CN115987130A - Improved data-driven energy storage inverter current prediction control method and system - Google Patents
Improved data-driven energy storage inverter current prediction control method and system Download PDFInfo
- Publication number
- CN115987130A CN115987130A CN202310109273.9A CN202310109273A CN115987130A CN 115987130 A CN115987130 A CN 115987130A CN 202310109273 A CN202310109273 A CN 202310109273A CN 115987130 A CN115987130 A CN 115987130A
- Authority
- CN
- China
- Prior art keywords
- current
- basic voltage
- voltage vector
- vectors
- energy storage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Inverter Devices (AREA)
Abstract
The invention discloses an improved current prediction control method and system for a data-driven energy storage inverter, which comprises the following steps: establishing a mathematical model in a static coordinate system of the energy storage inverter, analyzing to obtain a plurality of basic voltage vectors, and synthesizing a plurality of virtual vectors; estimating and compensating the measurement noise; establishing a current gradient relation corresponding to a plurality of basic voltage vectors, and updating the current gradient of the residual vector; predicting current, and analyzing to obtain a plurality of value function values corresponding to the basic voltage vectors and a plurality of virtual vector action time; and analyzing to obtain a plurality of virtual vectors corresponding to the current predicted values and the virtual voltage vector with the minimum value function value, and taking the virtual voltage vector with the minimum value function value as the optimal voltage vector to act on the next control period. The problem of current gradient update stagnation is solved, and current spikes are avoided; the problem that the prediction result is influenced by measurement noise is solved; the robustness of parameter change is improved, and the ripple of output current is reduced.
Description
Technical Field
The invention relates to the technical field of energy storage inverters, in particular to an improved current prediction control method and system of a data-driven energy storage inverter.
Background
In the field of power electronics, data-driven predictive current control is generally used for an energy storage inverter in order to eliminate the influence of system parameters on predictive control, and when the data-driven predictive current control is performed on the energy storage inverter, because the updated gradient through current measurement is only suitable for an applied vector, and the current gradients corresponding to other residual vectors keep old values, the problem that the updating of the current gradients is stopped exists in the data-driven predictive current control of the energy storage inverter, and the stopping of the updating of the current gradients can cause a large current spike in the output current of the energy storage inverter.
In the prior art, when the problem that the updating of the current gradient is stopped in the data-driven prediction current control method of the energy storage inverter is solved, the influence of measurement noise on the current gradient is not considered, so that the current prediction result is easily influenced by the measurement noise, and the error is large; and the number of vectors used is limited, resulting in large output current ripple.
Therefore, a multi-vector energy storage inverter data driving prediction current control method capable of considering measurement noise is needed, and the problem that current gradient updating is stopped in energy storage inverter data driving prediction can be solved.
Disclosure of Invention
The invention aims to provide an improved data-driven energy storage inverter current prediction control method and system, which can realize real-time and rapid update of all current gradients in a lookup table, estimate and compensate measurement noise by using an extended state observer, and solve the problems of large current ripple and easiness in being influenced by measurement noise in the traditional data-driven prediction current control method.
In order to solve the technical problem, the invention provides an improved current prediction control method of a data-driven energy storage inverter, which comprises the following steps:
s1, establishing a mathematical model of an energy storage inverter in a static coordinate system, obtaining a plurality of basic voltage vectors according to the states of three-phase switching tubes of the energy storage inverter, using a current gradient as a prediction model of the energy storage inverter and synthesizing a plurality of virtual vectors;
s2, estimating and compensating the measurement noise of the power grid current measured by the energy storage inverter;
s3, establishing a current gradient relation corresponding to the plurality of basic voltage vectors, and updating current gradients corresponding to residual vectors;
s4, primary current prediction is carried out, and a plurality of value function values corresponding to the plurality of basic voltage vectors and action time of the plurality of virtual vectors are obtained through value function equation analysis;
and S5, analyzing and obtaining current predicted values corresponding to the virtual vectors according to the updated current gradient and the action time of the virtual vectors, obtaining a virtual voltage vector with the minimum value function value through the cost function, and acting the virtual voltage vector as an optimal voltage vector to the next control period.
Optionally, the plurality of basic voltage vectors are eight basic voltage vectors, which are u respectively 0 (0,0,0)、u 1 (1,0,0)、u 2 (1,1,0)、u 3 (0,1,0)、u 4 (0,1,1)、u 5 (0,0,1)、u 6 (1,0,1)、u 7 (1,1,1)；
The mathematical model of the energy storage inverter in the static coordinate system is as follows:
wherein u is x For the output voltage of the energy storage inverter, i g Is the output current of the energy storage inverter, e g The voltage of the power grid side of the energy storage inverter is L, a filter inductor is L, a filter resistor is R, and t is time;
the plurality of virtual vectors is six virtual vectors.
Optionally, in S2, an extended state observer is established to estimate and compensate noise of the measured grid current, where the extended state observer can measure current error feedback, and the extended state observer is:
analyzing the extended state observer yields:
wherein i e For an estimate of the output current of the energy storage inverter, [ delta ] i e For actually measuring the estimated value of the current gradient, δ 1 For feedback error gain of sampled current, delta 2 For sampling feedback error gain, T, of grid current s To control the period, i g For the output current of the energy-storage inverter i e – i g For current error, t is time.
Optionally, the current gradient update formula is expressed as:
wherein Δ i m Current gradient of basic voltage vector for first application, t m Is the action time of the first applied basic voltage vector, A is the proportionality coefficient of the first applied basic voltage vector, t n For the action time of the basic voltage vector of the second application, B is the proportionality coefficient of the basic voltage vector of the second application, Δ i n For the current gradient of the second applied basic voltage vector,. DELTA.i z Current gradient, Δ i, of zero voltage vector x To sample the current gradient of the current,. DELTA.i y Is the current gradient of the residual voltage vector,. DELTA.i 1 Is the current gradient, Δ i, of a first elementary voltage vector of said plurality of elementary voltage vectors 4 Is the current gradient of a fourth basic voltage vector of said plurality of basic voltage vectors, k-1 being the last control moment, T s To control the period, t 1 The action time of a first one of said elementary voltage vectors, k-2 being the last control moment,. DELTA.i 2,6 A current gradient, t, for a second and a sixth basic voltage vector of the plurality of basic voltage vectors z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient, t, for a first and an eighth basic voltage vector of the plurality of basic voltage vectors 4 Is the action time, Δ i, of a fourth elementary voltage vector of said plurality of elementary voltage vectors 3,5 A current gradient, Δ i, for a third and a sixth elementary voltage vector of said plurality of elementary voltage vectors 1,4 A current gradient of a second basic voltage vector and a fourth basic voltage vector of the plurality of basic voltage vectors.
