CN114063443A - Grid-connected inverter control method for improving BP (Back propagation) setting fractional order PID (proportion integration differentiation) - Google Patents
Grid-connected inverter control method for improving BP (Back propagation) setting fractional order PID (proportion integration differentiation) Download PDFInfo
- Publication number
- CN114063443A CN114063443A CN202111454118.8A CN202111454118A CN114063443A CN 114063443 A CN114063443 A CN 114063443A CN 202111454118 A CN202111454118 A CN 202111454118A CN 114063443 A CN114063443 A CN 114063443A
- Authority
- CN
- China
- Prior art keywords
- current
- grid
- fractional order
- coefficient
- neural network
- 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
- 238000000034 method Methods 0.000 title claims abstract description 19
- 230000004069 differentiation Effects 0.000 title abstract description 8
- 230000010354 integration Effects 0.000 title abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims abstract description 32
- 230000001939 inductive effect Effects 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 17
- 239000003990 capacitor Substances 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 6
- 230000000946 synaptic effect Effects 0.000 claims description 6
- 238000013016 damping Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 14
- XOFYZVNMUHMLCC-ZPOLXVRWSA-N prednisone Chemical compound O=C1C=C[C@]2(C)[C@H]3C(=O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 XOFYZVNMUHMLCC-ZPOLXVRWSA-N 0.000 description 14
- 238000004088 simulation Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000001808 coupling effect Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B11/00—Automatic controllers
- G05B11/01—Automatic controllers electric
- G05B11/36—Automatic controllers electric with provision for obtaining particular characteristics, e.g. proportional, integral, differential
- G05B11/42—Automatic controllers electric with provision for obtaining particular characteristics, e.g. proportional, integral, differential for obtaining a characteristic which is both proportional and time-dependent, e.g. P. I., P. I. D.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Control Of Ac Motors In General (AREA)
Abstract
The invention relates to a grid-connected inverter control method for improving BP (back propagation) setting fractional order PID (proportion integration differentiation). A fractional order PID controller is adopted for inductive current outer ring control, a proportional compensator is adopted for capacitive current inner ring control of a three-phase LCL (lower control level) grid-connected inverter, and a proportional coefficient K in the fractional order PID controller is controlled by adopting a BP neural networkpIntegral coefficient KiDifferential coefficient KdAnd carrying out parameter setting on the integral order lambda and the differential order mu. The multivariable fractional order PID control is acted on the outer ring of the inductive current, so that the system has better robustness, and the THD value of the grid-connected current is greatly reduced; and a proportional controller is used as a capacitance current inner ring to realize double-closing control. Aiming at the problem of parameter setting caused by the introduction of a fractional order PID algorithm, the improved BP neural network algorithm is adopted to identify and optimize parameters, so that the stable control of grid-connected inversion is realized, and the robustness of the system is improved.
Description
Technical Field
The invention relates to an inverter control technology, in particular to a grid-connected inverter control method for improving BP setting fractional order PID.
Background
With the development of renewable energy grid-connected power generation technology, the control technology of a high-power grid-connected inverter becomes a hotspot of research. The energy research and development field is mainly embodied in three expression forms of light, heat and chemical utilization. Of these three forms, the application of photovoltaics to cell technology is the most critical form. Solar energy resources in most regions in China are abundant, and the average sunshine amount can reach more than 2000 hours in a year-round unit. In the aspect of photovoltaic power generation, China has sufficient resource reserves. On the other hand, photovoltaic power generation is a sustainable and renewable energy source, and can solve the power utilization problem of areas with difficult power transmission, distribution, transmission and utilization. Plays an important role in the development of poor areas in China.
Conventional control of three-phase inverters has been sophisticated, including PID control, state feedback control, and PR control. The PID control is mostly applied to a multi-loop control link, the effect of single closed-loop control is limited, and in a double closed-loop control link, the difference value of a reference voltage and an actual voltage is tracked through a PID controller.
The three-phase inverter mainly outputs sinusoidal signals, and the total harmonic distortion factor of the three-phase inverter is particularly important in the evaluation of the three-phase inverter. The three-phase inverter control system has complicated time-varying property, coupling property and nonlinearity, and is difficult to control.
