CN117937412B - Current stabilizing control method for rectifier unit of photovoltaic direct current access electrolytic aluminum power supply system - Google Patents
Current stabilizing control method for rectifier unit of photovoltaic direct current access electrolytic aluminum power supply system Download PDFInfo
- Publication number
- CN117937412B CN117937412B CN202311688363.4A CN202311688363A CN117937412B CN 117937412 B CN117937412 B CN 117937412B CN 202311688363 A CN202311688363 A CN 202311688363A CN 117937412 B CN117937412 B CN 117937412B
- Authority
- CN
- China
- Prior art keywords
- photovoltaic
- rectifier unit
- sub
- electrolytic aluminum
- current
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 229910052782 aluminium Inorganic materials 0.000 title claims abstract description 39
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 title claims abstract description 39
- 230000000087 stabilizing effect Effects 0.000 title claims abstract description 7
- 230000001105 regulatory effect Effects 0.000 claims abstract description 25
- 230000002787 reinforcement Effects 0.000 claims abstract description 14
- 230000009471 action Effects 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 8
- 230000001276 controlling effect Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 7
- 230000005855 radiation Effects 0.000 claims description 3
- 238000009434 installation Methods 0.000 claims description 2
- 238000013178 mathematical model Methods 0.000 claims description 2
- 229920006395 saturated elastomer Polymers 0.000 claims 1
- 238000013528 artificial neural network Methods 0.000 abstract description 11
- 230000008569 process Effects 0.000 abstract description 8
- 230000003287 optical effect Effects 0.000 abstract description 5
- 230000007547 defect Effects 0.000 abstract description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 abstract description 2
- 229910052799 carbon Inorganic materials 0.000 abstract description 2
- 230000005611 electricity Effects 0.000 abstract description 2
- 238000003912 environmental pollution Methods 0.000 abstract description 2
- 230000033228 biological regulation Effects 0.000 description 9
- 238000010248 power generation Methods 0.000 description 9
- 238000004519 manufacturing process Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 210000002569 neuron Anatomy 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000002156 mixing Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- TWNQGVIAIRXVLR-UHFFFAOYSA-N oxo(oxoalumanyloxy)alumane Chemical compound O=[Al]O[Al]=O TWNQGVIAIRXVLR-UHFFFAOYSA-N 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000013082 photovoltaic technology Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
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
- H02J1/00—Circuit arrangements for dc mains or dc distribution networks
- H02J1/10—Parallel operation of dc sources
- H02J1/102—Parallel operation of dc sources being switching converters
-
- C—CHEMISTRY; METALLURGY
- C25—ELECTROLYTIC OR ELECTROPHORETIC PROCESSES; APPARATUS THEREFOR
- C25C—PROCESSES FOR THE ELECTROLYTIC PRODUCTION, RECOVERY OR REFINING OF METALS; APPARATUS THEREFOR
- C25C3/00—Electrolytic production, recovery or refining of metals by electrolysis of melts
- C25C3/06—Electrolytic production, recovery or refining of metals by electrolysis of melts of aluminium
- C25C3/08—Cell construction, e.g. bottoms, walls, cathodes
-
- C—CHEMISTRY; METALLURGY
- C25—ELECTROLYTIC OR ELECTROPHORETIC PROCESSES; APPARATUS THEREFOR
- C25C—PROCESSES FOR THE ELECTROLYTIC PRODUCTION, RECOVERY OR REFINING OF METALS; APPARATUS THEREFOR
- C25C3/00—Electrolytic production, recovery or refining of metals by electrolysis of melts
- C25C3/06—Electrolytic production, recovery or refining of metals by electrolysis of melts of aluminium
- C25C3/16—Electric current supply devices, e.g. bus bars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/092—Reinforcement learning
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01F—MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES
- H01F29/00—Variable transformers or inductances not covered by group H01F21/00
- H01F29/02—Variable transformers or inductances not covered by group H01F21/00 with tappings on coil or winding; with provision for rearrangement or interconnection of windings
-
- 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
- H02J1/00—Circuit arrangements for dc mains or dc distribution networks
- H02J1/10—Parallel operation of dc sources
- H02J1/12—Parallel operation of dc generators with converters, e.g. with mercury-arc rectifier
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