Optionally, in S4, the multiple basic voltage vectors are added to the sampled current to perform current prediction, where the current prediction formula is:
wherein i iαβ Is the power grid current under the static coordinate system, k +1 is the next control moment, i αβ Is a sampled current in a stationary coordinate system,. DELTA.i iαβ The current gradient corresponding to the ith basic voltage vector is obtained, and k is the corresponding moment;
the cost function is:
wherein G is a value of a cost function, i refα For the reference current, i, of the energy storage inverter in the alpha-static coordinate system gα For the power grid current of the energy storage inverter under an alpha static coordinate system, k +1 is the next control moment, i refβ For the reference current i of the energy-storage inverter in the beta stationary coordinate system gβ And the grid current of the energy storage inverter under a beta static coordinate system is obtained.
Optionally, in S5, substituting the value obtained by predicting the current into the cost function to obtain a corresponding cost function value;
the action time of the plurality of virtual vectors is as follows:
wherein, t m Acting time of basic voltage vector for first application, t n Acting time of basic voltage vector for second application, t z The action times of a first and an eighth basic voltage vector of said basic voltage vectors, G n Value of the fundamental voltage vector for the second application, G z Value of the zero-voltage vector, G m Value of the cost function of the basic voltage vector for the first application, T s Is a control cycle.
Optionally, when three basic voltage vectors are applied in each control cycle, the predicted value of the current at the time k +1 is:
wherein i g Is the output current of the energy storage inverter, t m The action time of the first applied basic voltage vector, k +1 the next control moment, k the current control moment, T s To control the period,. DELTA.i m A current gradient, t, for an mth basic voltage vector of the plurality of basic voltage vectors n The action time of the basic voltage vector for the second application of the basic voltage vector,. DELTA.i n Is the current gradient of the nth basic voltage vector of the plurality of basic voltage vectors, k is the current control time, t z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient being a first and an eighth base voltage vector of the plurality of base voltage vectors;
the predicted value of the current at the time k +2 is:
wherein i g For the output current of the energy storage inverter, k +2 is the next control time, k +1 is the next control time, t m The action time, T, of the basic voltage vector for the first application of the basic voltage vector s To control the period,. DELTA.i m A current gradient, t, for an mth basic voltage vector of the plurality of basic voltage vectors n The action time of the basic voltage vector for the second application of the basic voltage vector,. DELTA.i n Is the current gradient of the nth basic voltage vector of the plurality of basic voltage vectors, t z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient of a first basic voltage vector and an eighth basic voltage vector of the plurality of basic voltage vectors.
The invention also provides an improved data-driven energy storage inverter current prediction control system, which comprises:
the parameter setting module is used for establishing a mathematical model of the energy storage inverter in a static coordinate system, obtaining a plurality of basic voltage vectors according to the state of a three-phase switching tube of the energy storage inverter, and taking a current gradient as a prediction model of the energy storage inverter and synthesizing a plurality of virtual vectors;
the noise estimation module is used for estimating and compensating the measurement noise of the grid current measured by the energy storage inverter;
the current gradient updating module is used for establishing a current gradient relation corresponding to the plurality of basic voltage vectors and updating current gradients corresponding to other residual vectors;
and the current prediction module is used for obtaining a plurality of cost function values corresponding to the plurality of basic voltage vectors through the analysis of a cost function equation, obtaining the action time of the plurality of virtual vectors, obtaining a current prediction value corresponding to the plurality of virtual vectors through the analysis according to the updated current gradient and the action time of the plurality of virtual vectors, obtaining a virtual voltage vector with the minimum value function value through the cost function, and acting the virtual voltage vector as an optimal voltage vector to the next control period.
Compared with the prior art, the invention at least has the following beneficial effects:
according to the method, a model is established according to the coordinate relation of voltage vectors through an improved data-driven energy storage inverter current prediction control method, and the current gradient is rapidly updated in real time through a current gradient updating formula, so that the problem that the current gradient updating is stagnated in the energy storage inverter data-driven prediction current control is solved, the current spike of the output current of the energy storage inverter is eliminated, and the prediction error is reduced;
the measurement noise of the measured power grid current in the energy storage inverter data drive prediction current control is considered, the measurement noise of the measured power grid current is estimated and compensated, the problem that the energy storage inverter data drive prediction current control is easily affected by the measurement noise is solved under the condition of high dynamic response speed, and the problem of large current prediction error is further solved;
furthermore, three vectors are applied in each control period, so that the robustness of parameter change is improved, the voltage vector acted in each control period is increased under the condition of not increasing a lookup table, the influence of sampling disturbance and the ripple of output current are reduced, and the quality of the output current is improved;
further, the current prediction of the improved data-driven energy storage inverter current prediction control method is based on the current gradients stored in the lookup table, real-time update of all the current gradients in the lookup table is realized without any system parameters, and good current performance can still be maintained when the parameters are mismatched.
Drawings
FIG. 1 is a schematic control flow diagram according to an embodiment of the present invention;
FIG. 2 is a topology diagram of an energy storage inverter according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a synthesized three-vector according to an embodiment of the present invention;
FIG. 4 is a control block diagram of an extended state observer according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a current gradient synthesis method according to an embodiment of the present invention;
FIG. 6 is a schematic diagram illustrating real-time current gradient updating according to an embodiment of the present invention;
FIG. 7a is a first current gradient simulation diagram of a current gradient update method in the prior art;
FIG. 7b is a current gradient simulation diagram II of a current gradient update method in the prior art;
FIG. 7c is a current gradient simulation diagram III of a current gradient update method in the prior art;
FIG. 7d is a current gradient simulation diagram of a current gradient update method in the prior art;
FIG. 7e is a simulation diagram I of the current gradient update method according to an embodiment of the present invention;
FIG. 7f is a simulation diagram of a current gradient update method according to an embodiment of the present invention;
FIG. 8a is a current gradient plot of three vector data driven predictive current control injected noise in accordance with an embodiment of the present invention;
FIG. 8b is a current gradient plot for three vector data driven predictive current control based on an extended state observer in accordance with the present invention;
FIG. 9a is a graph of a prior art current gradient update experiment at 10A;
FIG. 9b is a graph of a prior art current gradient update experiment at 10A;
FIG. 9c is a graph of an experiment for updating current gradient of three-vector data-driven predictive current control according to an embodiment of the present invention,
FIG. 9d is a current gradient update experimental graph of three-vector data-driven predictive current control based on the extended state observer according to the embodiment of the present invention;
FIG. 10a is an experimental diagram of three-vector data driving prediction current control when the parameters of the filter inductor controller are not doubled with the actual parameters in the prior art;
FIG. 10b is an experimental graph of three-vector data-driven predictive current control based on an extended state observer when the filter inductor controller parameters are not multiplied by the actual parameters in accordance with an embodiment of the present invention;
FIG. 11a is an experimental diagram of a three-vector data-driven predictive current control when the reference current is changed from 5A to 10A in the prior art;
FIG. 11b is an experimental graph of three vector data driven predictive current control when the reference current is changed from 5A to 10A in accordance with one embodiment of the present invention;
FIG. 11c is an experimental diagram of the three vector data-driven predictive current control based on the extended state observer when the reference current is changed from 5A to 10A in accordance with one embodiment of the present invention.
