CN112909944B - Hybrid electric energy treatment device and method - Google Patents

Hybrid electric energy treatment device and method Download PDF

Info

Publication number
CN112909944B
CN112909944B CN202110089911.6A CN202110089911A CN112909944B CN 112909944 B CN112909944 B CN 112909944B CN 202110089911 A CN202110089911 A CN 202110089911A CN 112909944 B CN112909944 B CN 112909944B
Authority
CN
China
Prior art keywords
current
inverter
igbt
mosfet
harmonic
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.)
Active
Application number
CN202110089911.6A
Other languages
Chinese (zh)
Other versions
CN112909944A (en
Inventor
彭子舜
戴瑜兴
朱方
胡文
陈宇
李民英
章纯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Zhicheng Champion Group Co Ltd
Original Assignee
Guangdong Zhicheng Champion Group Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangdong Zhicheng Champion Group Co Ltd filed Critical Guangdong Zhicheng Champion Group Co Ltd
Priority to CN202110089911.6A priority Critical patent/CN112909944B/en
Publication of CN112909944A publication Critical patent/CN112909944A/en
Application granted granted Critical
Publication of CN112909944B publication Critical patent/CN112909944B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/01—Arrangements for reducing harmonics or ripples
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/40—Arrangements for reducing harmonics

Abstract

The invention discloses a hybrid electric energy management device, which comprises a harmonic control module and a hybrid active filter unit, wherein the hybrid active filter unit comprises an IGBT inverter and an MOSFET inverter which are connected in parallel, wherein the direct current input sides of the IGBT inverter and the MOSFET inverter are connected in parallel and share the same direct current capacitor; the IGBT inverter and the MOSFET inverter are connected in parallel at the alternating current output side, and the filter inductors at the alternating current output side are connected separately; the harmonic control module comprises an IGBT harmonic control unit and an MOSFET harmonic control unit, and is respectively used for controlling the IGBT inverter to remove low-order harmonic current in the power grid current and controlling the MOSFET inverter to remove residual harmonic current in the power grid current. The invention adopts a Si IGBT/SiC MOSFET hybrid active filter unit; compared with an active filter device with a single power device, the device has the advantages of the single power device, and has the characteristics of lower switching loss, higher short-circuit redundancy capability, lower cost and the like.