- H02M7/00—Conversion of ac power input into dc power output; Conversion of dc power input into ac power output
- H02M7/02—Conversion of ac power input into dc power output without possibility of reversal
- H02M7/04—Conversion of ac power input into dc power output without possibility of reversal by static converters
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P13/00—Arrangements for controlling transformers, reactors or choke coils, for the purpose of obtaining a desired output
- H02P13/06—Arrangements for controlling transformers, reactors or choke coils, for the purpose of obtaining a desired output by tap-changing; by rearranging interconnections of windings
-
- 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
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
- Y02E10/56—Power conversion systems, e.g. maximum power point trackers
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Chemical & Material Sciences (AREA)
- Theoretical Computer Science (AREA)
- Organic Chemistry (AREA)
- Metallurgy (AREA)
- Materials Engineering (AREA)
- Physics & Mathematics (AREA)
- Electrochemistry (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Control Of Electrical Variables (AREA)
Abstract
The invention discloses a steady flow control method of a rectifier unit of a photovoltaic direct current access electrolytic aluminum power supply system, which comprises the following steps: rectifier bridge, on-load regulating transformer and saturation reactor adjusting component. The patent provides a realization mode of total modulation and split modulation, and improves the defect of poor stability of closed-loop control of the system in the total modulation mode; the current stabilizing control method of the system when the photovoltaic is dominant is provided, the current stabilizing control process of the rectifier unit is guided by utilizing the optical power prediction system and the reinforcement learning algorithm based on the artificial neural network, and the position of the tap of the on-load voltage regulating transformer is dynamically regulated. The related technology greatly lightens the dependence of the electrolytic aluminum industry on traditional fossil energy sources, reduces carbon emission and environmental pollution, stabilizes the cost uncertainty caused by electricity price fluctuation to a certain extent, and improves the market competitiveness of electrolytic aluminum enterprises.
Description
Technical Field
The invention relates to a steady flow control method of a rectifier unit of a power supply system, in particular to a steady flow control method of a rectifier unit of a photovoltaic direct current access electrolytic aluminum power supply system.
Background
Aluminum is a light metal with good strength and corrosion resistance, and is widely used in the fields of aerospace, automobile manufacturing, railway, construction, packaging and the like. China is one of the internationally largest aluminum production and consumption countries, and governments have set forth various policies to encourage the development of the electrolytic aluminum industry. The electrolytic aluminum processing process comprises the following steps: the aluminum oxide is reduced to aluminum at the cathode through the direct current power supply, a large amount of electric energy is consumed in the whole process, and the dependence on a power grid is high.
In recent years, the development of distributed photovoltaic technology provides a new idea for the electrolytic aluminum industry. The distributed photovoltaic system is raised, so that part of required power can be provided for electrolytic aluminum production, unused electric energy can be fed back to a power grid in non-production time, and efficient utilization of energy is realized. In addition, the utilization of photovoltaic power generation greatly lightens the dependence of the electrolytic aluminum industry on traditional fossil energy sources, reduces carbon emission and environmental pollution, stabilizes the cost uncertainty caused by electricity price fluctuation to a certain extent, and improves the market competitiveness of the electrolytic aluminum industry.
The research shows that the electrolytic aluminum factory has high requirement on the stability of a power supply system. Because photovoltaic power generation has larger fluctuation, the output power is very easy to be influenced by environmental factors, engineering technicians need to adopt a rectification system, and the problem of current stabilizing control of an electrolytic aluminum load after photovoltaic access is solved.