Detailed Description
The improved data-driven energy storage inverter current predictive control method and system of the present invention will now be described in greater detail with reference to the schematic drawings in which preferred embodiments of the invention are shown, it being understood that those skilled in the art may modify the invention herein described while still achieving the advantageous results of the invention. Accordingly, the following description should be construed as broadly as possible to those skilled in the art and not as limiting the invention.
The invention is described in more detail in the following paragraphs by way of example with reference to the accompanying drawings. Advantages and features of the present invention will become apparent from the following description and from the claims. It is to be noted that the drawings are in a very simplified form and are not to precise scale, which is merely for the purpose of facilitating and distinctly claiming the embodiments of the present invention.
The invention discloses an improved current prediction control method of a data-driven energy storage inverter, which comprises the following steps, please refer to fig. 1:
s1, establishing a mathematical model of an energy storage inverter in a static coordinate system, obtaining a plurality of basic voltage vectors according to the states of three-phase switching tubes of the energy storage inverter, using a current gradient as a prediction model of the energy storage inverter, and synthesizing a plurality of virtual vectors, wherein the plurality of virtual vectors refer to fig. 3.
And S2, estimating and compensating the measurement noise of the energy storage inverter for measuring the power grid current.
And S3, establishing a current gradient relation corresponding to the plurality of basic voltage vectors, and updating current gradients corresponding to other residual vectors, wherein a schematic diagram of a current gradient synthesis mode is shown in FIG. 5, and a schematic diagram of a current gradient real-time updating is shown in FIG. 6.
And S4, performing primary current prediction, and analyzing through a cost function equation to obtain a plurality of cost function values corresponding to the plurality of basic voltage vectors and action time of the plurality of virtual vectors.
And S5, analyzing and obtaining current predicted values corresponding to the virtual vectors according to the updated current gradient and the action time of the virtual vectors, obtaining a virtual voltage vector with the minimum value function value through the cost function, and acting the virtual voltage vector as an optimal voltage vector to the next control period.
In a specific example, the plurality of basic voltage vectors are eight basic voltage vectors, each of which is u 0 (0,0,0)、u 1 (1,0,0)、u 2 (1,1,0)、u 3 (0,1,0)、u 4 (0,1,1)、u 5 (0,0,1)、u 6 (1,0,1)、u 7 (1, 1), please refer to FIG. 3 for a related diagram.
The mathematical model of the energy storage inverter in the static coordinate system is as follows:
wherein u is x For the output voltage of the energy-storage inverter, i g Is the output current of the energy storage inverter, e g And the voltage of the power grid side of the energy storage inverter is L, the filter inductor is L, the filter resistor is R, and t is time.
The plurality of virtual vectors is six virtual vectors, as shown in fig. 3.
Further, referring to fig. 4, in S2, an extended state observer is established to estimate and compensate the noise of the measured grid current, where the extended state observer is capable of measuring a current error feedback, and the extended state observer is:
analyzing the extended state observer yields:
wherein i e For an estimate of the output current of the energy storage inverter, [ delta ] i e For actually measuring the estimated value of the current gradient, δ 1 For feedback error gain of sampled current, delta 2 For sampling feedback error gain, T, of grid current s To control the period, i g For the output current of the energy-storage inverter i e – i g For current error, t is time.
As an example, referring to fig. 5 and 6, the current gradient update formula can be expressed as:
wherein Δ i m Current gradient of basic voltage vector for first application, t m Is the action time of the first applied basic voltage vector, A is the proportionality coefficient of the first applied basic voltage vector, t n For the action time of the basic voltage vector of the second application, B is the proportionality coefficient of the basic voltage vector of the second application, Δ i n For the current gradient of the second applied basic voltage vector,. DELTA.i z Current gradient, Δ i, of zero voltage vector x To sample the current gradient of the current,. DELTA.i y Is the current gradient, Δ i, of the residual voltage vector 1 Is the current gradient, Δ i, of a first elementary voltage vector of said plurality of elementary voltage vectors 4 Is the current gradient of a fourth basic voltage vector of said plurality of basic voltage vectors, k-1 is the last control instant, T s To control the period, t 1 The action time of a first one of said elementary voltage vectors, k-2 being the last control moment,. DELTA.i 2,6 A current gradient, t, for a second and a sixth basic voltage vector of the plurality of basic voltage vectors z Is a first basic voltage vector of the plurality of basic voltage vectorsAnd the action time, Δ i, of the eighth fundamental voltage vector z A current gradient, t, for a first and an eighth basic voltage vector of the plurality of basic voltage vectors 4 Is the action time, Δ i, of a fourth elementary voltage vector of said plurality of elementary voltage vectors 3,5 A current gradient, Δ i, for a third and a sixth elementary voltage vector of said plurality of elementary voltage vectors 1,4 A current gradient of a second basic voltage vector and a fourth basic voltage vector of the plurality of basic voltage vectors.
In one embodiment of the present invention, in S4, the plurality of basic voltage vectors are added to the sampled current to perform current prediction, and the current prediction formula is:
wherein i iαβ Is the power grid current under the static coordinate system, k +1 is the next control moment, i αβ Is a sampling current in a stationary coordinate system,. DELTA.i iαβ The current gradient corresponding to the ith basic voltage vector is shown, and k is the corresponding moment.
The cost function is:
wherein G is a value of a cost function, i refα For the reference current i of the energy-storage inverter in the alpha stationary coordinate system gα For the power grid current of the energy storage inverter under an alpha static coordinate system, k +1 is the next control moment, i refβ For the reference current, i, of the energy storage inverter in a beta stationary coordinate system gβ And the grid current of the energy storage inverter under a beta static coordinate system is obtained.
Further, in S5, the value obtained by predicting the current is substituted into the cost function to obtain a corresponding cost function value.
The action time of the plurality of virtual vectors is as follows:
wherein, t m Is a basic voltage vector u m Time of action of t n Is a basic voltage vector u n Time of action of t z The action times of a first and an eighth basic voltage vector of said basic voltage vectors, G n Is u n Value of (G) z Is u z Value of (G) m Is u m Value of (T) s Is a control cycle.
Specifically, when three basic voltage vectors are applied per control cycle, the predicted value of the current at the time k +1 is:
wherein i g Is the output current of the energy storage inverter, t m Is a basic voltage vector u m K +1 is the next control time, k is the current control time, T s To control the period,. DELTA.i m A current gradient of an mth basic voltage vector of the plurality of basic voltage vectors, t is t n Is a basic voltage vector u m Time of action,. DELTA.i n Is the current gradient of the nth basic voltage vector of the plurality of basic voltage vectors, k is the current control time, t z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient of a first basic voltage vector and an eighth basic voltage vector of the plurality of basic voltage vectors.
The predicted value of the current at the time k +2 is:
wherein i g For the output current of the energy storage inverter, k +2 is the next control timeMoment, k +1 is the next control moment, t m The action time, T, of the basic voltage vector for the first application of the basic voltage vector s To control the period,. DELTA.i m A current gradient, t, for an mth basic voltage vector of the plurality of basic voltage vectors n The action time of the basic voltage vector for the second application of the basic voltage vector,. DELTA.i n Is the current gradient of the nth basic voltage vector of the plurality of basic voltage vectors, t z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient of a first basic voltage vector and an eighth basic voltage vector of the plurality of basic voltage vectors.