Disclosure of Invention
Aiming at the problem of high-power grid-connected total harmonic distortion, the grid-connected inverter control method for improving BP setting fractional order PID is provided, so that the system has good dynamic and steady response and robustness, and the reference quantity is quickly tracked.
The technical scheme of the invention is as follows: grid-connected inverter control method for improving BP (back propagation) setting fractional order PID (proportion integration differentiation), wherein a fractional order PID controller is adopted for inductive current outer ring control, and a proportional compensator pair is adopted for a capacitive current inner ringThe three-phase LCL grid-connected inverter is controlled, and the current i on the capacitor is measured through the current transformercThe proportional controller is selected to eliminate harmonic component of the current high-frequency switch, the resonance peak is inhibited to increase the damping of the system, and the current i on the grid-connected inductor2With a given reference current i2 *Comparing, and obtaining the capacitance current reference signal i of the inner ring by the difference value through a fractional order PID controllerc *Reference signal i of capacitance currentc *And the actual current i of the inner loop of the capacitor currentcComparing the voltage signals to obtain a voltage signal U of the inverter after compensation is carried out by a proportional controllerinv(ii) a Proportional coefficient K in fractional order PID controller by BP neural networkpIntegral coefficient KiDifferential coefficient KdAnd carrying out parameter setting on the integral order lambda and the differential order mu.
Further, the input of the BP neural network is an inductive current signal i given to a grid-connected inductor2 *And error E ═ i2-i2 *E is a given inductor current signal i2 *And the current signal i obtained by actual measurement2A difference of (d); the hidden layer is selected from five neurons, and the output layer is Kp、Ki、KdAnd an integral order λ, a derivative order μ;
network input layer input is oiThe input of the hidden layer is net(2) mThe output is o(2) mThe input of the output layer is net(3) n;
The output is o(3) n(ii) a The neural network continuously adjusts the synaptic weight omega of the input layerimAnd hidden layer synaptic weight omegamnReducing the error E;
the input and output of the network hidden layer are respectively
Wherein k is the number of iterations;
the activation function of the hidden layer neuron selects a positive and negative symmetric Sigmoid function,
the input and output of the network output layer are respectively:
output of network layer o(3) 1,o(3) 2,o(3) 3,o(3) 4,o(3) 5Respectively corresponding to 5 parameters K of a fractional order PID controllerp、Ki、KdAnd an integral order λ, a derivative order μ;
the activation function of the output layer selects a non-negative Sigmoid function:
the weight coefficient of the neural network is adjusted by a gradient descent method through real-time input of an error E (t), the adjustment of the weight coefficient is stopped when the error is within a required range, and the weight coefficient adjustment formula which is transmitted from the error of an output layer to a hidden layer is deduced as follows:
and adjusting the weighting coefficient from the hidden layer to the output layer according to the formula, wherein k is the current iteration number, eta is the learning rate, and alpha is the inertia coefficient.
Further, the BP neural network error satisfies the condition: e (k) -E (k-1) is less than or equal to 1 x 10-6When the iteration is stopped, the neural network stops iterating, and the obtained five parameters are substituted into the Oustaloup filter to carry out the fractional order PIλDμA digital implementation is performed.
Further, the learning rate eta and the inertia coefficient alpha in the BP neural network are regulated by the method:
when | Δ E (k)/Δ E (k-1) | >1,
when | Δ e (k) | < 1/Δ e (k-1) |
Coefficient K0Is generally in the range of [ -1,1 [)]Selecting K0Most suitably 0.5; alpha is alpha0As an initial value, take alpha0=1。
The invention has the beneficial effects that: the grid-connected inverter control method for adjusting the fractional order PID in BP is improved, and the multivariable fractional order PID is controlled to act on the outer ring of the inductive current, so that the system has better robustness, and the THD value of the grid-connected current is greatly reduced; and a proportional controller is used as a capacitance current inner ring to realize double-closing control. Aiming at the problem of parameter setting caused by the introduction of a fractional order PID algorithm, the improved BP neural network algorithm is adopted to identify and optimize parameters, so that the stable control of grid-connected inversion is realized, and the robustness of the system is improved.