At present, most of electrolytic aluminum rectifying systems adopt a mode of on-load voltage regulating transformer, diode rectification and saturation reactor regulation, the on-load voltage regulating transformer is used for coarse adjustment, the action time is as long as 15-30 seconds, frequent actions cannot be performed, and the saturation reactor is used for fine adjustment, but the regulating range is smaller. The rectifying system generally adopts a mode of rectifying by a plurality of rectifying units in parallel so as to improve the reliability. There are generally two modes of total blending and split blending. And under the total regulation mode, the steady flow system performs closed-loop control on the total output current of all the rectifying units, and synchronously regulates the output of each rectifying unit. Under the split-regulation mode, each rectifier unit respectively carries out closed-loop control on the output current of the rectifier unit, and adjusts the control output of the rectifier unit. The total adjustment mode is relatively simple to operate, but the degree of freedom of operators is low; under the split-regulation mode, operators can give different output currents according to the aging states of different rectifying units, so that a plurality of aluminum factories adopt the split-regulation mode.
Some current stabilizing control methods of rectifier units of photovoltaic direct current access electrolytic aluminum power supply systems also exist in the market at present, for example: the Chinese patent 202310581963.4 discloses a power supply system and a method based on the fact that photovoltaic power generation is connected to an electrolytic aluminum direct current bus, the technical scheme directly connects the photovoltaic power generation to an alternating current power grid, the defects of complex system, low conversion efficiency and the like are overcome, and the state of a public power grid can be influenced in the implementation process; china patent 202210267727.0 discloses a method for using distributed photovoltaic power generation to power supply rectification system of electrolytic aluminum plant, in which the output current of photovoltaic system is directly measured and subtracted from the given value of steady flow system, however, the scheme cannot be used for controlling a dispatching unit, and when the power fluctuation range of photovoltaic system exceeds the regulation range of saturation reactor, the system cannot work. In addition, photovoltaic dc access systems are typically provided with an output current closed-loop PI regulator in the controller for closed-loop control of its output current. If the power of the photovoltaic system is large, the regulation of the rectifier unit is in control coupling with the control of the photovoltaic system, and the method for reducing the actual output current of the collected photovoltaic system in the steady-flow control of the rectifier unit can cause the instability of closed-loop control of the system.
Aiming at the defects of the photovoltaic direct current access scheme in the electrolytic aluminum power supply system, the patent provides an effective method for solving the problem of current stabilization control of an electrolytic aluminum load after photovoltaic direct current access.
Disclosure of Invention
Aiming at the problems, the invention provides a steady flow control method of a rectifier unit of a photovoltaic direct current access electrolytic aluminum power supply system, wherein the rectifier unit consists of a plurality of rectifier units, and each rectifier unit comprises an on-load voltage regulating transformer, a rectifier bridge and a reactor regulating assembly;
as a further technical scheme, the steady flow control method of the rectifier unit comprises two modes of total adjustment and split adjustment, and can be selected according to the setting;
the control method of the total modulation mode comprises the following steps: subtracting a given value of total output current of the photovoltaic system from a manual given current of the rectifying system, and controlling the given value as an actual given value of the rectifying system;
The control mode of the sub-modulation mode is as follows: firstly, reading the manual given value of each current sub-regulating unit, and summing to obtain a total given value; secondly, dividing the manual given value of each sub-regulating unit by the total given value to obtain the power duty ratio of the sub-regulating unit; and finally, multiplying the given value of the total output current of the photovoltaic system by the power duty ratio of each sub-regulator set to be used as the actual given value of the sub-regulator set, and executing sub-regulation control by the sub-regulator set.
As a further technical scheme, the total output current given value of the photovoltaic system is: all connected photovoltaic direct current systems output the sum of the given values of the current closed-loop regulator.
As a further technical scheme, according to the current operation working condition of the rectifier unit, the position of the tap of the on-load voltage regulating transformer of the rectifier unit is regulated, even if the regulation range of the rectifier unit is released, so that the system cannot excessively limit the photovoltaic output due to insufficient regulation range of the rectifier unit.
As a further technical scheme, an optical power prediction system is arranged, and the tap position of the on-load regulating transformer of the rectifier unit is adjusted according to the predicted change range of the photovoltaic output power, so that the rectifier unit has an adjusting range matched with the change range of the photovoltaic output power.