And obtaining a virtual voltage vector with the minimum value function value through the value function and the current predicted value, and acting the virtual voltage vector as an optimal voltage vector in the next control period.
In another aspect of the present invention, there is also included an improved data-driven energy storage inverter current predictive control system comprising:
the parameter setting module is used for establishing a mathematical model of the energy storage inverter in a static coordinate system, obtaining a plurality of basic voltage vectors according to the states of three-phase switching tubes of the energy storage inverter, and taking a current gradient as a prediction model of the energy storage inverter and synthesizing a plurality of virtual vectors;
the noise estimation module is used for estimating and compensating the measurement noise of the energy storage inverter for measuring the power grid current;
the current gradient updating module is used for establishing a current gradient relation corresponding to the plurality of basic voltage vectors and updating current gradients corresponding to other residual vectors;
and the current prediction module is used for obtaining a plurality of value function values corresponding to the plurality of basic voltage vectors through the analysis of a value function equation, obtaining the action time of the plurality of virtual vectors, analyzing and obtaining current prediction values corresponding to the plurality of virtual vectors according to the updated current gradient and the action time of the plurality of virtual vectors, obtaining a virtual voltage vector with the minimum value function value through the value function, and acting the virtual voltage vector as an optimal voltage vector to the next control period.
In an embodiment of the present invention, the method for predicting and controlling the current of the data-driven energy storage inverter is performed as follows, specifically referring to fig. 1:
step one, according to the topology of the energy storage inverter shown in fig. 2, a mathematical model of the energy storage inverter in a static coordinate system is established, and eight basic voltage vectors are obtained according to the three-phase switch tube state of the energy storage inverter, wherein the eight basic voltage vectors are u 0 (0,0,0)、u 1 (1,0,0)、u 2 (1,1,0)、u 3 (0,1,0)、u 4 (0,1,1)、u 5 (0,0,1)、u 6 (1,0,1)、u 7 (1, 1), using the current gradient as a prediction model of the energy storage inverter, and synthesizing to obtain six virtual vectors, please refer to fig. 3.
Specifically, although the data-driven predictive current control method in the prior art can use the measured current gradient as a predictive model of the energy storage inverter, thereby eliminating the influence of model parameters, the output current performance of the data-driven predictive current control method in the prior art is affected by the number of application vectors, the current gradient update stagnation, and the sampling disturbance. In order to solve the above problem, the present embodiment synthesizes a virtual voltage vector as shown in fig. 3 based on eight basic voltage vectors of the energy storage inverter.
Three-phase power grid current i obtained by sampling abc (ii) subjecting to Clark transformation to obtain i αβ :
Wherein i α Is the grid current i in the alpha stationary coordinate system β Is the grid current i in a beta static coordinate system a For a-phase grid current, i b For b-phase grid current, i c Is the c-phase grid current.
Establishing a mathematical model according to the energy storage inverter topology shown in fig. 2, and obtaining the mathematical model of the two-level energy storage inverter in a static coordinate system as follows:
wherein u is x = [u xα ，u xβ ] T For the output voltage of the energy-storage inverter, i g = [i gα ，i gβ ] T For the output current of the energy-storage inverter, e g = [e gα ，e gβ ] T The voltage of the grid side of the energy storage inverter is L, the filter inductance is L, and the parasitic resistance is R.
According to the forward eulerian method, the prediction model of the energy storage inverter at the time k +1 is as follows:
wherein i g For the output current of the energy storage inverter, k +1 is the next control time, k is the current control time, T s For control period, L is filter inductance, R is parasitic resistance, and Δ i x (k) Is a vector u x (k) Gradient of current under influence u x For the output voltage of the energy-storage inverter, e g The grid side voltage of the energy storage inverter.
To eliminate the effect of model parameters on the predicted current, the current gradient can be expressed as:
wherein Δ i x (k-1) is the current gradient at the previous time, k-1 is the previous control time, i g And k is the output current of the energy storage inverter and the current control moment.
As can be seen from the above equation, the current gradient measured at time k should correspond to the vector applied from time k-1 to time k, and the obtained result is stored in the lookup table for current prediction in the subsequent steps.
In summary, the predicted current based on the measured current gradient can be expressed as:
wherein i g Is the output current of the energy storage inverter, k is the current control time, k +1 is the next control time, Δ i x (k) The current gradient at the present moment.
Secondly, applying an extended state observer, referring to fig. 4, estimating and compensating measurement noise of the measured power grid current, reducing the influence of measurement disturbance on a prediction result, and ensuring the stability of the output current performance; the extended state observer with measured current error feedback is represented as:
wherein i e Is i g Is estimated by Δ i e Is Δ i s Estimate of i e – i g As a current error, δ 1 For feedback error gain of sampled current, delta 2 The feedback error gain of the current of the power grid is sampled.
The above formula can be discretized as:
wherein i e (k-1) is an estimate of the last moment in the grid current, i g (k-1) is a sampling value of the power grid current at the previous moment, and err (k-1) is an error between the power grid current estimation value and the sampling value at the previous moment; i.e. i e (k) As an estimate of the current time of the grid, Δ i e (k-1) is an estimated value of the current gradient at the previous moment; Δ i e (k) The current gradient is an estimated value of the current moment; delta 1 For feedback error gain of sampled current, delta 2 For sampling grid currentsFeeding error gain; t is s Is a control cycle.
To derive the error feedback gain, the matrix form of the extended state observer can be expressed as:
wherein i e (k-1) is an estimate of the last moment in the grid current, i g (k-1) is a sampling value of the power grid current at the previous moment, and err (k-1) is an error between the power grid current estimation value and the sampling value at the previous moment; i.e. i e (k) As an estimate of the current time of the grid, Δ i e (k-1) is an estimated value of the current gradient at the previous moment; delta i e (k) The current gradient is an estimated value of the current moment; delta 1 For feedback error gain of sampled current, delta 2 Sampling feedback error gain of the power grid current; t is s Is a control cycle.
The characteristic polynomial of the above matrix can be expressed as:
wherein z is the observer state vector and I is the second order identity matrix.
In order to obtain the fastest dynamic performance, both eigenvalues of the above-mentioned characteristic polynomial are chosen to be zero to obtain a dead-beat characteristic, i.e.:
analyzing and calculating the characteristic polynomial to obtain:
wherein z is the observer state vector, G is the value of the cost function, δ 1 For feedback error gain of sampled current, delta 2 Sampling feedback error gain of the power grid current; t is s Is a control cycle.
And thirdly, referring to fig. 5 and 6, establishing a current gradient relationship corresponding to different vectors, and updating current gradients corresponding to remaining vectors according to the current gradient relationship corresponding to the different vectors.
The virtual vector applied between the time k-1 and the time k consists of three elementary vectors u m 、u n And u z In composition, the resulting current gradient Δ i is measured x (k-1) by [ Delta ] i m (k-1)、△i n (k-1) and Δ i z (k-1).