Drawings
FIG. 1 is a circuit topology structure diagram of a three-phase LCL grid-connected inverter;
FIG. 2 is a schematic diagram of the structure of the control loop of the inductor current outer loop and the capacitor current inner loop of the method of the present invention;
FIG. 3 is a functional block diagram of a fractional order PID control;
FIG. 4 is a diagram of a BP neural network architecture provided by the present invention;
FIG. 5 is a control block diagram of a BP neural network tuning fractional order PID controller of the present invention;
FIG. 6 is a simulation waveform diagram of grid-connected voltage in a grid-connected inverter control method for BP neural network tuning fractional order PID provided by the present invention;
FIG. 7 is a simulation waveform diagram of a sudden change of current from 10A to 15A in the grid-connected inverter control method for improving BP neural network setting fractional order PID provided by the invention;
fig. 8 is a simulation result diagram of the total harmonic distortion rate in the grid-connected inverter control method for improving the BP neural network setting fractional order PID provided by the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
The topology control structure of the three-phase LCL type grid-connected inverter is shown in figure 1, wherein the LCL type filter is included in figure 1 and mainly comprises an inverter side inductor L1Filter capacitor CfAnd a network side inductor L2Constitution S1~S6A switching tube IGBT of the grid-connected inverter; u shapedcIs DC side voltage, C is DC side filter capacitance icIs a capacitance current i1Is a direct side current, i2Is the grid-connected current.
As shown in fig. 2, the structural schematic diagram of the control loop of the inductive current outer loop and the capacitive current inner loop is constructed by applying a fractional order PID controller to the inductive current outer loop control, and the capacitive current inner loop is controlled by a proportional compensator. Constructing a three-phase LCL grid-connected inverter, and measuring the current i on a capacitor through a current transformercThe proportional controller is selected to eliminate harmonic component of the current high-frequency switch, the resonance peak is inhibited to increase the damping of the system, and the current i on the grid-connected inductor2With a given reference current i2 *Comparing, and obtaining the capacitance current reference signal i of the inner ring by the difference value through a fractional order PID controllerc *Reference signal i of capacitance currentc *And the actual current i of the inner loop of the capacitor currentcComparing the voltage signals to obtain a voltage signal U of the inverter after compensation is carried out by a proportional controllerinv。Fractional order PID controller, G, being an outer loop of the inductor currentcA proportional controller of an inner loop of the capacitance current.
Gc=KcIn which K isp,Ki,KdProportional coefficient, integral coefficient and differential coefficient of the fractional order PID controller are respectively; λ and μ are the integral and differential order of the fractional order PID controller in current, KcIs a proportional control coefficient. G in FIG. 21(s)=1/(L1s+R1),G2(s)=1/Cfs,G3(s)=1/(L2s+R2) Ug is the grid side voltage, R1Is an inverter side inductor resistance, R2Is a net side inductor resistor.
The conventional PID controller parameter setting is often poor in setting, poor in performance, not wide in control range and poor in adaptability to operation conditions, and a fractional order control theory and a PID controller setting theory are combined. The general format of the fractional order PID controller is abbreviated as PIλDμThe lambda-order integral and the mu-order differential are taken, and two adjustable parameters are added compared with an integral-order PID controller. When both λ and μ take 1, it is an integer solution PID controller. Fractional order PIλDμThe controller has greater flexibility and wider adaptability to the controlled system. Since the order can be arbitrarily selected, the fractional order PIλDμThe controller can be selected to a greater extent than a conventional integer order PID controller. Fractional order PIλDμThe controller is insensitive to the change of the controller parameter and the controlled system parameter, and when the parameter of the controller is set and the corresponding parameter changes in a certain range, the fractional order PIλDμThe controller can still effectively control without re-setting parameters, and has basically equivalent control effect and stronger robustness.
FIG. 3 shows a functional block diagram of a fractional order PID control, fractional order PIλDμController and conventional PIDThe controller structure is the same, and fractional orders are respectively introduced in integration and differentiation. The control range of the PID controller is expanded. The differential equation of the fractional order PID controller is:
the integration order λ and the differentiation order μ should satisfy 0< λ, μ < 2.