Description

Hybrid electric energy treatment device and method
Technical Field
The invention relates to the field of power electronic devices and control, in particular to a hybrid electric energy management device and method.
Background
With the vigorous promotion of renewable energy development in China, more and more novel energy generating sets are connected to a power grid, and the power grid is in a double-high form, namely the new energy is connected to a high proportion and power electronic equipment is connected to a high proportion. The new energy unit generally adopts a power electronic device as a grid-connected interface, so that a large amount of harmonic waves can be introduced into a power grid, and the power grid environment is polluted. At present, a power electronic device is used as one of the technologies which restrict the performance of an active filter device, so that the research on the comprehensive control technology and device for the electric energy quality with high performance, high power density, large capacity, low cost and high reliability requirements is of great importance for the development of power electronics and the improvement of the electric energy quality in China.
In fact, both Si IGBT and SiC MOSFET active filter devices are difficult to meet these stringent requirements effectively; new active filtering devices are required to achieve effective management of hybrid power in the grid.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a hybrid electric energy management device and a hybrid electric energy management method.
In order to achieve the purpose, the invention adopts the following technical scheme: a hybrid electric energy management device comprises a harmonic control module and a hybrid active filter unit, wherein the hybrid active filter unit comprises an IGBT inverter and an MOSFET inverter which are connected in parallel, wherein the direct current input sides of the IGBT inverter and the MOSFET inverter are connected in parallel and share the same direct current capacitor; the IGBT inverter and the MOSFET inverter are connected in parallel at the alternating current output side, and the filter inductors at the alternating current output side are connected separately;
the harmonic control module comprises an IGBT harmonic control unit and an MOSFET harmonic control unit, and is respectively used for controlling the IGBT inverter to remove low-order harmonic current in the power grid current and controlling the MOSFET inverter to remove residual harmonic current in the power grid current.
Furthermore, the voltage level of the IGBT inverter is the same as that of the MOSFET inverter, the switching frequency of the IGBT inverter is lower than that of the MOSFET inverter, and the rated current of the IGBT inverter is larger than that of the MOSFET inverter.
A hybrid electric energy treatment method comprises the following steps:
s01: low-order harmonic current of the power grid current is filtered by the IGBT harmonic control unit; the method specifically comprises the following steps:
s011: the power grid current is filtered by a low-pass filter to remove higher harmonics, and then a fundamental component is subtracted;
s012: then adding a voltage loop PI control output quantity introduced with a conventional direct current droop control strategy to obtain low-order harmonic current;
s013: the IGBT inverter is controlled to remove low-order harmonic current in the power grid current by adopting a self-adaptive model predictive control strategy aiming at the IGBT inverter, so that the primary treatment of the power grid current is realized;
s02: the residual harmonic current of the power grid current is filtered by the MOSFET harmonic control unit; the method specifically comprises the following steps:
s021: the power grid current is filtered by a low-pass filter to remove higher harmonics, and then the power grid current is subtracted;
s022: then adding a voltage loop PI control output quantity introduced with a conventional direct current droop control strategy to obtain residual harmonic current;
s023: the self-adaptive model predictive control strategy for the MOSFET inverter is adopted to control the MOSFET inverter to remove the residual harmonic current in the power grid current, so as to further control the power grid current; the IGBT inverter and the MOSFET inverter are connected in parallel, and the direct current input sides of the IGBT inverter and the MOSFET inverter are connected in parallel and share the same direct current capacitor; the IGBT inverter and the MOSFET inverter are connected in parallel at the alternating current output side, and the filter inductor at the alternating current output side is connected separately.
Further, step S013 specifically includes:
s0131: an improved wolf algorithm is adopted, a multi-objective optimization function applied to the IGBT harmonic control unit is designed, and an optimal weight coefficient of self-adaptive model predictive control in the IGBT harmonic control unit is obtained;
s0132: substituting the optimal weight coefficient into the cost function, and combining with the prediction model to obtain the optimal switching vector of the IGBT inverter;
s0133: and controlling the IGBT inverter by using the optimal switching vector to filter low-order harmonic current in the power grid current.
Furthermore, the improved gray wolf algorithm comprises a leading-layer wolf group, a second-layer wolf group, a subordinate wolf group and a lowest-layer wolf group, wherein the leading-layer wolf group, the second-layer wolf group and the subordinate wolf group adopt the gray wolf algorithm; half of the wolfs in the lowest-level wolf group adopt a wolf algorithm path updating scheme, and the other half of the wolfs adopt a cuckoo algorithm path updating scheme.
Further, the step S0131 specifically includes:
t01: initializing the improved wolf algorithm and entering an optimization iteration loop;
t02: updating the particles in the leading-layer wolf group, the second-layer wolf group, the subordinate wolf group and the lowest-layer wolf group to obtain the latest weight coefficient of the adaptive model predictive control in the IGBT harmonic control unit; outputting the latest weight coefficient values into a cost function: meanwhile, the current and the voltage of the power grid are obtained, and an adaptive value corresponding to the latest particle is calculated through a multi-objective optimization function;
t03: judging whether the iteration times are less than or equal to the set times, if so, comparing the particles with the minimum adaptive values in all the current groups with the historical optimal particles, updating the optimal particles, and returning to the step T01; and if the number of iterations is equal, selecting the weight coefficient with the minimum adaptive value for output, namely the optimal weight coefficient.
Further, step S023 specifically includes:
s0231: an improved wolf algorithm is adopted, a multi-objective optimization function applied to the MOSFET harmonic control unit is designed, and an optimal weight coefficient of self-adaptive model prediction control in the MOSFET harmonic control unit is obtained;