The current gradient can be expressed as:
wherein Δ i x (k-1) is the current gradient at the previous moment, i g Is the output current of the energy storage inverter, k is the current control time, k-1 is the last control time, t m The action time, T, of the basic voltage vector for the first application of the basic voltage vector s To control the period,. DELTA.i m A current gradient, t, for an mth basic voltage vector of the plurality of basic voltage vectors n The action time of the basic voltage vector for the second application of the basic voltage vector,. DELTA.i n Is the current gradient of the nth basic voltage vector of the plurality of basic voltage vectors, t z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient of a first basic voltage vector and an eighth basic voltage vector of the plurality of basic voltage vectors.
t z A 2 is u 0 Or u 7 Time of action of u m And u n Is a base vector, t m And t n Are each u m And u n The action time of (1).
Current gradient Δ i m (k-1)、△i n (k-1) and Δ i z (k-1) are unknown quantities, wherein the current gradient Δ i x (k-1) can be regarded as a sum [ u ] x (k-1) - Ri g (k-1) - e g (k-1)]T s and/L is equal.
Selecting a voltage vector u m 、u n And u z And obtaining a current gradient equation at the k-1 moment as follows:
similarly, at time k-2,. DELTA.i m (k-2)、△i n (k-2) and Δ i z The current gradient equation between (k-2) can be expressed as:
and (3) calculating to obtain a new circuit gradient equation:
wherein the content of the first and second substances,
further, the current gradient Delta i is obtained through analysis and calculation m (k-1)、△i n (k-1) and Δ i z (k-1) is:
wherein L is filter inductance, Δ i m Current gradient of basic voltage vector for first application, t m Is the action time of the first applied basic voltage vector, A is the proportionality coefficient of the first applied basic voltage vector, t n For the action time of the basic voltage vector of the second application, B is the proportionality coefficient of the basic voltage vector of the second application, Δ i n For the current gradient of the second applied basic voltage vector,. DELTA.i x For sampling currentFlow gradient,. DELTA.i y Is the current gradient of the residual voltage vector,. DELTA.i 1 Is the current gradient, Δ i, of a first elementary voltage vector of said plurality of elementary voltage vectors 4 Is the current gradient of a fourth basic voltage vector of said plurality of basic voltage vectors, k-1 being the last control moment, T s To control the period, t 1 The action time of a first one of said elementary voltage vectors, k-2 being the last control moment,. DELTA.i 2,6 A current gradient, t, for a second and a sixth basic voltage vector of the plurality of basic voltage vectors z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient, t, for a first and an eighth basic voltage vector of the plurality of basic voltage vectors 4 Is the action time, Δ i, of a fourth elementary voltage vector of said plurality of elementary voltage vectors 3,5 A current gradient, Δ i, for a third and a sixth elementary voltage vector of said plurality of elementary voltage vectors 1,4 A current gradient of a second basic voltage vector and a fourth basic voltage vector of the plurality of basic voltage vectors.
Through the steps, the current gradient values under the action of the three virtual voltage vectors applied in the previous control period are updated.
Similarly, a current gradient equation of the residual vector and the zero vector is established to update the residual four residual vectors u y (k-1) Current gradient value Δ i y (k-1). The update equation for the residual current gradient is derived as:
where C is the scale factor of the residual vector,. DELTA.i y Is the current gradient, Δ i, of the residual voltage vector z The current gradients of a first basic voltage vector and an eighth basic voltage vector of the plurality of basic voltage vectors are obtained, k-1 is the last control moment, and k-2 is the last control moment; and is
Where C is the scale factor of the residual vector, u y (k-1) is the coordinate component of the residual vector in the last control period, u y (k-2) is the coordinate component of the last control period of the residual vector, u z (k-1) is the coordinate component of the zero vector in the last control period, u z And (k-2) is the coordinate component of the last control period of the zero vector.
When the denominators of a, B and C of the current gradient formula are not 0, the values of a, B and C are about equal to 1 because the dc voltage is kept constant in one control period.
Thus, Δ i m (k-1)、△i n (k-1)、△i z (k-1) and Δ i y The value of (k-1) can be expressed as:
wherein Δ i m For the current gradient of the first applied basic voltage vector,. DELTA.i n For the current gradient of the second applied basic voltage vector,. DELTA.i z A current gradient, Δ i, for a first and an eighth of said plurality of elementary voltage vectors x To sample the current gradient of the current,. DELTA.i y Current gradient of residual voltage vector, t m Acting time of basic voltage vector for first application, t n Acting time of basic voltage vector for second application, T s For the control period, k-1 is the last control time, and k-2 is the last control time.
However, the above update equation fails when the denominators a, B, and C of the above current gradient equation approach 0.
At this time, a△i 4 (k-1) and Δ i 3 (k-1)，△i 5 (k-1) equation relationship, thereby realizing Δ i 4 And (k-1) updating.
When Δ i 1 (k-1) and Δ i 4 (k-1) when the current gradient as the application vector is updated, the update formula can be expressed as:
when Δ i 1 (k-1) and Δ i 4 (k-1) when the residual current gradient as the residual vector is updated, the update formula may be expressed as:
wherein Δ i m Current gradient of basic voltage vector for first application, t m Is the action time of the first applied basic voltage vector, A is the proportionality coefficient of the first applied basic voltage vector, t n For the action time of the basic voltage vector of the second application, B is the proportionality coefficient of the basic voltage vector of the second application, Δ i n For the current gradient of the second applied basic voltage vector,. DELTA.i z Current gradient, Δ i, of zero voltage vector x To sample the current gradient of the current,. DELTA.i y Is the current gradient, Δ i, of the residual voltage vector 1 Is the current gradient, Δ i, of a first elementary voltage vector of said plurality of elementary voltage vectors 4 Is the current gradient of a fourth basic voltage vector of said plurality of basic voltage vectors, k-1 being the last control moment, T s To control the period, t 1 The action time of a first one of said elementary voltage vectors, k-2 being the last control moment,. DELTA.i 2,6 A current gradient, t, for a second and a sixth basic voltage vector of the plurality of basic voltage vectors z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z For said plurality of basic voltage vectorsCurrent gradients of the first and eighth basic voltage vectors of magnitude, t 4 Is the action time, Δ i, of a fourth elementary voltage vector of said plurality of elementary voltage vectors 3,5 A current gradient, Δ i, for a third and a sixth elementary voltage vector of said plurality of elementary voltage vectors 1,4 A current gradient of a second basic voltage vector and a fourth basic voltage vector of the plurality of basic voltage vectors.
In conclusion, the method realizes the real-time updating of all the current gradients in each control period, and completely eliminates the stagnation phenomenon of the updating of the current gradients.
And step four, adding the current gradients corresponding to the eight basic voltage vectors and the sampling current to predict current, substituting the corresponding eight predicted currents into a cost function equation to obtain corresponding eight cost function values, and calculating to obtain the action time of each vector in the virtual vector due to the fact that the action time of the vector is in inverse proportion to the cost function value.
After the update of all current gradients is completed, the vector action time of the next control cycle is calculated.