Fractional order PI based on BP neural networkλDμThe parameter setting algorithm of the controller comprises the following steps:
as shown in fig. 4, the BP neural network structure diagram is constructed such that the neural network has a 2 × 5 × 5 structure, and two inputs are provided to the neural network, one is an inductive current signal i given to a grid-connected inductor2 *The setting in the simulation is 15A, and the other is an error signal E, for a given inductor current signal i2 *And the current signal i obtained by actual measurement2The difference of (a). The hidden layer is selected from five neurons, and the output layer is Kp、Ki、KdAnd an integration order λ, a differentiation order μ.
Network input layer input is oiThe input of the hidden layer is net(2) mThe output is o(2) mThe input of the output layer is net(3) n(ii) a The output is o(3) n. The neural network continuously adjusts the synaptic weight omega of the input layerimAnd hidden layer synaptic weight omegamnReduces the error E. Error e (t) ═ i2(t)-i2 *(t),i2(t) is the actually measured value of the inductance current, i2 *(t) is a given inductor current reference value.
The input and output of the network hidden layer are respectively
Where k is the number of iterations.
The activation function of the hidden layer neuron selects a positive and negative symmetric Sigmoid function.
The input and output of the network output layer are respectively
Output of network layer o(3) 1,o(3) 2,o(3) 3,o(3) 4,o(3) 5Respectively corresponding to 5 parameters K of a fractional order PID controllerp、Ki、KdAnd an integration order λ, a differentiation order μ.
The activation function of the output layer selects a non-negative Sigmoid function:
and (3) adjusting the weight coefficient of the neural network by inputting the error E (t) in real time and adopting a gradient descent method, and stopping the adjustment of the weight coefficient when the error is within the required range. The weight coefficient adjustment formula for the output layer error back-propagation to the hidden layer is derived as follows:
and adjusting a weighting coefficient from the hidden layer to the output layer according to the formula, wherein k is the current iteration number, eta is the learning rate, and alpha is an inertia coefficient, and the formula can be expanded as follows:
Δ u (k) is the output difference of two consecutive fractional order PID controllers, and y (k) is the inductance current value set as the input of the neural network.
Due to the fact thatIt cannot be determined that the sign function sgn is used for replacing the sign function sgn, and the error between the sign function sgn and the sign function sgn is compensated by the learning rate. The same method is adopted to adjust the weight of the hidden layer.
When the condition is satisfied: e (k) -E (k-1) is less than or equal to 1 x 10-6And then, the neural network stops iteration, outputs five parameters of the fractional order PID controller, and digitally realizes the fractional order PID controller by using an improved Oustaloup filtering algorithm.
The discretization fractional order PID control algorithm:
the effect of the fractional order PID controller can be achieved by the S-function in Matlab. The control block diagram of the BP neural network setting fractional order PID controller is shown in FIG. 5, in the whole control loop, the input of the fractional order PID controller is error E, and the output is reference signal i of the inner ring of the capacitance current controlc *。
In a neural network algorithm, selection of an inertia coefficient and a learning rate has great influence on a controlled object, and a controller for fixing the inertia coefficient and the learning rate is easy to generate large overshoot in a control process and has an oscillation problem. And setting the inertia coefficient and the learning rate according to the trend of error change. When the error delta E (k) >0 exists, the error curve has a rising trend, which indicates that the phenomena of output overshoot or output far from expectation and the like of the controller exist, and the values of the two coefficients should be reduced; when the continuous error ratio is reduced, the slow response exists or the output is approaching to the expected value of the controlled object, in order to further ensure that the control result does not have rapid change, the learning rate eta and the inertia coefficient alpha need to be adjusted, and the two coefficients are properly increased to enable the curve to continuously keep the change trend; when Δ E (k) ═ Δ E (k-1) ═ 0, it is indicated that the system output has stabilized.
When | Δ E (k)/Δ E (k-1) | >1,
when | Δ e (k) | < 1/Δ e (k-1) |
Coefficient K0Is generally in the range of [ -1,1 [)]Selecting K from the text after multiple tests0Most suitably 0.5; take alpha01 is an initial value.
Constructing a three-phase LCL type grid-connected inverter, setting a grid-connected current value, and obtaining a reference signal of a current inner ring through a fractional order PID controller of an inductance current outer ring after the difference is made between the grid-connected current value and an actual value of a grid-connected current;
and subtracting the reference signal of the obtained current inner ring from the actual capacitance current, compensating by a proportional compensator to obtain a voltage signal at the side of the inverter, converting the obtained voltage signal by Clark, inputting the converted voltage signal to an SVPWM module to obtain a driving signal for switching on and off a switching tube, and realizing the stable control of the mathematical model of the topological structure of the inverter.
