nd step was calculated the control signal u of next bat according to the 2.1st step to the process in the 2.6th step ₁(k+1):

The 2.1st step utilized formula V to calculate the output voltage measured value of next bat

Filter inductance electric current measured value with next bat

U wherein ₁(k) be the control signal of current bat, i ₀(k) be the load current of current bat:

$\left[\begin{array}{c}{\hat{u}}_0(k+1)\\ {\hat{i}}_i(k+1)\end{array}\right]=(A_s-H_s{C}_s)\left[\begin{array}{c}{\hat{u}}_0\left(k\right)\\ {\hat{i}}_1\left(k\right)\end{array}\right]+B_s\left[\begin{array}{c}{u}_1\left(k\right)\\ i_0\left(k\right)\end{array}\right]+H_s{C}_s\left[\begin{array}{c}{u}_0\left(k\right)\\ i_1\left(k\right)\end{array}\right]---V$

$A_s=\left[\begin{array}{cc}{e}^{-\frac{r}{2L}T}{\cos \mathrm{\&omega;}}_{d}T+\frac{r}{{2\mathrm{L\&omega;}}_d}{e}^{-\frac{r}{2L}T}{\sin \mathrm{\&omega;}}_{d}T& \frac{1}{{\mathrm{C\&omega;}}_d}{e}^{-\frac{r}{2L}T}{\sin \mathrm{\&omega;}}_{d}T\\ -\frac{1}{{\mathrm{L\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T& e^{-\frac{r}{2L}T}\cos {\mathrm{\&omega;}}_{d}T-\frac{r}{{2\mathrm{L\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T\end{array}\right]=\left[\begin{array}{cc}{\mathrm{\&phi;}}_{11}& {\mathrm{\&phi;}}_{12}\\ {\mathrm{\&phi;}}_{21}& {\mathrm{\&phi;}}_{22}\end{array}\right]$

B _s＝[B ₁?B ₂]

$B_1=\left[\begin{array}{c}{e}^{-\frac{r}{2L}T}(-{\cos \mathrm{\&omega;}}_{d}T-\frac{r}{{2\mathrm{L\&omega;}}_d}\sin {\mathrm{\&omega;}}_{d}T)+1\\ \frac{1}{{\mathrm{L\&omega;}}_d}{e}^{-\frac{r}{2L}T}{\sin \mathrm{\&omega;}}_{d}T\end{array}\right]=\left[\begin{array}{c}{b}_{11}\\ b_{21}\end{array}\right]$

$B_2=\left[\begin{array}{c}r(e^{-\frac{r}{2L}T}\cos {\mathrm{\&omega;}}_{d}T+\frac{r}{{2\mathrm{L\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T-1)-\frac{1}{{\mathrm{C\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T\\ -e^{-\frac{r}{2L}T}\cos {\mathrm{\&omega;}}_{d}T-\frac{r}{{2\mathrm{L\&omega;}}_d}{e}^{-\frac{r}{2L}T}\sin {\mathrm{\&omega;}}_{d}T+1\end{array}\right]=\left[\begin{array}{c}{b}_{12}\\ b_{22}\end{array}\right]$

C _s＝[1?0]

Natural frequency of oscillation for inverter

Damped oscillation frequency for inverter

Wherein L is total filter inductance of inverter output, and C is total filter capacitor of inverter output, and r is the equivalent damping resistance of inverter; H _sFor the feedback gain matrix of state observer, according to (A _s-H _sC _s) characteristic value select feedback gain matrix H than the fast principle more than 3 times of the closed loop characteristic value of inverter _s

The 2.2nd step utilized the formula VI to calculate the load current measured value of next bat

$\left[\begin{array}{c}{\widehat{\stackrel{.}{i}}}_o(k+1)\\ {\hat{i}}_o(k+1)\end{array}\right]=(A_d-H_d{C}_d)\left[\begin{array}{c}{\widehat{\stackrel{.}{i}}}_o\left(k\right)\\ {\hat{i}}_o\left(k\right)\end{array}\right]+H_d{C}_d\left[\begin{array}{c}{\stackrel{.}{i}}_o\left(k\right)\\ i_o\left(k\right)\end{array}\right]---\mathrm{VI}$

$A_d=\left[\begin{array}{cc}1& 0\\ T& 1\end{array}\right]$

C _d＝[0 1]

H _dFor the feedback gain matrix of disturbance observer, according to (A _d-H _dC _d) characteristic value select feedback gain matrix H than the fast principle more than 5 times of the closed loop characteristic value of inverter _d

The 2.3rd step utilized the formula VII to calculate the voltage error signal measured value of next bat

U wherein _r(k+1) be the reference quantity of next bat:

${\hat{e}}_1(k+1)=u_r(k+1)-{\hat{u}}_0(k+1)---\mathrm{VII}$

The 2.4th step utilized the formula VIII to calculate the given signal u of electric current of next bat when no-load current feedovers _Ir(k+1), when load-current feedforward, utilize the formula IX to calculate the predetermined signal u of electric current of next bat _Ir1(k+1):

$u_{\mathrm{ir}}(k+1)=[k_{\mathrm{vp}}+k_{\mathrm{vi}}\mathrm{Tz}/(z-1)+k_{\mathrm{vd}}(z-1)/\left(\mathrm{Tz}\right)]{\hat{e}}_1(k+1)---\mathrm{VIII}$

$u_{\mathrm{ir}1}(k+1)=[k_{\mathrm{vp}}+k_{\mathrm{vi}}\mathrm{Tz}/(z-1)+k_{\mathrm{vd}}(z-1)/\left(\mathrm{Tz}\right)]{\hat{e}}_1(k+1)+{\hat{i}}_o(k+1)---\mathrm{IX}$

The predetermined signal u of the electric current of next bat _Ir1(k+1) be the given signal u of electric current of next bat behind the process amplitude limit _Ir(k+1); K wherein _Vp, k _Vi, k _VdBe respectively ratio, integration, the differential coefficient of voltage pid number controller; Z is the discrete domain operator;

The 2.5th step utilized the formula X to calculate the current error signal e of next bat ₂(k+1), wherein, when no-load current feedovers, the electric current measured value of next bat

Filter inductance electric current measured value for next bat

Or the filter capacitor electric current measured value of next bat

When load-current feedforward, the electric current measured value of next bat

Filter inductance electric current measured value for next bat

$e_2(k+1)=u_{\mathrm{ir}}(k+1)+\hat{i}(k+1)---X$

The 2.6th step utilized the formula XI to calculate the control signal u of next bat ₁(k+1):

u ₁(k+1)＝[k _ip+k _iiTz/(z-1)]e ₂(k+1) Ⅺ

Wherein, k _Ip, k _IiBe respectively ratio, the integral coefficient of current PI digitial controller, z is the discrete domain operator;

k _Ip, k _Vd, k _Ii, k _Vp, k _ViCalculate according to the following equation and obtain:

k _ip＝k ₅

k _vd＝-k ₄·T/k ₅

k _IiBe equation

Real root;

k _vp＝[k ₁+k ₄+(k ₄·T/k ₅-C)k _ii]/k ₅，

k _v＝k ₃/k _ii

Wherein, C is total filter capacitor of inverter output, k ₁, k ₂, k ₃, k ₄, k ₅Utilize state matrix and input matrix after the inverter state equation discretization, and discrete domain expectation closed-loop pole matrix computations obtains;

The 3rd step was utilized the control signal u of the 2nd next bat of calculating of step ₁(k+1) inverter is controlled;

The 4th step made k=k+1, repeated～the 3 step of the 1st step, until shutdown.

Images