Firstly, current prediction is carried out according to the current gradient updated by each voltage vector:
wherein i iαβ Is the power grid current under the static coordinate system, k +1 is the next control moment, i αβ Is a sampling current in a static coordinate system, delta i iαβ The current gradient corresponding to the ith basic voltage vector, k is the corresponding time, delta i iαβ (k) For the current gradient corresponding to the ith fundamental voltage vector, i =1,2, ..., 8,i iαβ And (k + 1) is a predicted current corresponding to the ith basic voltage vector.
Wherein the action time of each vector can be expressed as:
wherein, t m Acting time of basic voltage vector for first application, t n Acting time of basic voltage vector for second application, t z The action times of a first and an eighth basic voltage vector of said basic voltage vectors, G n Value of the fundamental voltage vector for the second application, G z Value of the zero-voltage vector, G m Value of the cost function of the basic voltage vector for the first application, T s Is a control period; g m 、G n And G z Are each u m 、u n And u z A value of the cost function; u. of z Represents u 0 And u 7 。
Said value the function is:
wherein G is a value of a cost function, i refα For the reference current i of the energy-storage inverter in the alpha stationary coordinate system gα For the power grid current of the energy storage inverter under an alpha static coordinate system, k +1 is the next control moment, i refβ For the reference current, i, of the energy storage inverter in a beta stationary coordinate system gβ And the grid current of the energy storage inverter under a beta static coordinate system is obtained.
And step five, obtaining current predicted values corresponding to the six virtual vectors through analysis and calculation according to the updated current gradient and time obtained in the step five, substituting the current predicted values into the cost function for evaluation, obtaining a virtual voltage vector with the minimum cost function value as an optimal voltage vector, and applying the optimal voltage vector to the next control period.
When three basic voltage vectors are applied per control cycle, the predicted current at time k +1 is expressed as:
to compensate for the control delay of one cycle, the predicted current at time k +2 is expressed as:
wherein i g Is the output current of the energy storage inverter, t m The action time of the first applied basic voltage vector, k +2 the next control moment, k +1 the next control moment, k the current control moment, T s To control the period,. DELTA.i m A current gradient, t, for an mth basic voltage vector of the plurality of basic voltage vectors n The action time of the basic voltage vector for the second application of the basic voltage vector,. DELTA.i n Is the current gradient of the nth basic voltage vector of the multiple basic voltage vectors, k is the current control time, t z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient of a first basic voltage vector and an eighth basic voltage vector of the plurality of basic voltage vectors.
Substituting the 6 predicted currents obtained by the above formula into the cost function to perform optimization, and applying the virtual vector with the lowest cost function value as the optimal vector in the next control period, wherein the control block diagram is shown in fig. 6.
In a specific experiment, in order to verify the effectiveness of the improved data-driven energy storage inverter current prediction control method, the steady-state performance of the improved data-driven energy storage inverter current prediction control method under different updating methods is tested and compared with the tested data-driven prediction current control method in the prior art.
In the first experiment, please refer to fig. 7a to 7f, in which the method for controlling the three-vector data-driven predictive current of the energy storage inverter according to the embodiment of the present invention has better performance than the prior art.
In the second experiment, please refer to fig. 8a to 8b, which show that the method for controlling the three-vector data-driven prediction current of the energy storage inverter according to the embodiment of the present invention has better performance than the prior art.
In experiments three to five, matlab/Simulink simulations were used to verify the effectiveness of the proposed strategy, using the following table for the relevant parameters:
in the third experiment, please refer to fig. 9a, the long-time stagnation phenomenon occurs in the update method in the prior art, which causes the current spike in the output current ripple. Referring to fig. 9b, although the prior art update method effectively improves the stagnation phenomenon, the total harmonic distortion is reduced from 2.77% to 2.42%, but the current spike still exists in the output current ripple because the stagnation is not completely eliminated. Referring to fig. 9c, in the method for controlling a three-vector data-driven prediction current according to an embodiment of the present invention, the total harmonic distortion is reduced to 2.20%; further, referring to fig. 9d, in the method for controlling the prediction current driven by the three-vector data based on the extended state observer according to the embodiment of the present invention, the total harmonic distortion is further reduced to 1.71%, the stagnation phenomenon is eliminated, and the prediction error caused by the sampling disturbance is suppressed. In summary, the updating method and the sampling disturbance suppression method of the energy storage inverter three-vector data driving prediction current control method have the effect superior to that of the prior art.
In the fourth experiment, please refer to fig. 10A-10b, the reference current is set to 10A, and the filter inductor controller parameter is not doubled with the actual parameter, specifically, referring to fig. 10A, when the three-vector data driving prediction current control method in the prior art is applied, as the inductance parameter error increases, the current ripple and the prediction error increase with the increase of the inductance parameter. In particular, referring to FIG. 10a, it can be seen that when L is 0 Near 0.2L, the prediction error of the current is the largest, and the current ripple is distorted at this time. Referring to FIG. 10b, the three-vector data-driven predictive current control method of the prior art is similar to that of the prior artBy comparison, the current quality of the three-vector data-driven prediction current control method based on the expansion state sensor is not affected by parameter mismatch. In conclusion, compared with the data-driven predictive current control method in the prior art, the three-vector data-driven predictive current control method for the energy storage inverter has better parameter robustness.
In the fifth experiment, the reference current is set to be 10A from 5A, please refer to fig. 11a to 11c, fig. 11a shows the three-phase current waveform, the dynamic response and the prediction error of the three-vector data driving prediction current control method in the prior art, fig. 11b shows the three-phase current waveform, the dynamic response and the prediction error of the three-vector data driving prediction current control method according to the present invention, and fig. 11c shows the three-phase current waveform, the dynamic response and the prediction error of the three-vector data driving prediction current control method based on the extended state observer according to the present invention; it can be seen that the total harmonic distortion of the phase a current of the three-vector data driving predictive current control method in the prior art is reduced from 3.85% to 1.91%; the current performance of the three-vector data driving prediction current control method is influenced by sampling disturbance, the total harmonic distortion is slightly higher than that of the three-vector data driving prediction current control method in the prior art, and specifically, the total harmonic distortion is reduced to 2.20% from 4.42%; the current ripple of the present invention's extended state observer-based three-vector data-driven predictive current control method is improved, the total harmonic distortion is reduced from 3.31% to 1.71%. In conclusion, the three-vector data drive prediction current control method based on the extended state observer can effectively inhibit sampling disturbance, does not have the defect that the three-vector data drive prediction current control method in the prior art is influenced by system nonlinear factors, has small current ripple and prediction error, and is superior to the three-vector data drive prediction current control method in the prior art. Further, the error gain feedback in the extended state observer based on the three-vector data drive prediction current control method of the extended state observer aims at the fastest dynamic response, and the response speed is 1.32ms. In conclusion, the three-vector data drive prediction current control method based on the extended state observer is superior to the three-vector model-free prediction current control method in the prior art.
In the actual experiment, the experimental parameters are consistent with the simulation parameters.
In conclusion, the improved data-driven energy storage inverter current prediction control method has better performance than the energy storage inverter data-driven prediction current control method in the prior art in both simulation and actual experiments.