The transfer function of a fractional order PID controller is as follows:a simulation experiment was performed thereon.
TABLE 1
Parameter(s) | Numerical value |
DC side voltage Vdc/ |
700 |
Inverter side inductor L1/mH | 3 |
Grid-connected side inductor L2/mH | 1.5 |
Filter capacitor Cf/uF | 14.1 |
Inverter side inductor resistor R1/Ω | 0.1 |
Network side inductance resistance R2/Ω | 0.1 |
Switching frequency fs/ |
10 |
Table 1 is a simulation system parameter list, and the simulation experimental evaluations are as follows:
FIG. 6 is a graph of the results of the fractional order dual loop control output of the present invention showing: the output voltage under the fractional order double-loop control reaches the same phase and meets the requirement of high sine degree of output waveform of the grid-connected voltage, and the waveform is ensured not to be distorted at the zero-crossing point and the peak value. The control strategy can well filter harmonic waves and improve the grid-connected electric energy quality.
FIG. 7 is a dynamic performance test waveform under external disturbance, which is a current mutation simulation waveform, and when the simulation runs to 0.1s and the sine wave is at the peak top and is most prone to distortion, the current is increased from 10A to 15A. The current recovers to a steady state after 0.05 s. Under the same conditions, the traditional PID controller needs about 0.1s of time. According to the analysis, under the same external disturbance condition, the time required for restoring the integral order PID controller from the influence generated by the fluctuation is longer, and the fractional order PID controller has stronger anti-interference capability.
Fig. 8 is a waveform diagram of the total harmonic distortion of the grid-connected current of the present invention, and it can be seen from the waveform diagram that under the control of the fractional order PID, the THD value of the grid-connected current is 0.76%, and compared with the total harmonic distortion controlled by the conventional PID controller being 1.81%, the system has fewer higher harmonics, and the waveform of the system is better.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (4)
1. A grid-connected inverter control method for improving BP setting fractional order PID is characterized in that an inductive current outer ring control adopts a fractional order PID controller, a capacitance current inner ring adopts a proportion compensator to control a three-phase LCL grid-connected inverter, and a current i on a capacitor is measured through a current transformercThe proportional controller is selected to eliminate harmonic component of the current high-frequency switch, the resonance peak is inhibited to increase the damping of the system, and the current i on the grid-connected inductor2With a given reference current i2 *Comparing, and obtaining the capacitance current reference signal i of the inner ring by the difference value through a fractional order PID controllerc *Reference signal i of capacitance currentc *And the actual current i of the inner loop of the capacitor currentcComparing the voltage signals to obtain a voltage signal U of the inverter after compensation is carried out by a proportional controllerinv(ii) a Proportional coefficient K in fractional order PID controller by BP neural networkpIntegral ofCoefficient KiDifferential coefficient KdAnd carrying out parameter setting on the integral order lambda and the differential order mu.
2. The grid-connected inverter control method for improving BP setting fractional order PID according to claim 1, characterized in that the BP neural network input is an inductive current signal i given on a grid-connected inductor2 *And error E ═ i2-i2 *E is a given inductor current signal i2 *And the current signal i obtained by actual measurement2A difference of (d); the hidden layer is selected from five neurons, and the output layer is Kp、Ki、KdAnd an integral order λ, a derivative order μ;
network input layer input is oiThe input of the hidden layer is net(2) mThe output is o(2) mThe input of the output layer is net(3) n(ii) a The output is o(3) n(ii) a The neural network continuously adjusts the synaptic weight omega of the input layerimAnd hidden layer synaptic weight omegamnReducing the error E;
the input and output of the network hidden layer are respectively
Wherein k is the number of iterations;
the activation function of the hidden layer neuron selects a positive and negative symmetric Sigmoid function,
the input and output of the network output layer are respectively:
output of network layer o(3) 1,o(3) 2,o(3) 3,o(3) 4,o(3) 5Respectively corresponding to 5 parameters K of a fractional order PID controllerp、Ki、KdAnd an integral order λ, a derivative order μ;
the activation function of the output layer selects a non-negative Sigmoid function:
the weight coefficient of the neural network is adjusted by a gradient descent method through real-time input of an error E (t), the adjustment of the weight coefficient is stopped when the error is within a required range, and the weight coefficient adjustment formula which is transmitted from the error of an output layer to a hidden layer is deduced as follows:
and adjusting the weighting coefficient from the hidden layer to the output layer according to the formula, wherein k is the current iteration number, eta is the learning rate, and alpha is the inertia coefficient.
3. The grid-connected inverter control method for improving the BP setting fractional order PID according to claim 2, characterized in that the BP neural network error satisfies the condition: e (k) -E (k-1) is less than or equal to 1 x 10-6When the iteration is stopped, the neural network stops iterating, and the obtained five parameters are substituted into the Oustaloup filter to carry out the fractional order PIλDμA digital implementation is performed.
4. The grid-connected inverter control method for improving the BP setting fractional order PID according to claim 2, characterized in that the learning rate η and the inertia coefficient α in the BP neural network are adjusted by:
when | Δ E (k)/Δ E (k-1) | >1,
when | Δ e (k) | < 1/Δ e (k-1) |
Coefficient K0Is generally in the range of [ -1,1 [)]Selecting K0Most suitably 0.5; alpha is alpha0As an initial value, take alpha0=1。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111454118.8A CN114063443A (en) | 2021-12-01 | 2021-12-01 | Grid-connected inverter control method for improving BP (Back propagation) setting fractional order PID (proportion integration differentiation) |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111454118.8A CN114063443A (en) | 2021-12-01 | 2021-12-01 | Grid-connected inverter control method for improving BP (Back propagation) setting fractional order PID (proportion integration differentiation) |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114063443A true CN114063443A (en) | 2022-02-18 |
Family
ID=80228516
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111454118.8A Pending CN114063443A (en) | 2021-12-01 | 2021-12-01 | Grid-connected inverter control method for improving BP (Back propagation) setting fractional order PID (proportion integration differentiation) |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114063443A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115009278A (en) * | 2022-08-08 | 2022-09-06 | 潍柴动力股份有限公司 | Cruise control method, device, equipment and storage medium |
CN115498851A (en) * | 2022-08-23 | 2022-12-20 | 嘉兴索罗威新能源有限公司 | Intelligent current control method for inverter of photovoltaic system |
CN116184812A (en) * | 2023-04-24 | 2023-05-30 | 荣耀终端有限公司 | Signal compensation method, electronic equipment and medium |
CN117596462A (en) * | 2023-11-21 | 2024-02-23 | 长春理工大学中山研究院 | SMA motor, camera and anti-shake control method of camera |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102566426A (en) * | 2011-12-29 | 2012-07-11 | 大连交通大学 | Fractional order parameter adjustment controller algorithm of PI<alpha>D<beta> controller |
CN106168815A (en) * | 2016-07-27 | 2016-11-30 | 中冶南方工程技术有限公司 | A kind of acid liquor temperature control system based on Neural network PID and method |
CN110262219A (en) * | 2019-06-14 | 2019-09-20 | 广东工业大学 | A kind of motor PID automatic setting method based on BP neural network |
CN110531612A (en) * | 2019-08-16 | 2019-12-03 | 佛山科学技术学院 | A kind of parameter tuning method of Fractional Order PID Controller |
CN113541196A (en) * | 2021-08-10 | 2021-10-22 | 太原理工大学 | Fractional order control method for single-phase LC type grid-connected inverter |
-
2021
- 2021-12-01 CN CN202111454118.8A patent/CN114063443A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102566426A (en) * | 2011-12-29 | 2012-07-11 | 大连交通大学 | Fractional order parameter adjustment controller algorithm of PI<alpha>D<beta> controller |
CN106168815A (en) * | 2016-07-27 | 2016-11-30 | 中冶南方工程技术有限公司 | A kind of acid liquor temperature control system based on Neural network PID and method |
CN110262219A (en) * | 2019-06-14 | 2019-09-20 | 广东工业大学 | A kind of motor PID automatic setting method based on BP neural network |
CN110531612A (en) * | 2019-08-16 | 2019-12-03 | 佛山科学技术学院 | A kind of parameter tuning method of Fractional Order PID Controller |
CN113541196A (en) * | 2021-08-10 | 2021-10-22 | 太原理工大学 | Fractional order control method for single-phase LC type grid-connected inverter |
Non-Patent Citations (3)
Title |
---|