s0232: substituting the optimal weight coefficient into the cost function, and combining with the prediction model to obtain the optimal switching vector of the MOSFET inverter;
s0233: and controlling the MOSFET inverter by utilizing the optimal switching vector to filter residual harmonic current in the power grid current.
Furthermore, the improved gray wolf algorithm comprises a leading-layer wolf group, a second-layer wolf group, a subordinate wolf group and a lowest-layer wolf group, wherein the leading-layer wolf group, the second-layer wolf group and the subordinate wolf group adopt the gray wolf algorithm; half of the gray wolves in the lowest-level wolf group adopt a gray wolves algorithm path updating scheme, and the other half of the gray wolves adopt a cuckoo algorithm path updating scheme;
further, step S0231 specifically includes:
b01: initializing the improved wolf algorithm and entering an optimization iteration loop;
b02: updating the particles in the leading-layer wolf group, the second-layer wolf group, the subordinate wolf group and the lowest-layer wolf group to obtain the latest weight coefficient of the adaptive model predictive control in the MOSFET harmonic control unit; outputting the latest weight coefficient values into a cost function: meanwhile, the current and the voltage of the power grid are obtained, and an adaptive value corresponding to the latest particle is calculated through a multi-objective optimization function;
b03: judging whether the iteration times are less than or equal to the set times, if so, comparing the particles with the minimum adaptive values in all the current groups with the historical optimal particles, updating the optimal particles, and returning to the step T01; and if the number of iterations is equal, selecting the weight coefficient with the minimum adaptive value for output, namely the optimal weight coefficient.
Further, when the grid current is a three-phase grid current, and the IGBT inverter and the MOSFET inverter are two-level three-phase inverters, the step S0132 or step S0232 specifically includes: obtaining a relational expression of the voltage vector, the inductive current and the predicted current of the IGBT inverter or the MOSFET inverter:
Figure BDA0002912041320000051
wherein v (k) represents 7 switching vectors; l, L1Is a filter inductor; i.e. iICOM(k) Is an IGBT inverter current vector; i.e. iIp(k +1) is a predicted current vector of the IGBT inverter; i.e. iMCOM(k) Is a MOSFET inverter current vector; i.e. iMp(k +1) is a MOSFET inverter predicted current vector; the IGBT inverter command current iIr(k +1) obtaining a current vector i by stationary coordinate axis transformationIrα(k +1) and iIrβ(k + 1); the MOSFET inverter is controlled to command the current iMr(k +1) obtaining a current vector i by stationary coordinate axis transformationMrα(k +1) and iMrβ(k+1)；
Establishing a cost function of the adaptive model predictive control in the IGBT harmonic control unit and a cost function of the adaptive model predictive control in the MOSFET harmonic control unit:
Figure BDA0002912041320000052
in the formula iIpα(k +1) and iIpβ(k +1) is the harmonic current value of the IGBT inverter obtained through model prediction; i.e. iMpα(k +1) and iMpβ(k +1) is a harmonic current value of the MOSFET inverter obtained through model prediction; mu.s1、μ2、μ3Weight coefficient, mu, for adaptive model predictive control in IGBT harmonic control units4、μ5And mu6Weight coefficients for adaptive model predictive control in a MOSFET harmonic control unit; λ is the net side power factor; p is network side active power; q is the network side reactive power; and the switching vector corresponding to the minimum cost function is the optimal switching vector.
The invention has the beneficial effects that: the invention adopts a Si IGBT/SiC MOSFET hybrid active filter unit; compared with an active filter device with a single power device, the device has the advantages of the single power device, and has the characteristics of lower switching loss, higher short-circuit redundancy capability, lower cost and the like.
Drawings
FIG. 1 is a block diagram of the structure and control of a hybrid power management device according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of harmonic extraction applied to a hybrid power management device according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of predictive control applied to a hybrid power management device according to an embodiment of the present invention;
FIG. 4 is a flow chart of an optimization of the improved graying algorithm according to an embodiment of the present invention.
Detailed Description
The invention will be further described with reference to the accompanying drawings and the detailed description below:
the invention provides a hybrid electric energy management device which comprises a harmonic control module and a hybrid active filter unit, wherein the hybrid active filter unit comprises an IGBT inverter and an MOSFET inverter which are connected in parallel. Specifically, the IGBT inverter is a large-capacity Si IGBT inverter, and the MOSFET inverter is a small-capacity SiC MOSFET inverter. The high-capacity Si IGBT inverter adopts lower switching frequency to prevent the increase of the switching loss of the hybrid active filter unit, the low-capacity SiC MOSFET inverter adopts higher switching frequency to further improve the power density of the Si IGBT/SiC MOSFET hybrid active filter unit, and the redundancy capability of the hybrid electric energy management device can be further increased on the premise of preventing the cost of the Si IGBT/SiC MOSFET hybrid active filter unit from being too high. Aiming at a Si IGBT inverter, a low-order harmonic current is obtained, then an adaptive model predictive control strategy aiming at the IGBT inverter is adopted to realize the function of inhibiting the low-order harmonic current of an active filter device, and meanwhile, the Si IGBT is enabled to bear the low-order heavy current harmonic; for the SiC MOSFET inverter, the residual harmonic current is obtained, and then the suppression function of the residual harmonic current of the active filter unit is realized by adopting a self-adaptive model prediction control strategy for the MOSFET inverter; the combination of the two can also realize effective correction of the power factor. Therefore, the invention can enable the treatment device to effectively meet the harsh requirements of high performance, high power density, large capacity, low cost, high reliability and the like by combining a large-capacity Si IGBT inverter, a small-capacity SiC MOSFET inverter and a corresponding harmonic control strategy.