As a further technical scheme, the on-load voltage-regulating transformer tap position of the rectifier unit is regulated by adopting an artificial neural network reinforcement learning algorithm, and the artificial neural network is trained according to the history control error, so that the system can not excessively regulate the on-load voltage-regulating transformer tap frequently, and can not excessively limit the photovoltaic output due to insufficient regulating range of the rectifier unit.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. The steady flow control method of the total regulation mode is improved, the steady flow control deduction is carried out by replacing the actual measurement value of the output current with the given value of the output current of the photovoltaic system, and the closed loop control stability of the system is improved;
2. a steady flow control method of a split-tuning mode is provided;
3. a steady flow control method suitable for a system with larger power occupation of a photovoltaic system is provided.
Drawings
The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention.
Fig. 1: scheme structure diagram of photovoltaic direct current access electrolytic aluminum power supply system.
Fig. 2: an implementation framework based on an optical power prediction algorithm.
Fig. 3: an example of an implementation of an artificial neural network reinforcement learning algorithm.
Detailed Description
For the purpose of making apparent the objects, technical solutions and advantages of the present invention, the present invention will be further described in detail with reference to the following examples and the accompanying drawings, wherein the exemplary embodiments of the present invention and the descriptions thereof are for illustrating the present invention only and are not to be construed as limiting the present invention.
Example 1
FIG. 1 shows a system composition of a photovoltaic direct current access electrolytic aluminum power supply system rectifier unit, the rectifier unit is composed of a plurality of rectifier units, each rectifier unit comprises an on-load voltage regulating transformer, a rectifier bridge and a reactor regulating assembly;
the steady flow control method of the rectifier unit comprises two modes of total adjustment and split adjustment, and can be selected according to the setting;
The control method of the total modulation mode comprises the following steps: subtracting a given value of total output current of the photovoltaic system from a manual given current of the rectifying system, and controlling the given value as an actual given value of the rectifying system;
The control mode of the sub-modulation mode is as follows:
s1: reading the manual given value of each current sub-regulating unit, and summing to obtain a total given value;
S2: dividing the manual given value of each sub-regulating unit by the total given value to obtain the power duty ratio of the sub-regulating unit;
S3: and multiplying the given value of the total output current of the photovoltaic system by the power duty ratio of each sub-regulating unit to be used as the actual given value of the sub-regulating unit, and executing sub-regulating control by the sub-regulating unit.
The total output current given value of the photovoltaic system is as follows: all connected photovoltaic direct current systems output the sum of the given values of the current closed-loop regulator.
And according to the current operation working condition of the rectifier unit, the tap position of the on-load voltage regulating transformer of the rectifier unit is regulated, and even if the regulation range of the rectifier unit is released, the system can not limit the photovoltaic output due to insufficient regulation range of the rectifier unit.
Example two
Fig. 2 shows an implementation framework of an optical power prediction module, where the rectifier unit can adjust the tap position of the on-load tap-changing transformer of the rectifier unit according to the predicted change range of the photovoltaic output power based on the optical power prediction system, so that the rectifier unit has an adjustment range matched with the change range of the photovoltaic output power.
The prediction method of the photovoltaic output power comprises the following steps: a prediction method based on statistical characteristics, a time sequence analysis method, an artificial intelligent algorithm, a prediction method based on a physical model and the like. In this embodiment, taking an Artificial Neural Network (ANN) algorithm as an example, a detailed process of photovoltaic output power prediction is shown, and the specific implementation steps are as follows:
S1: data preprocessing: the historical operation data of the photovoltaic system is collected, and mainly comprises: meteorological data such as photovoltaic power generation capacity, solar radiation intensity, temperature, humidity and the like and mounting characteristics of a solar panel;
s2: data preprocessing: carrying out normalization processing on the data, subsequently removing abnormal values and error data, and randomly dividing a data set into a training set, a verification set and a test set;
S3: selecting parameters of a neural network model: selecting a proper neural network structure, and determining the quantity of neurons of an input layer, hidden layers, neurons and neurons of an output layer; in this example, the amount to be predicted is the photovoltaic output power value, so the number of neurons in the output layer is selected to be 1;
s4: model training and verification: selecting an Adam optimizer;
S5: the algorithm is deployed at the rectifier unit end of the electrolytic aluminum power supply system;
s6: iterative optimization is carried out, and the prediction accuracy of the model is continuously improved.
The photovoltaic power generation power prediction algorithm can help enterprises predict the power generation amount in a period of time in the future, so that decisions can be made in advance, the direct current power grid can be ensured to stably run, and the safety of the electrolytic aluminum process is improved.
Example III
Fig. 3 shows an implementation example of an artificial neural network reinforcement learning algorithm, in which an artificial neural network reinforcement learning algorithm is used to adjust the tap position of the on-load tap changer of the rectifier unit, and the artificial neural network is trained according to the historical control error, so that the system does not adjust the tap of the on-load tap changer too frequently, and does not limit the output of the photovoltaic too much due to the insufficient adjustment range of the rectifier unit.
Reinforcement learning (Reinforcement Learning) is the training of machine learning models to make a series of decisions, with the emphasis that under certain circumstances an Agent takes appropriate action to maximize the desired reward. In the smart grid decision, the specific implementation steps of the reinforcement learning algorithm are as follows:
S1: problem description and environment definition: mainly comprises a state space, an action space and a reward signal; wherein, the state space mainly represents state parameters, voltage fluctuation amplitude and the like related to the electrolytic aluminum production process; the action space mainly expresses scheduling instructions, tap positions of the on-load tap-changing transformer and the like;
s2: suitable reinforcement learning algorithms: the project implementation process adopts Deep Q Networks (DQN) algorithm;
S3: model training: training an RL model by adopting a OpenAI Gym framework, continuously injecting a dataset, continuously adjusting the learning rate and the design rule of a reward function, performing iterative optimization on a reinforcement learning algorithm, and continuously improving the capability of the algorithm;
s4: and (3) verification and test: verifying a learned strategy in Gym simulation environments, testing in a subsequent actual environment, and evaluating the performance of an algorithm;
s5: algorithm deployment: after a satisfactory reinforcement learning model is obtained, the result is deployed into a finishing unit and is applied to a photovoltaic direct current access electrolytic aluminum power supply system.
Example IV
When the tap position of the on-load voltage regulating transformer is optimized by adopting a deep reinforcement learning algorithm, the specific implementation process is as follows:
1. determination of the state space: the parameters selected were:
Energy source output: the solar power generation of each node, 、、、 ；
Load: real-time load demand for electrolytic aluminum plants；
Output voltage of on-load regulating transformer；
2. The tap position of the on-load voltage regulating transformer is adopted in the action space: the tap connections to different turns positions of the secondary winding can produce different output voltage levels. The higher tap typically corresponds to a higher output voltage, while the lower tap typically corresponds to a lower output voltage. When a reinforcement learning algorithm is used in the photovoltaic direct current access electrolytic aluminum power supply system, the tap position of the on-load voltage regulating transformer is used as an action space, and the variation range is 0-1; wherein 0: representing that the tap is connected to the lowest position; 1: representing the tap connected to the highest position.
3. State transfer function: establishing a mathematical model, describing the relationship among solar energy input power, load change and on-load tap position of a voltage regulating transformer in a direct current power grid, and finally obtaining the next state from the current state and action:
Expression of solar input power:
Wherein, In order to obtain the intensity of the solar radiation,In order to be able to determine the temperature,In order to be a degree of humidity,Power prediction is by an artificial neural network as a function of the installation location, with specific reference to example two.
Load power demandIs provided by an actual electrolytic aluminum plant.
The system power balance equation is:
Wherein, Indicating the position of the tapAnd system losses in terms of on-load transformer output voltage.
The relation between the output voltage of the on-load transformer and the tap position is as follows:
Wherein, Is the input voltage of the on-load transformer, the value of which is equal to the output voltage of the photovoltaic system,Is a function of the characteristics of the transformer.
Bonus function: combining with practical operation experience, introducing an energy cost control model when designing a reward function, and smoothing load peaks in the power system as much as possible; state regularization is introduced, and stability of the algorithm is improved. Dynamic adjustment of weights: the reward weight is adjusted step by step to balance different learning objectives.
The foregoing detailed description of the embodiments and the technical solutions and advantageous effects of the present invention will be presented only by way of example, and it should be understood that the foregoing description is not intended to limit the scope of the invention, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention should be construed as being included in the scope of the invention.
Claims (4)
1. A current stabilizing control method for a rectifier unit of a photovoltaic direct current access electrolytic aluminum power supply system is characterized by comprising the following steps of: the rectifying unit consists of a plurality of rectifying units, and each rectifying unit comprises a rectifying bridge, an on-load voltage regulating transformer and a saturated reactor regulating assembly; the steady flow control method of the rectifier unit comprises two modes of total adjustment and split adjustment, and can be selected according to the setting;
the control method of the total modulation mode comprises the following steps: subtracting a given value of total output current of the photovoltaic system from a manual given current of the rectifying system, and controlling the given value as an actual given value of the rectifying system;
The control mode of the sub-modulation mode is as follows: firstly, reading the manual given value of each current sub-regulating unit, and summing to obtain a total given value; dividing the manual given value of each sub-regulating unit by the total given value to obtain the power duty ratio of the sub-regulating unit; finally, multiplying the given value of the total output current of the photovoltaic system by the power ratio of each sub-regulator set to be used as a given subtraction value of the sub-regulator set, subtracting the given subtraction value from the given value of the sub-regulator set to be used as an actual given value of the sub-regulator set, and executing sub-regulation control by the sub-regulator set;
The steady flow control method adopts a deep reinforcement learning algorithm; establishing a mathematical model, describing the relationship among solar input power, load change and on-load tap position of a direct-current power grid, and predicting the next state according to the current state and action to obtain a system balance equation:
The solar energy input power is as follows:
；
Wherein, For the intensity of solar radiation,/>Is the temperature/>Is humidity,/>As a function of the installation location,/>、、/> Solar energy generating capacity for each node;
The system balance equation is:
；
Wherein, For real-time load demand of electrolytic aluminum plants,/>Is the system loss; /(I)For tap position,/>Is the output voltage of the on-load tap changing transformer.
2. The method for controlling the steady flow of the rectifier unit of the photovoltaic direct current access electrolytic aluminum power supply system according to claim 1 is characterized in that: the total output current given value of the photovoltaic system is as follows: all connected photovoltaic direct current systems output the sum of the given values of the current closed-loop regulator.
3. The method for controlling the steady flow of the rectifier unit of the photovoltaic direct current access electrolytic aluminum power supply system according to claim 1 is characterized in that: and the method also comprises the step of adjusting the tap position of the on-load voltage-regulating transformer of the rectifier unit according to the current operation working condition of the rectifier unit, so that the system cannot excessively limit the photovoltaic output due to insufficient adjustment range of the rectifier unit.
4. The method for controlling the steady flow of the rectifier unit of the photovoltaic direct current access electrolytic aluminum power supply system according to claim 1 is characterized in that: the method also comprises the step of setting a light power prediction system, and adjusting the tap position of the on-load voltage regulating transformer of the rectifier unit according to the predicted change range of the photovoltaic output power, so that the rectifier unit has an adjusting range matched with the change range of the photovoltaic output power.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311688363.4A CN117937412B (en) | 2023-12-11 | 2023-12-11 | Current stabilizing control method for rectifier unit of photovoltaic direct current access electrolytic aluminum power supply system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311688363.4A CN117937412B (en) | 2023-12-11 | 2023-12-11 | Current stabilizing control method for rectifier unit of photovoltaic direct current access electrolytic aluminum power supply system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117937412A CN117937412A (en) | 2024-04-26 |
| CN117937412B true CN117937412B (en) | 2024-06-25 |
Family
ID=90761864
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311688363.4A Active CN117937412B (en) | 2023-12-11 | 2023-12-11 | Current stabilizing control method for rectifier unit of photovoltaic direct current access electrolytic aluminum power supply system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117937412B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102944998A (en) * | 2012-11-08 | 2013-02-27 | 国电南瑞科技股份有限公司 | Method for modeling electrolytic aluminium rectification system |
| CN110535146A (en) * | 2019-08-27 | 2019-12-03 | 哈尔滨工业大学 | The Method for Reactive Power Optimization in Power of Policy-Gradient Reinforcement Learning is determined based on depth |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE2218806C3 (en) * | 1972-04-18 | 1975-02-27 | Siemens Ag, 1000 Berlin Und 8000 Muenchen | Rectifier system for feeding aluminum electrolysis furnaces |
| US8106537B2 (en) * | 2008-07-01 | 2012-01-31 | Satcon Technology Corporation | Photovoltaic DC/DC micro-converter |
| US20140197736A1 (en) * | 2013-01-15 | 2014-07-17 | Osram Sylvania Inc. | Filament detection circuit |
| CN116231616A (en) * | 2022-12-12 | 2023-06-06 | 国家电投集团云南国际电力投资有限公司 | Distributed photovoltaic direct current access electrolytic aluminum power supply system |
| CN116417986A (en) * | 2023-05-23 | 2023-07-11 | 青海省投资集团有限公司 | Power supply system and method based on photovoltaic power generation and access to electrolytic aluminum direct current bus |
-
2023
- 2023-12-11 CN CN202311688363.4A patent/CN117937412B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102944998A (en) * | 2012-11-08 | 2013-02-27 | 国电南瑞科技股份有限公司 | Method for modeling electrolytic aluminium rectification system |
| CN110535146A (en) * | 2019-08-27 | 2019-12-03 | 哈尔滨工业大学 | The Method for Reactive Power Optimization in Power of Policy-Gradient Reinforcement Learning is determined based on depth |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117937412A (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112232980B (en) | Regulation and control method for heat pump unit of regional energy heat supply system | |
| CN107301472B (en) | Distributed photovoltaic planning method based on scene analysis method and voltage regulation strategy | |
| CN105846461B (en) | Control method and system for large-scale energy storage power station self-adaptive dynamic planning | |
| Yu et al. | Stochastic optimal relaxed automatic generation control in non-markov environment based on multi-step $ Q (\lambda) $ learning | |
| CN114336702B (en) | Wind-solar storage station group power distribution collaborative optimization method based on double-layer random programming | |
| CN109765787B (en) | Power distribution network source load rapid tracking method based on intraday-real-time rolling control | |
| Hong et al. | Optimal VAR planning considering intermittent wind power using Markov model and quantum evolutionary algorithm | |
| CN110783925A (en) | Two-stage deployment power grid regulation and control model center control system and method | |
| CN114243794A (en) | Wind, light and water multi-energy complementary system flexibility requirement quantification and coordination optimization method | |
| CN117439101B (en) | Intelligent network for interaction of new energy and flexible load in power grid | |
| CN113536694B (en) | Robust optimization operation method, system and device for comprehensive energy system and storage medium | |
| CN106786702A (en) | Full range modeling for mixed energy storage system predicts energy dispatching method | |
| CN117852927A (en) | Source network charge storage integrated planning production simulation method, system and storage medium | |
| CN116722605B (en) | Power distribution network scheduling optimization method based on Internet of things | |
| CN117937412B (en) | Current stabilizing control method for rectifier unit of photovoltaic direct current access electrolytic aluminum power supply system | |
| CN116961008A (en) | Micro-grid capacity double-layer optimization method considering power spring and load demand response | |
| CN116154875A (en) | Photovoltaic power station active power optimization distribution method based on TCN and error function | |
| CN115983492A (en) | Hydrogen-containing multi-energy system operation carbon emission optimization decision method, system and medium | |
| CN115619153A (en) | Response method and device considering distributed resource operation characteristics under dynamic aggregation | |
| CN114218803A (en) | Distributed power supply grid connection method considering source load response characteristics | |
| CN114301062A (en) | Distributed energy optimization treatment device | |
| CN117713211B (en) | Photovoltaic grid-connected intelligent scheduling method and system based on environmental analysis | |
| CN118249420B (en) | Light storage direct-soft system power scheduling method and device based on model prediction | |
| Zhang et al. | Optimal Scheduling Strategy for Power Systems Containing Offshore Wind Farms Considering Wind Power Uncertainty | |
| CN117974365B (en) | Multi-objective operation optimization method and system for electric heating comprehensive energy coupling 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 |