In conclusion, the invention establishes a model according to the coordinate relation of the voltage vector through the improved data-driven energy storage inverter current prediction control method, and carries out real-time and rapid updating on the current gradient through the current gradient updating formula, thereby solving the problem that the updating of the current gradient is stagnated in the model-free prediction current control of the energy storage inverter, eliminating the current spike of the output current of the energy storage inverter and reducing the prediction error.
The method has the advantages that the measurement noise of the measured power grid current in the model-free prediction current control of the energy storage inverter is considered, the measurement noise of the measured power grid current is estimated and compensated, the problem that the model-free prediction current control of the energy storage inverter is easily affected by the measurement noise is solved under the condition of high dynamic response speed, and the problem of large current prediction error is further solved.
Three vectors are applied in each control period, so that the robustness of parameter change is improved, the voltage vector acted in each control period is increased under the condition of not increasing a lookup table, the influence of sampling disturbance and the ripple of output current are reduced, and the quality of the output current is improved.
Furthermore, the current prediction of the data-driven energy storage inverter current prediction control method is based on the current gradients stored in the lookup table, so that the real-time update of all the current gradients in the lookup table is realized without any system parameter, and the good current performance can be still maintained when the parameters are mismatched.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (8)
1. An improved current prediction control method for a data-driven energy storage inverter is characterized by comprising the following steps:
s1, establishing a mathematical model of an energy storage inverter in a static coordinate system, obtaining a plurality of basic voltage vectors according to the state of a three-phase switching tube of the energy storage inverter, using a current gradient as a prediction model of the energy storage inverter and synthesizing a plurality of virtual vectors;
s2, estimating and compensating the measurement noise of the energy storage inverter for measuring the power grid current;
s3, establishing a current gradient relation corresponding to the plurality of basic voltage vectors, and updating current gradients corresponding to residual vectors;
s4, primary current prediction is carried out, and a plurality of value function values corresponding to the plurality of basic voltage vectors and action time of the plurality of virtual vectors are obtained through value function equation analysis;
and S5, analyzing and obtaining current predicted values corresponding to the virtual vectors according to the updated current gradient and the action time of the virtual vectors, obtaining a virtual voltage vector with the minimum value function value through the cost function, and acting the virtual voltage vector as an optimal voltage vector to the next control period.
2. The improved data-driven energy storage inverter current prediction control method as claimed in claim 1 wherein the plurality of fundamental voltage vectors is eight fundamental voltage vectors, each being u 0 (0,0,0)、u 1 (1,0,0)、u 2 (1,1,0)、u 3 (0,1,0)、u 4 (0,1,1)、u 5 (0,0,1)、u 6 (1,0,1)、u 7 (1,1,1)；
The mathematical model of the energy storage inverter in the static coordinate system is as follows:
wherein u is x For the output voltage of the energy storage inverter, i g Is the output current of the energy storage inverter, e g The voltage of the power grid side of the energy storage inverter is L, a filter inductor is L, a filter resistor is R, and t is time;
the plurality of virtual vectors is six virtual vectors.
3. The improved data-driven energy storage inverter current prediction control method of claim 1, wherein in S2, an extended state observer capable of measuring current error feedback is established to estimate and compensate for noise of the measured grid current, the extended state observer being:
analyzing the extended state observer yields:
wherein i e For an estimated value of the output current of the energy storage inverter,. DELTA.i e For actually measuring the estimated value of the current gradient, δ 1 For feedback error gain of sampled current, delta 2 For sampling the feedback error gain, T, of the grid current s To control the period, i g For the output current of the energy-storage inverter i e – i g For current error, t is time.
4. The improved data-driven energy storage inverter current prediction control method of claim 1, wherein the current gradient update formula is expressed as:
wherein Δ i m Current gradient of basic voltage vector for first application, t m Is the action time of the first applied basic voltage vector, A is the proportionality coefficient of the first applied basic voltage vector, t n For the duration of the action of the second applied basic voltage vector, B is the proportionality factor of the second applied basic voltage vector,. DELTA.i n For the current gradient of the second applied basic voltage vector,. DELTA.i z Current gradient, Δ i, of zero voltage vector x To sample the current gradient of the current,. DELTA.i y Is the current gradient, Δ i, of the residual voltage vector 1 Is the current gradient, Δ i, of a first elementary voltage vector of said plurality of elementary voltage vectors 4 Is the current gradient of a fourth basic voltage vector of said plurality of basic voltage vectors, k-1 being the last control moment, T s To control the period, t 1 The action time of a first one of said elementary voltage vectors, k-2 being the last control moment,. DELTA.i 2,6 A current gradient, t, for a second and a sixth basic voltage vector of the plurality of basic voltage vectors z Is a first basic voltage vector sum of the multiple basic voltage vectorsDuration of action of the eighth basic voltage vector,. DELTA.i z A current gradient, Δ i, for a first and an eighth of said plurality of elementary voltage vectors x For sampling the measured current gradient, t 4 Is the action time, Δ i, of a fourth elementary voltage vector of said plurality of elementary voltage vectors 3,5 A current gradient, Δ i, for a third and a sixth elementary voltage vector of said plurality of elementary voltage vectors 1,4 A current gradient of a second basic voltage vector and a fourth basic voltage vector of the plurality of basic voltage vectors.
5. The improved data-driven energy storage inverter current prediction control method of claim 1, wherein in S4, the plurality of basic voltage vectors are added to the sampled current for current prediction, and the current prediction formula is:
wherein i iαβ Is the power grid current under the static coordinate system, k +1 is the next control moment, i αβ Is a sampling current in a stationary coordinate system,. DELTA.i iαβ The current gradient corresponding to the ith basic voltage vector is obtained, and k is the corresponding moment;
the cost function is:
wherein G is a value of a cost function, i refα For the reference current i of the energy-storage inverter in the alpha stationary coordinate system gα For the power grid current of the energy storage inverter under an alpha static coordinate system, k +1 is the next control moment, i refβ For the reference current i of the energy-storage inverter in the beta stationary coordinate system gβ And the grid current of the energy storage inverter under a beta static coordinate system is obtained.
6. The improved data-driven energy storage inverter current prediction control method of claim 1, wherein in S5, substituting the current prediction value into the cost function results in a corresponding cost function value;
the action time of the plurality of virtual vectors is as follows:
wherein, t m Acting time of basic voltage vector for first application, t n Acting time of basic voltage vector for second application, t z The action times of a first and an eighth basic voltage vector of said basic voltage vectors, G n Value of the fundamental voltage vector for the second application, G z Value of the zero-voltage vector, G m Value of the cost function of the fundamental voltage vector for the first application, T s Is a control cycle.