孙昌茂: "输入电压时变的多并网逆变器稳定性研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, pages 47 - 48 * |
金海光: "基于相邻交叉耦合的直线电梯电磁悬浮控制***研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, pages 27 - 29 * |
顾晓虎: "基于 BP 神经网络的LCL型三相并网逆变器控制***设计", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, pages 40 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115009278A (en) * | 2022-08-08 | 2022-09-06 | 潍柴动力股份有限公司 | Cruise control method, device, equipment and storage medium |
CN115009278B (en) * | 2022-08-08 | 2022-11-29 | 潍柴动力股份有限公司 | Cruise control method, device, equipment and storage medium |
CN115498851A (en) * | 2022-08-23 | 2022-12-20 | 嘉兴索罗威新能源有限公司 | Intelligent current control method for inverter of photovoltaic system |
CN115498851B (en) * | 2022-08-23 | 2023-04-25 | 嘉兴索罗威新能源有限公司 | Intelligent current control method for inverter of photovoltaic system |
CN116184812A (en) * | 2023-04-24 | 2023-05-30 | 荣耀终端有限公司 | Signal compensation method, electronic equipment and medium |
CN117596462A (en) * | 2023-11-21 | 2024-02-23 | 长春理工大学中山研究院 | SMA motor, camera and anti-shake control method of camera |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114063443A (en) | Grid-connected inverter control method for improving BP (Back propagation) setting fractional order PID (proportion integration differentiation) | |
CN106230257B (en) | A kind of two-way DC converter feedback linearization contragradience sliding-mode control | |
CN102593862B (en) | Photovoltaic grid-connected inverter and control method thereof | |
CN103972922A (en) | Photovoltaic grid connection control method on basis of improved quasi-resonance control and repeated control | |
CN105958527A (en) | Photovoltaic grid connected inverter multiparameter hysteresis current composite control method | |
Chelli et al. | Study on shunt active power filter control strategies of three-phase grid-connected photovoltaic systems | |
Yu et al. | Multi-step predictive current control for NPC grid-connected inverter | |
CN106292283B (en) | A kind of adaptive fuzzy integral sliding mode control method of photovoltaic combining inverter | |
CN107896071A (en) | A kind of three phase combined inverter based on neutral net | |
CN113411002B (en) | Single-phase inverter control system and method based on sliding mode variable structure of extreme learning machine | |
Tsang et al. | PLL-less single stage grid-connected photovoltaic inverter with rapid maximum power point tracking | |
CN117097189A (en) | Control method of photovoltaic inverter | |
Chigane et al. | Experimental assessment of integral-type terminal sliding mode control designed for a single-phase grid-interlinked PV system | |
CN109888824B (en) | Photovoltaic grid-connected inversion control method based on predictive control | |
CN110176769B (en) | Micro-grid group harmonic compensation method based on two-stage control | |
CN105896591B (en) | Photovoltaic combining inverter self-adaptation control method | |
CN105978018A (en) | LC grid-connected inverter control method | |
Abu-Zaher et al. | Dual-input configuration of three-phase split-source inverter for photovoltaic systems with independent maximum power point tracking | |
Azzam-Jai et al. | Adaptive adaline neural PQ strategy-based multipurpose PV interfaced shunt active power filter | |
Liu et al. | Research on adaptive control of grid-connected PV inverters in weak grid | |
CN112187079B (en) | Single-phase three-level converter system and control method thereof | |
Eslampanah et al. | Artificial Neural Networks for Control of Three Phase Grid Connected T-Type Inverter | |
Zhou et al. | Research on grid-connected photovoltaic inverter based on quasi-PR controller adjusting by dynamic diagonal recurrent neural network | |
Han et al. | A grid-connected current control technique of single-phase voltage source inverter based on BP neural network | |
Garikapati et al. | ANFIS Controlled MMC-UPQC to Mitigate Power Quality Problems in Solar PV Integrated Power 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 |