The direct-current input sides of the IGBT inverter and the MOSFET inverter are connected in parallel and share the same direct-current capacitor; the IGBT inverter and the MOSFET inverter are connected in parallel at the alternating current output side, and the filter inductors at the alternating current output side are connected separately; the voltage grade of the IGBT inverter is the same as that of the MOSFET inverter, the switching frequency of the IGBT inverter is far lower than that of the MOSFET inverter, and the rated current of the IGBT inverter is far larger than that of the MOSFET inverter. The harmonic control module comprises an IGBT harmonic control unit and an MOSFET harmonic control unit which are respectively used for controlling the IGBT inverter to remove low-order harmonic current in the power grid current and controlling the MOSFET inverter to remove residual harmonic current in the power grid current.
FIG. 1 shows the structure and control block diagram of a Si IGBT/SiC MOSFET hybrid power management device. In FIG. 1, va、vb、vcIs the grid voltage; i.e. isIs the current of the power grid; l isa、Lb、LcIs an alternating current filter inductor; u shapecdIs a direct current capacitance feedback value; u shapecfIs a direct current given value; i.e. iIIs the harmonic current of the SiIGBT inverter; i.e. iMIs the SiC MOSFET inverter harmonic current; i.e. iIrIs a harmonic reference current of the SiIGBT inverter; i.e. iMrIs a SiC MOSFET inverter harmonic reference current; i isMdc、IIdcThe direct current of the Si IGBT inverter and the direct current of the SiC MOSFET inverter are respectively. Fig. 1 mainly includes a hybrid active filter unit formed by connecting a large-capacity Si IGBT inverter (taking a two-level three-phase inverter as an example) and a small-capacity SiC MOSFET inverter in parallel, and a harmonic control module for controlling the IGBT inverter and the MOSFET inverter to remove low-order harmonic current and residual harmonic current in grid current. The high-capacity Si IGBT inverter adopts lower switching frequency to prevent the increase of the switching loss of the hybrid active filter device, the low-capacity SiC MOSFET inverter adopts higher switching frequency to further improve the power density of the Si IGBT/SiC MOSFET hybrid active filter device, and the low-capacity Si IGBT inverter and the low-capacity SiC MOSFET inverter can further increase the redundancy capability of the device on the premise of preventing the cost of the Si IGBT/SiC MOSFET hybrid active filter unit from being too high. Aiming at a Si IGBT inverter, a low-order harmonic current is obtained, then an adaptive model predictive control strategy aiming at the IGBT inverter is adopted to realize the function of inhibiting the low-order harmonic current of an active filter device, and meanwhile, the Si IGBT is enabled to bear the low-order heavy current harmonic; for the SiC MOSFET inverter, the residual harmonic current is obtained, and then the suppression function of the residual harmonic current of the active filter device is realized by adopting a self-adaptive model prediction control strategy for the MOSFET inverter; the combination of the two can also realize effective correction of the power factor.
Referring to fig. 1-4, the present invention provides a method for treating hybrid electric energy, which specifically comprises:
s01: low-order harmonic current of the power grid current is filtered by the IGBT harmonic control unit; firstly, filtering higher harmonics of the power grid current through a low-pass filter, and then subtracting a fundamental component; then adding a voltage loop PI control output quantity introduced with a conventional direct current droop control strategy to obtain low-order harmonic current; finally, a self-adaptive model predictive control strategy for the IGBT inverter is adopted to control the IGBT inverter to remove low-order harmonic current in the power grid current;
as shown in fig. 2, fig. 2 is a method for separating low harmonic current from residual harmonic current, which includes a voltage loop and a harmonic acquisition unit. In fig. 2, the PLL is a phase locked loop; the LPF is a low-pass filter; u shapecdIs a direct current capacitance feedback value; u shapecfIs a direct current given value; i isMdc、IIdc、DM、DIThe direct current side current of the Si IGBT inverter, the direct current side current of the SiC MOSFET inverter, a direct current droop control coefficient in a voltage ring of the Si IGBT inverter and a direct current droop control coefficient in a voltage ring of the SiC MOSFET inverter are respectively; i isa0、Ib0、Ic0Is a direct current component; i.e. iaf、ibf、icfIs the fundamental component; i.e. ialh、iblh、iclhLow order harmonic currents; i.e. iIra、iIrb、iIrcIs a harmonic reference current of the SiIGBT inverter; i.e. iMra、iMrb、iMrcIs a SiC MOSFET inverter harmonic reference current. With reference to fig. 2, step S01 specifically includes the following steps:
s011: the high order harmonic wave of the power grid current is filtered by a low pass filter; grid side current is(three phases are each i)sa、isbAnd isc) Obtaining each phase current i containing direct current component, fundamental component and low-order harmonic current through LPF linkal、ibl、icl：
Figure BDA0002912041320000081
In the formula Ia0、Ib0、Ic0A direct current component for each term; i.e. iaf、ibfAnd icfAs the fundamental component of each phase; i.e. ialh、iblhAnd iclhIs the low order harmonic current of each phase.
Current i on network sidesa、isbAnd iscPerforming Fourier transform to obtain a fundamental amplitude; then using a phase locked loop to obtain a fundamental current (i) in the grid currentaf、ibfAnd icf). Will ial、ibl、iclAnd iaf、ibf、icfAre subtracted.
S012: DC capacitor voltage feedback value UcdSag link (I)Idc×DMOr IMdc×DM) With given value of DC capacitor voltage UcfObtaining output control quantity through voltage loop PI control, and adding the output quantity into the instruction current; obtaining low harmonic current i of network side currentalh、iblh、iclh。
S013: by adopting a self-adaptive model predictive control strategy aiming at the IGBT inverter, the IGBT inverter is controlled to remove low-order harmonic current in the power grid current, and the command current i of the Si IGBT inverter can be obtainedIr(three phases are each i)Ira、iIrbAnd iIrc) And the primary treatment of the power grid current is realized. Step S013 specifically includes:
s0131: an improved wolf algorithm is adopted, a multi-objective optimization function applied to the IGBT harmonic control unit is designed, and an optimal weight coefficient of self-adaptive model predictive control in the IGBT harmonic control unit is obtained; the improved gray wolf algorithm comprises a first-collar hierarchical wolf group, a second-level wolf group, a subordinate wolf group and a lowest-level wolf group, wherein the first-collar hierarchical wolf group, the second-level wolf group and the subordinate wolf group adopt the gray wolf algorithm; half of the gray wolves in the lowest-level wolf group adopt the original gray wolves algorithm updating scheme, and the other half of the gray wolves adopt the cuckoo algorithm path updating scheme. The obtaining of the optimal weight coefficient for adaptive model predictive control in the IGBT harmonic control unit specifically includes:
t01: initializing an improved wolf algorithm and entering an optimization iteration loop;
t02: updating the particles in the leading-layer wolf group, the second-layer wolf group, the subordinate wolf group and the lowest-layer wolf group to obtain the latest weight coefficient of the adaptive model predictive control in the IGBT harmonic control unit; outputting the latest weight coefficient values into a cost function: meanwhile, the current and the voltage of the power grid are obtained, and an adaptive value corresponding to the latest particle is calculated through a multi-objective optimization function;
t03: judging whether the iteration times are less than or equal to the set times, if so, comparing the particles with the minimum adaptive values in all the current groups with the historical optimal particles, updating the optimal particles, and returning to the step T01; and if the number of iterations is equal, selecting the weight coefficient with the minimum adaptive value for output, namely the optimal IGBT weight coefficient.
S0132: substituting the optimal weight coefficient into the cost function, and combining with the prediction model to obtain the optimal switching vector of the IGBT inverter;
s0133: and the IGBT inverter is controlled by utilizing the optimal switching vector to remove low-order harmonic current in the power grid current.
S02: the residual harmonic of the power grid current is filtered by the MOSFET harmonic control unit; firstly, filtering higher harmonics of the power grid current through a low-pass filter, and subtracting the higher harmonics from the power grid current; then adding a voltage loop PI control output quantity introduced with a conventional direct current droop control strategy to obtain residual harmonic current; and finally, realizing that the MOSFET inverter absorbs low-power residual harmonic current by adopting self-adaptive model predictive control. With reference to fig. 2, the method specifically includes:
s021: the high order harmonic wave of the power grid current is filtered by a low pass filter; a synchronization step S011; the grid current is then subtracted.
S022: then adding a voltage loop PI control output quantity introduced with a conventional direct current droop control strategy; a synchronization step S012; to obtain residual harmonic currents.
S023: removing residual harmonic current in power grid current by using MOSFET inverter, namely command current i of MOSFET inverterMr(three phases are each i)Mra、iMrbAnd iMrc) I.e. the governed grid current. Step S023 specifically includes:
s0231: an improved wolf algorithm is adopted, a multi-objective optimization function applied to the MOSFET harmonic control unit is designed, and an optimal weight coefficient of self-adaptive model prediction control in the MOSFET harmonic control unit is obtained; the improved gray wolf algorithm comprises a first-collar hierarchical wolf group, a second-level wolf group, a subordinate wolf group and a lowest-level wolf group, wherein the first-collar hierarchical wolf group, the second-level wolf group and the subordinate wolf group adopt the gray wolf algorithm; half of the wolfs in the lowest-level wolf group adopt the original grey wolf algorithm updating scheme, and the other half of the wolfs adopt the cuckoo algorithm path updating scheme:
b01: initializing an improved wolf algorithm and entering an optimization iteration loop;
b02: updating the particles in the leading-layer wolf group, the second-layer wolf group, the subordinate wolf group and the lowest-layer wolf group to obtain the latest weight coefficient of the adaptive model predictive control in the MOSFET harmonic control unit; outputting the latest weight coefficient values into a cost function: meanwhile, the current and the voltage of the power grid are obtained, and an adaptive value corresponding to the latest particle is calculated through a multi-objective optimization function;
b03: judging whether the iteration times are less than or equal to the set times, if so, comparing the particles with the minimum adaptive values in all the current groups with the historical optimal particles, updating the optimal particles, and returning to the step T01; and if the number of iterations is equal, selecting the weight coefficient with the minimum adaptive value for output, namely the optimal weight coefficient.
S0232: and obtaining the optimal switching vector of the MOSFET inverter by adopting the optimal weight coefficient in combination with the filter inductance model and the corresponding cost function.
S0233: and substituting the optimal weight coefficient into the cost function, and combining with the prediction model to obtain the optimal switching vector of the MOSFET inverter.
Because the ideas of the adaptive model predictive control strategy in the IGBT harmonic control unit and the adaptive model predictive control strategy in the MOSFET harmonic control unit in S013 and S023 are basically the same, only some parameters are different, and the two are introduced together as follows:
fig. 4 is a flow chart of the optimization of the improved grayling algorithm. The idea of combining the improved Husky algorithm and the adaptive model predictive control strategy is as follows: a multi-objective optimization function reflecting two performance indexes of the power grid side current harmonic and the power factor of the Si/SiC hybrid active power filter device is established, and then a weight parameter in an adaptive model prediction control strategy is optimized through an improved Husky algorithm, so that the performance and the reliability of the device are improved.
The improved grey wolf algorithm updates the step length by adopting a nesting path scheme in the cuckoo algorithm for half of the grey wolfs in the subordinate level wolf group in the grey wolf group, and then performs re-value selection by randomly selecting part of the grey wolfs in the lowest level wolf group to realize effective improvement of the convergence precision and the convergence speed of the algorithm. The improved gray wolf algorithm comprises a first-collar rank, a second rank, a subordinate rank and a lowest rank four rank wolf groups, the optimal area is searched by mainly dispersing alpha (first-collar rank wolf group), beta (subordinate rank wolf group) and delta (lowest rank wolf group), and the algorithm has strong global searching capability due to strong random behavior in the optimizing process, and can effectively prevent the optimization from falling into the local optimal area. The position update equations of the first two hierarchical wolf clusters (the leading hierarchical wolf cluster and the second hierarchical wolf cluster) are:
Figure BDA0002912041320000121
wherein
Figure BDA0002912041320000122
In the formulae (1) and (2), rk(k-1, 2,3,4) represents [0,1 ]]Random numbers within the interval; n and nmaxRespectively representing the current iteration times and the maximum iteration times; s (n), sα(n) and sβ(n) each represents an individual particle in the improved grayling algorithm, i.e., represents a weight coefficient in the adaptive model predictive control. Because most of the better particles are in the top-level and second-level wolf clusters, the update methods used by these two clusters do not change.
The update equation of the subordinate level wolf group is as follows:
Figure BDA0002912041320000123
wherein
Figure BDA0002912041320000131
In formulae (3) and (4), the indices i and j represent the number of particles; k is a radical of1Represents the number of particles; r is5And r6Represents [0,1 ]]Random numbers within the interval; n is1And n2All represent the current iteration number; sδ(n) and sbest(n) represents the individual particles in the improved Husky algorithm and the optimal particles in the whole population respectively, namely represents the weight coefficient in the adaptive model prediction control; u and v are random numbers obeying a normal distribution; a is0B and b1Are all constants. Half of the wolfs in the subordinate stratum wolf group are updated by the original scheme, the other half of the wolf group are updated by the path optimizing scheme in the cuckoo algorithm, and the optimizing capability of the stratum particle can be improved by the combination of the two updating schemes.
The update equation of the lowest-level wolf group is:
Figure BDA0002912041320000132
in the formula, r7Represents [0,1 ]]Random numbers within the interval; p is a radical ofaRepresenting a probability value; s1(n) and s2(n) represents two randomly selected particle positions in the low-level wolf group. The positions of the particles are changed through a certain probability, so that the optimizing capability of the hierarchical particles can be effectively improved.
The establishment of the multi-target adaptive value function can effectively evaluate the optimization performance of the improved gray wolf algorithm, and the adaptive value function has the equation:
Figure BDA0002912041320000141
in the formula, taThe time of the source filter device operation is required for obtaining the adaptive value function; alpha is alphawAnd betawIs the weight value in the adaptive value function; i issah、IsbhAnd IschThe effective value of the total harmonic current at each phase network side is obtained; i issa1、Isb1And Isc1For each phase net side fundamental current effective value (and i)af、ibfAnd icfThe same). The smaller the adaptation value, the better the performance of the representative particle.
Fig. 3 is a schematic diagram of a predictive control strategy applied to a Si IGBT/SiC MOSFET hybrid active filter circuit, where the predictive control strategy includes two parts, namely an inverter model and a cost function. In fig. 3, v (k) is an inverter voltage vector; t issIs the sampling time; abs is a modulo operation, μ1、μ2、μ3、μ4、μ5And mu6Is a weight coefficient, iIp(k +1) is a predicted current vector of the SiIGBT inverter; i.e. iMp(k +1) is a predicted current vector of the SiCSMOSFET inverter; i.e. iIr(k +1) is Si IGBT inverter instruction current; i.e. iMr(k +1) is the SiCSMOSFET inverter command current. The inverter model is based on an inverter voltage vector established by the filter inductor, a relational expression of an inductor current and a predicted current, and predicts a current value of the next switching period according to the current value of the current switching period and a voltage vector corresponding to the switching vector; the cost function comprises a plurality of variables such as power grid power factors, inverter predicted currents, reference currents and the like and a plurality of weight coefficients, the influence of the plurality of variables such as errors between predicted values and reference values and power grid power factor conditions on the system performance under the corresponding weight coefficients is quantitatively evaluated, and the magnitude of the output value of the cost function in the current switching period is judged through a comparison unit; the smaller the output value of the cost function is, the closer the predicted current is to the reference current, the closer the power factor at the power grid side is to 1, and meanwhile, the switching vector corresponding to the current obtained through current prediction is output. The relevant content of the proposed predictive control is as follows:
definition of Sa、SbAnd ScIs each bridge of a Si IGBT three-phase full-bridge inverterThe gate control signal of the arm, the port outputs high level when in the switch state 1, and the port outputs zero level when in the switch state 0. According to different switch states, 7 switch vectors are generated, which are respectively: 000(111), 001, 010, 100, 110, 010, and 011; 7 switching vectors are again corresponding to V0～V6There are 7 voltage vectors in total.
Firstly, respectively establishing a SiIGBT inverter and a SiMOSFET inverter based on a filter inductance model to obtain a relational expression of an inverter voltage vector, an inductance current and a predicted current:
Figure BDA0002912041320000151
wherein v (k) represents 7 voltage vectors; l, L1Is a filter inductor; i.e. iICOM(k) Is a current vector of the SiIGBT inverter; i.e. iIp(k +1) is a predicted current vector of the SiIGBT inverter; i.e. iMCOM(k) Is a SiCSMOSFET inverter current vector; i.e. iMpAnd (k +1) is a predicted current vector of the SiCSMOSFET inverter. The Si IGBT inverter command current iIr(k +1) obtaining a current vector i by stationary coordinate axis transformationIrα(k +1) and iIrβ(k + 1); the SiCSMOSFET inverter command current iMr(k +1) obtaining a current vector i by stationary coordinate axis transformationMrα(k +1) and iMrβ(k + 1). Then, in order to accurately control the harmonic current and improve the network side power factor, a cost function expression which needs to be established is as follows:
Figure BDA0002912041320000152
in the formula iIpα(k +1) and iIpβ(k +1) is a harmonic current value of the Si IGBT inverter obtained through model prediction; i.e. iMpα(k +1) and iMpβ(k +1) is a harmonic current value of the SiC MOSFET inverter obtained through model prediction; mu.s1、μ2、μ3、μ4、μ5And mu6Is a weight coefficient; λ is the net side power factor; p is network side active power; q is net side absentWork power.
Updating the weight coefficient mu in the cost function in real time through the improved gray wolf algorithm1、μ2、μ3、μ4、μ5And mu6Function g can be performed in real timeIAnd gMAnd (4) optimizing. Judging the magnitude of the output value of the cost function in the current switching period through a comparison unit; the smaller the cost function of the current switching vector is, the closer the current predicted by the current switching vector is to the reference current, and the closer the grid-side power factor is to 1, so the inverter should select the switching vector.
According to the Si IGBT/SiC MOSFET hybrid electric energy management device provided by the invention, the harsh requirements of high performance, high power density, high capacity, low cost, high reliability and the like can be effectively met by the source filter device by combining the large-capacity Si IGBT inverter, the small-capacity SiC MOSFET inverter and the corresponding harmonic control strategy.
Various other modifications and changes may be made by those skilled in the art based on the above-described technical solutions and concepts, and all such modifications and changes should fall within the scope of the claims of the present invention.

Claims (5)

1. A hybrid electric energy treatment method is characterized by comprising the following steps:
s01: low-order harmonic current of the power grid current is filtered by the IGBT harmonic control unit; the method specifically comprises the following steps:
s011: the power grid current is filtered by a low-pass filter to remove higher harmonics, and then a fundamental component is subtracted;
s012: then adding a voltage loop PI control output quantity introduced with a conventional direct current droop control strategy to obtain low-order harmonic current;
s013: the IGBT inverter is controlled to remove low-order harmonic current in the power grid current by adopting a self-adaptive model predictive control strategy aiming at the IGBT inverter, so that the primary treatment of the power grid current is realized;
s02: the residual harmonic current of the power grid current is filtered by the MOSFET harmonic control unit; the method specifically comprises the following steps:
s021: the power grid current is filtered by a low-pass filter to remove higher harmonics, and then the power grid current is subtracted;
s022: then adding a voltage loop PI control output quantity introduced with a conventional direct current droop control strategy to obtain residual harmonic current;
s023: the self-adaptive model predictive control strategy for the MOSFET inverter is adopted to control the MOSFET inverter to remove the residual harmonic current in the power grid current, so as to further control the power grid current; the IGBT inverter and the MOSFET inverter are connected in parallel, and the direct current input sides of the IGBT inverter and the MOSFET inverter are connected in parallel and share the same direct current capacitor; the IGBT inverter and the MOSFET inverter are connected in parallel at the alternating current output side, and the filter inductors at the alternating current output side are connected separately;
the step S013 specifically includes:
s0131: an improved wolf algorithm is adopted, a multi-objective optimization function applied to the IGBT harmonic control unit is designed, and an optimal weight coefficient of self-adaptive model predictive control in the IGBT harmonic control unit is obtained;
s0132: substituting the optimal weight coefficient into the cost function, and combining with the prediction model to obtain the optimal switching vector of the IGBT inverter;
s0133: the optimal switching vector is utilized to control the IGBT inverter to filter low-order harmonic current in the power grid current;
the step S023 specifically includes:
s0231: an improved wolf algorithm is adopted, a multi-objective optimization function applied to the MOSFET harmonic control unit is designed, and an optimal weight coefficient of self-adaptive model prediction control in the MOSFET harmonic control unit is obtained;
s0232: substituting the optimal weight coefficient into the cost function, and combining with the prediction model to obtain the optimal switching vector of the MOSFET inverter;
s0233: and controlling the MOSFET inverter by utilizing the optimal switching vector to filter residual harmonic current in the power grid current.
2. The hybrid electric energy management method according to claim 1, wherein the improved grayling algorithm comprises a top-ranked wolf cluster, a second-ranked wolf cluster, a subordinate wolf cluster and a lowest-ranked wolf cluster, wherein the top-ranked wolf cluster, the second-ranked wolf cluster and the subordinate wolf cluster adopt a grayling algorithm; half of the wolfs in the lowest-level wolf group adopt a wolf algorithm path updating scheme, and the other half of the wolfs adopt a cuckoo algorithm path updating scheme.
3. The hybrid electric energy management method according to claim 2, wherein the step S0131 specifically includes:
t01: initializing the improved wolf algorithm and entering an optimization iteration loop;
t02: updating the particles in the leading-layer wolf group, the second-layer wolf group, the subordinate wolf group and the lowest-layer wolf group to obtain the latest weight coefficient of the adaptive model predictive control in the IGBT harmonic control unit; outputting the latest weight coefficient values into a cost function: meanwhile, the current and the voltage of the power grid are obtained, and an adaptive value corresponding to the latest particle is calculated through a multi-objective optimization function;
t03: judging whether the iteration times are less than or equal to the set times, if so, comparing the particles with the minimum adaptive values in all the current groups with the historical optimal particles, updating the optimal particles, and returning to the step T01; and if the number of iterations is equal, selecting the weight coefficient with the minimum adaptive value for output, namely the optimal weight coefficient.
4. The hybrid electric energy management method according to claim 2, wherein the step S0231 specifically includes:
b01: initializing the improved wolf algorithm and entering an optimization iteration loop;
b02: updating the particles in the leading-layer wolf group, the second-layer wolf group, the subordinate wolf group and the lowest-layer wolf group to obtain the latest weight coefficient of the adaptive model predictive control in the MOSFET harmonic control unit; outputting the latest weight coefficient values into a cost function: meanwhile, the current and the voltage of the power grid are obtained, and an adaptive value corresponding to the latest particle is calculated through a multi-objective optimization function;
b03: judging whether the iteration times are less than or equal to the set times, if so, comparing the particles with the minimum adaptive values in all the current groups with the historical optimal particles, updating the optimal particles, and returning to the step T01; and if the number of iterations is equal, selecting the weight coefficient with the minimum adaptive value for output, namely the optimal weight coefficient.
5. The hybrid electric energy management method according to claim 1, wherein when the grid current is a three-phase grid current, and the IGBT inverter and the MOSFET inverter are two-level three-phase inverters, the step S0132 or the step S0232 specifically includes: obtaining a relational expression of the voltage vector, the inductive current and the predicted current of the IGBT inverter or the MOSFET inverter:
Figure 268963DEST_PATH_IMAGE001
in the formula (I), the compound is shown in the specification,v(k) Represents 7 switching vectors;L、L 1 is a filter inductor;
Figure 406684DEST_PATH_IMAGE002
is an IGBT inverter current vector;
Figure 400047DEST_PATH_IMAGE003
predicting a current vector for the IGBT inverter;
Figure 154377DEST_PATH_IMAGE004
is a MOSFET inverter current vector;
Figure 891389DEST_PATH_IMAGE005
predicting a current vector for the MOSFET inverter; commanding current to IGBT inverteri Ir (k+1) Obtaining current vectors by stationary coordinate axis transformation
Figure 8249DEST_PATH_IMAGE006
And
Figure 438094DEST_PATH_IMAGE007
(ii) a Commanding current to MOSFET inverteri Mr (k+1) Obtaining current vectors by stationary coordinate axis transformation
Figure 679719DEST_PATH_IMAGE008
And
Figure 954843DEST_PATH_IMAGE009
；
establishing a cost function of the adaptive model predictive control in the IGBT harmonic control unit and a cost function of the adaptive model predictive control in the MOSFET harmonic control unit:
Figure 67155DEST_PATH_IMAGE010
in the formula (I), the compound is shown in the specification,
Figure 402321DEST_PATH_IMAGE011
and
Figure 131243DEST_PATH_IMAGE012
the harmonic current value of the IGBT inverter is obtained through model prediction;
Figure 210057DEST_PATH_IMAGE013
and
Figure 301510DEST_PATH_IMAGE014
the harmonic current value of the MOSFET inverter is obtained through model prediction;μ 1、μ 2、μ 3for the weighting coefficients of the adaptive model predictive control in the IGBT harmonic control unit,μ 4、μ 5andμ 6weighting coefficients for adaptive model predictive control in a MOSFET harmonic control unit；
Figure 73157DEST_PATH_IMAGE015
Is the grid-side power factor;Pactive power is the network side;Qthe network side reactive power is obtained; and the switching vector corresponding to the minimum cost function is the optimal switching vector.
CN202110089911.6A 2021-01-22 2021-01-22 Hybrid electric energy treatment device and method Active CN112909944B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110089911.6A CN112909944B (en) 2021-01-22 2021-01-22 Hybrid electric energy treatment device and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110089911.6A CN112909944B (en) 2021-01-22 2021-01-22 Hybrid electric energy treatment device and method

Publications (2)

Publication Number Publication Date
CN112909944A CN112909944A (en) 2021-06-04
CN112909944B true CN112909944B (en) 2021-12-07

Family

ID=76118520

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110089911.6A Active CN112909944B (en) 2021-01-22 2021-01-22 Hybrid electric energy treatment device and method

Country Status (1)

Country Link
CN (1) CN112909944B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114744670A (en) * 2022-03-17 2022-07-12 深圳市京泉华科技股份有限公司 Multi-type power device topology parallel power supply and control method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN201975789U (en) * 2011-01-18 2011-09-14 江苏斯菲尔电气股份有限公司 Harmonic current canceling module and module group
CN204720990U (en) * 2015-05-28 2015-10-21 东莞市绿能环保节能科技有限公司 Reactive power compensation active filter
CN107480913A (en) * 2017-09-06 2017-12-15 东北大学 A kind of distributed power source addressing constant volume system and method based on improvement grey wolf algorithm
CN109635492A (en) * 2018-12-27 2019-04-16 湖南科技大学 Based on the adaptive BBMC main circuit parameter preferred method of electric current quota
CN211183424U (en) * 2019-12-04 2020-08-04 广西电网有限责任公司电力科学研究院 Combined wide band domain harmonic treatment device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI347058B (en) * 2007-03-05 2011-08-11 Ablerex Electronics Co Ltd Active power filter apparatus

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN201975789U (en) * 2011-01-18 2011-09-14 江苏斯菲尔电气股份有限公司 Harmonic current canceling module and module group
CN204720990U (en) * 2015-05-28 2015-10-21 东莞市绿能环保节能科技有限公司 Reactive power compensation active filter
CN107480913A (en) * 2017-09-06 2017-12-15 东北大学 A kind of distributed power source addressing constant volume system and method based on improvement grey wolf algorithm
CN109635492A (en) * 2018-12-27 2019-04-16 湖南科技大学 Based on the adaptive BBMC main circuit parameter preferred method of electric current quota
CN211183424U (en) * 2019-12-04 2020-08-04 广西电网有限责任公司电力科学研究院 Combined wide band domain harmonic treatment device

Also Published As

Publication number Publication date
CN112909944A (en) 2021-06-04

Similar Documents

Publication Publication Date Title
CN107565576B (en) Reactive voltage optimization method for active power distribution network coordinated by multiple active management means
Carrizosa et al. Multi-terminal dc grid overall control with modular multilevel converters
CN112909944B (en) Hybrid electric energy treatment device and method
CN112383237B (en) Model prediction control method of grid-connected inverter
CN110867856A (en) LCL filter parameter optimization method under condition of considering VSC system harmonic stability
Rameshkumar et al. Real time implementation and analysis of enhanced artificial bee colony algorithm optimized PI control algorithm for single phase shunt active power filter
Zobaa et al. Comparative analysis of double-tuned harmonic passive filter design methodologies using slime mould optimization algorithm
Monfared et al. Optimal switching-sequence-based model predictive control for a hybrid multilevel STATCOM
CN108110760B (en) Micro-grid system harmonic coordination control method considering power grid harmonic waves and load harmonic waves
Arunsankar et al. Optimal controller for mitigation of harmonics in hybrid shunt active power filter connected distribution system: An EGOANN technique
CN113541144A (en) Harmonic virtual resistance technology-based active power grid harmonic resonance suppression method
Prasad et al. Mitigation of power quality problems using custom power devices: A review
Lin et al. Coordinated‐control strategy of scalable superconducting magnetic energy storage under an unbalanced voltage condition
Rao et al. Comparative analysis of SRF based Shunt Active Filter using Grey Wolf and Eagle Perching Optimization
CN113992043A (en) Novel finite control set model prediction control method applied to grid-connected inverter
Han et al. Simplified finite set model predictive control strategy of grid-connected cascade H-bridge converter
Reddy et al. Improving gain of real time PI controller using particle swarm optimization in active power filter
Das et al. Optimisation‐based improved switching technique for static VAr compensator considering minimum source harmonic injection
CN113485109B (en) Converter dynamic cascade control method and system based on variable priority
Zhou et al. Resonance Suppression Strategy of the Multi-inverter Grid-connected System Based on Current Given Correction
CN111835009B (en) Alternating current filter optimization method and device
CHENNAI Efficiency of Series APF based on Nine-level NPC Inverter Compared to Seven-level Configuration.
Luo et al. Optimized Inductive Filter Device Design for a Novel Transformer Based on Improved Immune Genetic Algorithm
Lin et al. Operation and control analysis of modular multilevel converter for VSC-HVDC application
CN117748893A (en) Circulation suppression method for multi-inverter parallel system

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
GR01 Patent grant
GR01 Patent grant