7. The improved data-driven energy storage inverter current prediction control method of claim 1, wherein when three fundamental voltage vectors are applied per control cycle, the predicted value of current at time k +1 is:
wherein i g Is the output current of the energy storage inverter, t m The action time of the first applied basic voltage vector, k +1 is the next control time, k is the current control time, T s To control the period,. DELTA.i m A current gradient, t, for an mth basic voltage vector of the plurality of basic voltage vectors n The action time of the basic voltage vector for the second application of the basic voltage vector,. DELTA.i n Is the nth basic voltage of the plurality of basic voltage vectorsCurrent gradient of the vector, k being the current control moment, t z For the action time of the first and eighth basic voltage vectors of the plurality of basic voltage vectors, Δ i z A current gradient being a first and an eighth base voltage vector of the plurality of base voltage vectors;
the predicted current value at the time k +2 is as follows:
wherein i g K +2 is the next control time, k +1 is the next control time, t is the output current of the energy storage inverter m The action time, T, of the basic voltage vector for the first application of the basic voltage vector s To control the period,. DELTA.i m A current gradient, t, for an mth basic voltage vector of the plurality of basic voltage vectors n The action time of the basic voltage vector for the second application of the basic voltage vector,. DELTA.i n Is the current gradient of the nth basic voltage vector of the plurality of basic voltage vectors, t z Is the action time, Δ i, of a first and an eighth of said plurality of elementary voltage vectors z A current gradient of a first basic voltage vector and an eighth basic voltage vector of the plurality of basic voltage vectors.
8. An improved data-driven energy storage inverter current predictive control system, comprising:
the parameter setting module is used for establishing a mathematical model of the energy storage inverter in a static coordinate system, obtaining a plurality of basic voltage vectors according to the states of three-phase switching tubes of the energy storage inverter, and taking a current gradient as a prediction model of the energy storage inverter and synthesizing a plurality of virtual vectors;
the noise estimation module is used for estimating and compensating the measurement noise of the grid current measured by the energy storage inverter;
the current gradient updating module is used for establishing a current gradient relation corresponding to the plurality of basic voltage vectors and updating current gradients corresponding to other residual vectors;
and the current prediction module is used for obtaining a plurality of value function values corresponding to the plurality of basic voltage vectors through the analysis of a value function equation, obtaining the action time of the plurality of virtual vectors, analyzing and obtaining current prediction values corresponding to the plurality of virtual vectors according to the updated current gradient and the action time of the plurality of virtual vectors, obtaining a virtual voltage vector with the minimum value function value through the value function, and acting the virtual voltage vector as an optimal voltage vector to the next control period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310109273.9A CN115987130A (en) | 2023-02-14 | 2023-02-14 | Improved data-driven energy storage inverter current prediction control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310109273.9A CN115987130A (en) | 2023-02-14 | 2023-02-14 | Improved data-driven energy storage inverter current prediction control method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115987130A true CN115987130A (en) | 2023-04-18 |
Family
ID=85974357
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310109273.9A Pending CN115987130A (en) | 2023-02-14 | 2023-02-14 | Improved data-driven energy storage inverter current prediction control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115987130A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116455288A (en) * | 2023-06-12 | 2023-07-18 | 国网上海能源互联网研究院有限公司 | No-parameter prediction control method, device, equipment and medium for electric locomotive |
| CN116545292A (en) * | 2023-06-30 | 2023-08-04 | 合肥安赛思半导体有限公司 | LC filtering inverter prediction control method and system |
| CN116707342A (en) * | 2023-08-08 | 2023-09-05 | 安徽大学 | Data-driven energy storage inverter voltage prediction control method |
| CN117595698A (en) * | 2024-01-19 | 2024-02-23 | 安徽大学 | Inverter operation parameter prediction method, device, equipment and storage medium |
-
2023
- 2023-02-14 CN CN202310109273.9A patent/CN115987130A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116455288A (en) * | 2023-06-12 | 2023-07-18 | 国网上海能源互联网研究院有限公司 | No-parameter prediction control method, device, equipment and medium for electric locomotive |
| CN116455288B (en) * | 2023-06-12 | 2023-08-29 | 国网上海能源互联网研究院有限公司 | No-parameter prediction control method, device, equipment and medium for electric locomotive |
| CN116545292A (en) * | 2023-06-30 | 2023-08-04 | 合肥安赛思半导体有限公司 | LC filtering inverter prediction control method and system |
| CN116707342A (en) * | 2023-08-08 | 2023-09-05 | 安徽大学 | Data-driven energy storage inverter voltage prediction control method |
| CN116707342B (en) * | 2023-08-08 | 2023-12-19 | 安徽大学 | Data-driven energy storage inverter voltage prediction control method |
| CN117595698A (en) * | 2024-01-19 | 2024-02-23 | 安徽大学 | Inverter operation parameter prediction method, device, equipment and storage medium |
| CN117595698B (en) * | 2024-01-19 | 2024-04-09 | 安徽大学 | Inverter operation parameter prediction method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115987130A (en) | Improved data-driven energy storage inverter current prediction control method and system | |
| Li et al. | Inductance surface learning for model predictive current control of switched reluctance motors | |
| CN108054972B (en) | Method for improving dynamic control performance of permanent magnet synchronous linear motor | |
| CN112865646A (en) | Dead-beat prediction control method for single current sensor of permanent magnet synchronous motor | |
| CN114172412A (en) | Non-parameter model prediction current control method for double three-phase permanent magnet motor | |
| CN112186791B (en) | Energy storage converter limited control set model prediction control method | |
| CN110649854A (en) | Surface-mounted permanent magnet synchronous motor continuum model prediction torque control method | |
| Zhang et al. | An improved model-free predictive current control of pwm rectifiers | |
| CN114531082B (en) | AESO-based dead-beat current prediction fuzzy control method for permanent magnet synchronous motor | |
| Wang et al. | An improved model free predictive current control for pmsm with current prediction error variations | |
| Pereira et al. | Model-free finite-set predictive current control with optimal cycle time for a switched reluctance motor | |
| CN111614279A (en) | Space vector modulation method and system based on model predictive control and inverter | |
| CN117254734A (en) | PMSM (permanent magnet synchronous motor) multimode switching model prediction control method, system and storage medium | |
| Tan et al. | Finite-control-set model predictive control of interleaved DC-DC boost converter Based on Kalman observer | |
| Yin et al. | A triple-voltage-vector model-free predictive current control scheme for voltage source inverters | |
| CN116015138A (en) | Motor control method and device, motor controller and permanent magnet synchronous motor | |
| CN114765445A (en) | Model-free prediction current control method and control system based on three voltage vectors | |
| CN109391165B (en) | Interference compensation method based on modular multilevel converter circulating current model | |
| Li et al. | A Robust Predictive Current Control Based on Incremental Model and Adaline N eural Network for PMSM Drives | |
| Zhang et al. | An improved model-free predictive current control method for PMSM drives based on extended control set and fast current difference updating | |
| KR20170137272A (en) | Current Control Methods for Single-Phase Voltage Source Inverters | |
| Biason et al. | Advances on analysis, modeling and accurate self-commissioning compensation of inverter dead-time distortion based on a physical model | |
| Al-kaf et al. | Model Predictive Current Control Using Single Layer Neural Network for PMSM Drives | |
| CN116846279A (en) | ESO-based open winding permanent magnet synchronous linear motor prediction current control method | |
| Sokola et al. | Combined impact of iron loss and main flux saturation on operation of vector controlled induction machines |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |