A voltage detection intelligent control peak-shaving and valley-filling device and method based on FPGA

Abstract

The present technology belongs to the technical field of power grid power supply and load balancing, and discloses a voltage detection intelligent control peak-shaving and valley-filling device and method based on FPGA. The present invention can detect the change of the AC voltage of the power grid in real time by setting the learning state, peak-shaving and valley-filling working mode, and profit working mode, and mobilize the generators and pumps in the many widely distributed pumped storage power stations within the range of 249V to 190V, and cooperate to complete the intelligent peak-shaving and valley-filling of the power grid power supply and load, so as to make the power supply resources of the power grid output electric energy as much as possible, improve the supply and demand balance of the power grid, and improve the efficiency of power generation equipment, which has important application value. At the same time, the present invention has the learning ability of artificial intelligence, and can re-learn according to the changes in the power output and load of the power grid, as well as the increase of pumped storage power stations, and constantly adapt to new changes in the power grid. And the core control function of the system is completed based on the FPGA chip, which has faster operation speed, high reliability, and system reprogrammability.

Description

Technical Field

[0001] The present technology belongs to the technical field of power grid power supply and load balancing, and in particular relates to a voltage detection intelligent control peak shaving and valley filling device and method based on FPGA. Background Art

[0002] As renewable energy plays an increasingly important role in the field of power supply applications, the installed capacity and power generation capacity of power generation, including thermal power, nuclear power, hydropower, tidal power, wind power, and photovoltaic power, are very impressive. However, its power generation efficiency is less than half of the power generation capacity. One of the important factors affecting power output is the intermittent and volatile nature of renewable energy generation, followed by the insufficient capacity of the power grid to absorb renewable energy, resulting in the abandonment of wind and solar power, and even the shutdown of thermal power, which reduces power generation efficiency. How to stabilize the voltage of the power grid by adjusting the power generation and power consumption of both the supply and demand sides of the power grid to achieve a dynamic balance, so as to avoid affecting some voltage-sensitive electrical equipment, has become a major problem in power supply of the power grid. Summary of the invention

[0003] In order to solve the above problems, the present invention provides a voltage detection intelligent control peak shaving and valley filling device based on FPGA. The present invention can intelligently control the working state of the generator set and pumping unit in the widely distributed pumped storage power station by detecting and sensing the grid voltage and using FPGA chip for machine learning, so as to change the power supply and load of the grid, thereby achieving the purpose of grid peak shaving and valley filling. The specific technical scheme of the voltage detection intelligent control peak shaving and valley filling device based on FPGA is as follows.

[0004] The voltage detection intelligent control peak-shaving and valley-filling device based on FPGA includes a clock control module, a step-down rectification and filtering module, an analog / digital conversion module, an analog / digital conversion control module, a power grid voltage detection module, a transformer control learning module, a hydropower control module, a pumping control module, and a wind power / photovoltaic power control module; the clock control module, the analog / digital conversion control module, the power grid voltage detection module, the transformer control learning module, the wind power / photovoltaic power control module, the hydropower control module, and the pumping control module are integrated in the FPGA chip; the FPGA chip is connected to an external clock; the step-down rectification and filtering module steps down the power grid voltage to a low voltage and performs full-wave rectification, and then performs filtering and voltage division to obtain a DC voltage signal to be measured; the analog / digital conversion module converts the DC voltage signal to be measured into a digital signal; the analog / digital conversion control module provides a chip selection signal for the analog / digital conversion module and converts and drives an external clock; and also converts the analog / digital The digital signal output serially by the module is converted into a parallel output signal, which is used as the input data of the grid voltage detection module; the grid voltage detection module converts the detected grid voltage data into the addressing address of the memory in the transformer control learning module; the transformer control learning module is used to learn the control data of the amount of power generation equipment that needs to be started when the grid voltage drops to balance the increased load of the grid, the control data of the amount of pumped storage equipment that needs to be started when the grid voltage rises to balance the reduced load of the grid, how to select wind power generators and photovoltaic generators that are in normal power generation state and / or in shutdown maintenance state, and the start and stop time of generators and pumps when in profit mode; the learning data is stored in the storage unit of the FPGA, and the hydropower control module, wind power / photovoltaic power control module, and pumping control module are controlled according to the learning content to respectively realize hydropower generation control, wind power generation control, photovoltaic power generation control, and pumping control.

[0005] The present invention also provides a voltage detection intelligent control peak shaving and valley filling method based on FPGA, which applies the voltage detection intelligent control peak shaving and valley filling device based on FPGA, and comprises the following steps:

[0006] S1. The grid voltage is stepped down to a low voltage and full-wave rectified, and then filtered and divided, so that the output at this time is a preset standard DC voltage when the grid voltage is 220V;

[0007] S2. Learn the control data of the amount of power generation equipment that needs to be started when the grid voltage drops to balance the increased load of the grid, and learn the control data of the amount of pumped storage equipment that needs to be started when the grid voltage rises to balance the reduced load of the grid; learn how to select wind power generators and photovoltaic generators that are in normal power generation state and / or in shutdown maintenance state; learn the start and stop time of generators and pumps when in profit mode; store the learning data in the storage unit of FPGA;

[0008] S3. Perform peak-shaving and valley-filling control: collect the grid voltage at preset time intervals; convert the collected grid voltage into addressing data of the storage unit of the FPGA; control the start and stop status of the generator and pump in the pumped-storage power station according to the range of the grid voltage by reading the corresponding storage unit; and the generator and pump do not work at the same time;

[0009] S4. Control the profit model: Use the learning data written in the storage unit of FPGA to control the pumped storage power station to pump water and store energy when the learned grid price is low, and release water to generate electricity at the specified time when the learned grid price is high; the wind power / photovoltaic motor power generation period is not restricted by time, and the generated electricity can be used for both electricity sales and pumping. The number of equipment started and stopped is controlled by the learning data, which is convenient for shutting down the motor when repairing the equipment.

[0010] When the peak load is reduced and the valley is filled, the task of the pumped storage power station is to start the pump to pump the water from the lower reservoir to the upper reservoir when the load of the power grid is low, convert the excess power of the power grid into the potential energy of water and store it, and consume a large amount of power to balance the load of the power grid. When the load of the power grid increases, the upper reservoir at a high position is used to store water to drive the generator to generate electricity, and input electricity to the power grid to balance the power supply demand of the power grid. When profit control is carried out, the task of the pumped storage power station is to generate as much electricity as possible, including using cheap electricity pumped and stored at night when the load of the power grid is low, generating electricity during the high electricity price period when the load of the power grid is peak during the day, and using the water surface of the reservoir to build a solar photovoltaic power station to generate electricity, and using the water surface of the reservoir to build a wind power station to generate electricity, so as to maximize the economic benefits. The generated electricity can be used for both electricity sales and pumping.

[0011] When the present invention performs peak shaving and valley filling control, the AC voltage of the power grid of 249V to 190V is detected in real time, and after the voltage reduction, rectification, filtering, and voltage dividing circuit, it is converted into a DC voltage of 3.237V to 2.47V to be detected, and then the digital signal is output after analog / digital conversion, and then converted into a memory address. Each address stores data corresponding to the voltage that should start or stop the generator and pump in the pumped storage power station to work. The detected real-time power grid voltage data, the control signal for the generator and pump obtained after analysis, is remotely transmitted to the pumped storage power stations distributed in various places, or is short-rangely transmitted to the local pumped storage power station, thereby realizing the start and stop control of the generator and pump.

[0012] Compared with the prior art, the present invention has the following beneficial effects:

[0013] The present invention can detect the change of AC voltage of the power grid in real time, and mobilize the generators and pumps in many widely distributed pumped storage power stations within the range of 249V to 190V, and cooperate to complete the intelligent peak-shaving and valley-filling of power supply and load of the power grid, so as to make the power supply resources of the power grid output electric energy as much as possible, improve the supply and demand balance of the power grid, and improve the efficiency of power generation equipment, which has important application value. At the same time, the present invention has the learning ability of artificial intelligence, and can relearn according to the changes of power output and load of the power grid, as well as the increase of pumped storage power stations, and constantly adapt to new changes in the power grid. The core control function of the system is completed based on FPGA chip, which has faster operation speed, high reliability, and system reprogrammability. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0015] Figure 1 This is a schematic diagram of the principle of the voltage detection intelligent peak shaving and valley filling device based on FPGA in Example 1.

[0016] Figure 2 This is a schematic diagram of the circuit principle of the buck rectifier and filter module of Example 1.

[0017] Figure 3 Schematic diagram of the principle of the analog / digital conversion module of Example 1.

[0018] Figure 4 This is a diagram of the main circuit modules inside the FPGA chip of Example 1.

[0019] Figure 5 Schematic diagram of the connection relationship of the long / short range transmission module of Example 2. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Example 1

[0022] like Figure 1As shown, this embodiment provides a voltage detection intelligent control peak-shaving and valley-filling device based on FPGA, including a clock control module and a buck rectifier filter module, an analog / digital conversion module, an analog / digital conversion control module, a power grid voltage detection module, and a transformer control learning module connected in sequence; wherein the buck rectifier filter module is externally connected to the power grid; the transformer control learning module is respectively connected to a hydropower control module, a pumping control module, and a wind power / photovoltaic power control module. The hydropower control module, the wind power / photovoltaic power control module, and the pumping control module respectively realize hydropower generation control, wind power generation control, photovoltaic power generation control, and pumping machine control through a long / short range transmission module. The clock control module is respectively connected to the analog / digital conversion module, the analog / digital conversion control module, and the power grid voltage detection module.

[0023] Among them, the clock control module, analog / digital conversion module, analog / digital conversion control module, grid voltage detection module, transformer control learning module, wind power / photovoltaic power control module, hydropower control module, and pumping control module are integrated in the FPGA chip. The crystal oscillator clock connected to the outside of the FPGA chip is a crystal oscillator.

[0024] The working principle of this peak shaving and valley filling device is as follows:

[0025] The clock control module divides the 100MHz square wave signal generated by the crystal oscillator into a 100kHz square wave clock signal, and provides a driving clock signal for the analog / digital conversion module, the analog / digital conversion control module, the grid voltage detection module, the transformer control learning module, the power generation control module, the pumping control module, and the wind power / photovoltaic power control module; the clock control module also generates 10-minute timing and 24-hour timing signals, and provides clock signals for the power generation control module and the pumping control module;

[0026] The analog / digital conversion control module provides chip selection signals and converts and drives external clocks for the analog / digital conversion module outside the FPGA chip; it also converts the 8-bit detected digital signal output serially by the analog / digital conversion module into an 8-bit parallel output signal as input data for the grid voltage detection module;

[0027] The grid voltage detection module converts the detected grid voltage data into the addressing address of the memory in the voltage transformation control learning module;

[0028] When the transformer control learning module is in the learning state, it obtains the learning content from the learning address and the learning data bus and stores it in the corresponding storage unit; when in the working state, it reads the content in the unit corresponding to the input address, determines whether to control the generator or the pump after analysis, and then outputs the control data in the unit to the power generation control module and the pumping control module;

[0029] The power generation control module and the pumping control module decode the input control data to determine which units to start or stop;

[0030] The wind power / photovoltaic power control module reads the data output by the transformer control learning module to determine which wind turbines and photovoltaic generators are to be operated or stopped.

[0031] As a specific implementation method, Figure 2 As shown, the step-down rectification and filtering module includes a transformer, a bridge rectifier circuit, a π-type filter circuit and a variable resistor voltage divider unit connected in sequence, and the voltage across the variable resistor voltage divider unit is used as a DC voltage to be measured.

[0032] The control method for converting the grid voltage into a DC voltage to be tested is as follows: the 220V AC voltage input from the grid is stepped down to a low-voltage AC signal U1 of about 9V to 18V by a transformer T, and then a pulsating DC signal U2 is outputted by a bridge rectifier circuit composed of a rectifier diode D1, a rectifier diode D2, a rectifier diode D3, and a rectifier diode D4, and then a DC signal U3 is obtained after the AC component is filtered out by a π-type filter circuit composed of a capacitor C1, a resistor R1, and a capacitor C2, and finally a 2.87V DC voltage V0 to be tested is obtained by voltage division by a variable resistor R2. The DC voltage V0 to be tested is not directly taken from the DC voltage at both ends of C2, but is taken from the voltage division of R2. The advantage is that the component values ​​of the transformer T, the rectifier diodes D1 to D4, the filter circuit capacitor C1, the resistor R1, and the capacitor C2 will have errors. To accurately obtain the 2.87V DC voltage V0 to be tested from both ends of C2, the accuracy of the components will be very high, which will increase the difficulty of circuit design and the cost of the device. When the variable resistor R2 is used for voltage division, as long as the two ends of the capacitor C2 have a rough value of about 3V, the precise voltage division value can be obtained by adjusting the variable resistor R2 to make V0=2.87V.

[0033] As a specific implementation method, Figure 3 As shown in the figure, the analog / digital conversion module is a TLC549 chip. The control process of converting the DC voltage to be tested into a digital signal is as follows: the rectified DC signal V0 to be tested is input from the ANIN pin of the TLC549 chip, and after the module conversion, an 8-bit digital signal is serially output from the DATAOUT pin. When the pin input is high level, the input analog voltage ANIN is linearly encoded using the pulse input from the CLKIN pin. When the pin input is low level, it is controlled by the CLKIN pin input clock beat to output the encoded digital signal corresponding to the grid voltage bit by bit, so that the subsequent FPGA chip can identify the voltage size. The main advantage of using the TLC549 chip as the analog / digital conversion is that the chip has fewer peripheral circuits when working, which is conducive to saving costs and improving circuit reliability, and the serial / parallel conversion of data is completed by the powerful FPGA.

[0034] like Figure 4 As shown, as a specific implementation method, this embodiment adopts the Vivado2019.2 development platform of XILINX Company of the United States, selects the Artix-7 series chip xc7a35tfgg484-2, and connects the main circuit module structures inside the FPGA chip through Verilog HDL programming design. Among them, timf is the clock control module U1, tlc549adc is the analog / digital conversion control module U2, vchk is the grid voltage detection module U3, vlearn is the transformer control learning module U4, vpwctr is the power generation control module U5, vpuctr is the pumping control module U6, and wspctr is the wind power / photovoltaic power control module U7.

[0035] The specific process of the FPGA chip controlling the voltage digital signal to be measured is as follows: the 100MHz crystal oscillator clock clk from the outside of the FPGA chip is divided and timed in the clock control module U1. The output frequency is an excitation clock signal clk100k of 100kHz, which provides a driving excitation signal for the analog / digital conversion control module U2, the grid voltage detection module U3, the transformer control learning module U4, the power generation control module U5, the pumping control module U6, and the wind power / photovoltaic power control module U7; the clk10m signal output by U1 with a period of 10 minutes provides the power generation control module U5 and the pumping control module U6 with an excitation signal for starting and stopping the generator and the pump every 10 minutes in the regulation mode; the cnt24h signal output by U1 with a period of 24 hours provides the power generation control module U5 and the pumping control module U6 with a reference timing signal for starting and stopping the generator and the pump every day in the profit mode.

[0036] Under the stimulation of the clock signal clk100k, the analog / digital conversion control module U2 generates a chip select signal csout with 16 clk100k cycles to control the operation of the analog / digital conversion chip TLC549, adopts the control clock clkout, and completes the work of converting the serial signal datain output from the TLC549 chip into an 8-bit parallel digital signal rev[7:0].

[0037] When the system is in working state, the grid voltage detection module U3 reads the 8-bit parallel digital signal rev[7:0] output by the analog / digital conversion control module under the stimulation of the clock signal clk100k, the chip selection signal csout, and the working state control signal rw=1, and converts it into the addressing address vramadr signal of the memory vram, thereby establishing the association between the grid voltage change and the memory address change. When the system is in the learning state rw=0, the external input address signal lrnadr[5:0] is converted into the memory addressing address vramadr.

[0038] Under the stimulation of clock signal clk100k, memory addressing address vramadr, and working state control signal rw=0, the transformer control learning module U4 writes the contents of the learning data bus lrn[7:0] into the corresponding memory unit vram[vramadr]. This includes the control data vram

[29] ~vram[0] of how to start the generator set when the grid voltage drops to 219V~190V; the control data vram

[31] ~vram

[60] of how to start the pumping unit when the grid voltage rises to 251V~222V; the data vram

[30] of the generator and the pumping unit not working when the grid voltage is normally 220V; the control vram

[61] of the wind power / photovoltaic generator set; the starting working time vram

[62] of the generator set in the profit mode; the stopping time vram

[63] of the generator set in the profit mode; the control of stopping the pumping unit when the generator is working and stopping the generator when the pumping unit is working is also completed in the transformer control learning module U4. The transformer control learning module U4 has 5 bus outputs to control the operation of generators, pumps, and wind power / photovoltaic generator sets, including: generator operation control bus vpwbus[7:0], pump operation control bus vpubus[7:0], generator start operation control bus wpstart[7:0], generator end operation control bus wpend[7:0], and wind power / photovoltaic generator set operation control bus wsbus[7:0].

[0039] Under the control of clock signals clk100k, clk10m, vpwbus[7:0], and workmdl=0, the power generation control module U5 adjusts the operation of the generator set every 10 minutes according to the detected grid voltage changes, and controls the start or stop of the work through the output bus pwctr[7:0]. After the 8-bit pwctr[7:0] control bus is fully decoded, there are 256 possible combinations and management of the generator set. When workmdl=1, under the control of clk100k, wpstart[7:0], wpend[7:0], and vpwbus[7:0] signals, it is compared with the current time indicated by the system timing bus cnt24, and intelligently determines whether it is in the high-price period of the grid or the low-price period of the grid, and then outputs pwctr[7:0] to control the power generation during the high-price period of the grid and stop the power generation during the low-price period of the grid.

[0040] The pumping control module U6, under the control of the clock signals clk100k, clk10m, vpubus[7:0], and workmdl=0, adjusts the operation of the pumping unit every 10 minutes according to the detected grid voltage changes, and controls the start or stop of the operation through the output bus puctr[7:0]. After the 8-bit puctr[7:0] control bus is fully decoded, there are 256 possible combinations and management of the pumping units. When workmdl=1, under the control of the clk100k, wpstart[7:0], wpend[7:0], and vpubus[7:0] signals, it is compared with the current time indicated by the system timing bus cnt24, and intelligently determines whether it is in the high-price period of the grid or the low-price period of the grid, and then outputs puctr[7:0] to control pumping and energy storage during the low-price period of the grid, and stop pumping during the high-price period of the grid.

[0041] Under the control of the clock signal clk100k and the wsbus[7:0] signal, the wind power / photovoltaic power control module U7 outputs the wind generator working control signal windctr[7:0] and the photovoltaic generator working control signal solarctr[7:0] when rw=1. The windctr[7:0] and solarctr[7:0] buses each have 8 bits of data, and after full decoding, 256 possible combinations and management of the wind power generator / photovoltaic generator group can be generated.

[0042] When the reset signal rst=0, each module in the FPGA chip can be reset. When rst=1, each module in the chip works normally according to the design requirements.

[0043] This embodiment also provides a voltage detection intelligent control peak shaving and valley filling method based on FPGA, which adopts the above-mentioned voltage detection intelligent control peak shaving and valley filling device based on FPGA.

[0044] The peak shaving and valley filling device has two modes, namely the peak shaving and valley filling working mode and the profit working mode. Figure 1 As shown, there is a working mode input signal workmdl outside the peak shaving and valley filling device for controlling the working mode of the system. When workmdl=0, it is in the peak shaving and valley filling working mode; when the working mode setting signal workmdl=1, it is in the profit working mode.

[0045] The method comprises the following steps:

[0046] S1. The grid voltage is stepped down to low voltage by a transformer and then full-wave rectified, and then filtered and divided, so that the output at this time is 2.87V standard DC voltage when the grid voltage is 220V.

[0047] Specifically, the 220V AC voltage input from the power grid is converted into a low-voltage AC signal of 18V to 9V by a transformer, converted into a pulsating DC signal by a bridge full-wave rectifier, filtered to obtain a stable DC signal, and finally a 2.87V stable DC voltage is obtained by a voltage divider resistor and output to the analog / digital conversion module for analog-to-digital conversion; the detection accuracy of the DC voltage to be detected output by the voltage divider resistor is 0.013V, and the corresponding AC detection accuracy is 1V; that is, for every 1V change in the AC voltage of the power grid, the working state of the generator and pumping unit in the pumped-storage power station is adjusted once.

[0048] The AC variation range of the power grid is 249V~190V, corresponding to the DC voltage to be detected before the input analog / digital conversion of 3.237V~2.47V, or the output value after the analog / digital conversion is rev=251~191=221±30, or the corresponding memory address is vramadr=30±30=60~0, thereby realizing the conversion from the AC mains voltage change to the control of the working status of the generator and pump in the pumped storage power station.

[0049] Specifically, when the analog / digital conversion output detection voltage data is rev=251~222, it means that the grid voltage has increased and the load is light, so the load should be increased, that is, pumped storage work should be carried out; when the analog / digital conversion output detection voltage data is rev=220~191, it means that the grid voltage has decreased and the load is heavy, so the power supply should be increased, that is, water discharge and power generation work should be carried out; when the voltage detection output signal rev=221, it means that the grid voltage has reached the standard value, the grid power generation and power consumption have reached a balance, and the generator and pump are not working; there are 30 gears to choose from for grid overvoltage and undervoltage, each memory unit has 8 bits, and 256 combinations of motors can be formed.

[0050] S2. Learn how many generators should be started to generate electricity for every 1V drop in the grid voltage when the system is working in the peak shaving and valley filling mode in order to restore the grid voltage to the normal voltage level of 220V; learn how many pumps should be started to pump water for every 1V increase in the grid voltage in order to restore the grid voltage to the normal voltage level of 220V; learn when to start and stop generators and pumps in the profit mode; learn which wind power generators and photovoltaic generators should be in normal power generation or in shutdown maintenance.

[0051] Specifically, Figure 1 As shown, the voltage detection intelligent peak shaving and valley filling device based on FPGA is provided with an external learning state control signal rw for controlling the working state of the system.

[0052] When the learning state control signal rw=0, the system works in the learning state.

[0053] The learning process is implemented by providing the address of the memory unit vram through the external input learning address signal lrnadr[5:0], and writing the control data of the generator set or pumping unit that should be started or shut down when the grid voltage changes by 1V into the memory unit vram[lrnadr] through the learning data bus lrn[7:0]. This data is an experimental data, which needs to be obtained through testing: when the grid changes by 1V, how many sets of generators or pumping units should be started to balance the grid voltage, which is an empirical data.

[0054] The external input learning address signals lrnadr=62 and lrnadr=63 provide the start and stop working time of the generator and pump of the vram

[62] and vram

[63] units used in the profit mode. lrnadr=61 provides the start and stop control of the wind power generator set and the photovoltaic generator set determined by the vram

[61] unit.

[0055] When the learning state control signal rw=1, the system is in working state. At this time, the operation of the system is controlled by whether the working mode signal workmdl is 0 or 1.

[0056] The FPGA chip acquires the knowledge of the time periods when the generator sets and pumping units in the pumped storage power station should be started or stopped through the initial learning process.

[0057] By building a 64-unit memory in the FPGA chip, each unit has an 8-bit operand, and when fully decoded, there can be up to 256 different combinations for controlling generators and pumping units; the address of the memory is manually set in the learning state, and the control data is input; in the working state, when the system is in the peak shaving and valley filling mode, the address converted from the detected grid voltage is used for addressing, and the data in the address corresponding to the external voltage is read out to control how many generators or pumping units should be started or stopped when the grid is at this voltage;

[0058] When the system is in the peak-shaving and valley-filling mode, when the address is 30, corresponding to the 220V standard grid voltage, the data in this unit is 0, indicating that the generator or pumping unit is not working; when the grid voltage is lower than the 220V standard grid voltage, the address is reduced by 1 from unit 30 for every 1V reduction, and a group of generators are started to work. When the voltage drops to or below 190V, the address is 0, and all generators are started to work; when the grid voltage is higher than the 220V standard grid voltage, the address is increased by 1 from unit 30 for every 1V increase, and a group of pumps are started to work. When the voltage is equal to or higher than 249V, the address is 60, and all pumps are started to work;

[0059] When the system is in profit mode, when the address is 61, bits 7 to 4 of the unit control the number of wind generators working, and bits 3 to 0 of the unit control the number of photovoltaic generators working; when the address is 62, bits 7 to 3 of the unit specify the hours at which the generator starts working in profit mode, working when the second bit is 1 and shutting down when it is 0; when the address is 63, bits 7 to 3 of the unit specify the hours at which the generator ends working in profit mode, working when the second bit is 1 and shutting down when it is 0.

[0060] S3. When in the peak-shaving and valley-filling working mode, perform peak-shaving and valley-filling control: convert the collected grid voltage into the addressing data of the storage unit of the FPGA; control the start and stop status of the generator and pump in the pumped storage power station according to the change range of the grid voltage by reading the corresponding storage unit; specifically, judge the grid voltage load at this time according to the analog-to-digital conversion output value range corresponding to the DC voltage change range to be detected, and control the start and stop status of the generator and pump according to the learning data. The generator and pump do not work at the same time.

[0061] Specifically, in this embodiment, when the system works in the intelligent peak shaving and valley filling working mode, the grid working voltage value is detected once every 10 minutes and the generator and pump of the pumped storage power station are started or stopped. The preset time interval of this embodiment can avoid too frequent start and stop of the generator and pump due to excessive detection.

[0062] When the grid voltage changes in the range of 249V to 221V, the generator control module vpwctr reads 30 control signals from the memory vram address 29 to 0 to control the operation of each generator. When the voltage changes over 249V, the control limit is reached and the control state at 249V is maintained. When the grid voltage changes in the range of 219V to 190V, the pump control module vpuctr reads 30 control signals from the memory vram address 31 to 60 to control the operation of each pump. When the voltage changes below 190V, the control limit is reached and the control state at 190V is maintained. When the grid voltage is the standard 220V, the control signal is read from the vram address 30 unit, and the generator and pump are controlled to not work.

[0063] In the peak-shaving and valley-filling working mode, the generator's power generation and the pump's water pumping are mutually exclusive, that is, when the generator is generating power, the pump stops working; when the pump is pumping water, the generator stops working. This avoids the pumped-storage power station from providing power while consuming power itself.

[0064] S4. When in profit mode, profit control is performed. The start and stop status of the generator and pump are controlled according to the preset time in the learning state. The difference in electricity prices during the day and night is utilized to pump water for energy storage when the electricity price is low, and release water for power generation when the electricity price is high.

[0065] Specifically, in this embodiment, when the system operates in the profit working mode, when the generator working start time stored in the memory unit 62 in the FPGA chip is consistent with the time output by the system timer, the generator starts working; when the generator working stop time stored in the 63 unit is consistent with the time output by the system timer, the generator stops working; in order to avoid the pumped-storage power station pumping and releasing water at the same time, the pump is set to stop working when the generator is working, and the generator is set to stop working when the pump is working.

[0066] In this mode, the start and stop status of each generator and pump is controlled by the preset time in the learning state. The high electricity price period of the day is the period when the generator generates electricity and outputs electricity, and the pump stops working; the low electricity price period of the day is the period when the pump pumps water to store energy, lifting the water from the lower reservoir to the upper reservoir to store the position potential energy of the water, and the generator stops working during this period. The specific working period of the generator and pump is controlled by the generator start working time data wpstart[7:3] output by the memory unit vram

[62] and the generator end working time data wpend[7:3] output by vram

[63] [7:3]. When the bit wpstart[2] and wpend[2] are 0, the generator and pump are not working and are in the equipment maintenance state; when the bit wpstart[2] and wpend[2] are 1, whether the generator and pump work is controlled by the time specified by wpstart[7:3] and wpend[7:3]. wpstart[7:3] sets the start time of the generator and the shutdown time of the pump, and wpend[7:3] sets the stop time of the generator and the working time of the pump.

[0067] Regardless of whether the working mode setting signal workmdl is 0 or 1, the start and stop of the wind power generator set and the photovoltaic generator set are not affected, and are only controlled by the shutdown data set in vram

[61] , so that all power generation equipment can generate as much electricity as possible to improve the utilization rate of power generation equipment. Wind power generation and photovoltaic power generation are not controlled by the system working mode, but only by which units stored in unit 61 are working and which units are stopped for maintenance; among them, the 7-4 bits of unit 61 control the wind power generator set, and the 3-0 bits control the photovoltaic generator set, each with 4 bits, and 16 control combinations can be obtained after full decoding.

[0068] When the system works in profit working mode, the generator starts working when the generator working start time stored in the memory unit 62 in the FPGA chip is consistent with the time output by the system timer; when the generator working stop time stored in the 63 unit is consistent with the time output by the system timer, the generator stops working; in order to avoid the pumped-storage power station pumping and releasing water at the same time, the pump is set to stop working when the generator is working, and the generator is set to stop working when the pump is working.

[0069] The FPGA-based voltage detection intelligent control peak-shaving and valley-filling device of this embodiment is located in a pumped storage power station, and the long / short-range transmission module is a short-range transmission method, such as RS232, RS484 and other wired serial buses, or one or more wireless transmission methods such as WIFI, Zigbee, etc. In this application, a single pumped storage power station can play a certain role in alleviating the peak-shaving and valley-filling of the power grid, and can achieve profits by using the day and night electricity price difference.

[0070] This embodiment provides a voltage detection intelligent control peak-shaving and valley-filling device based on FPGA, and designs a voltage detection intelligent control peak-shaving and valley-filling method. The control method specifically includes the control of the working mode input signal, the control of the learning state input signal, the control of converting the grid voltage into the DC voltage to be measured, the control of converting the DC voltage to be measured into a digital signal, the control of the FPGA chip on the digital signal of the voltage to be measured, and the control of the remote transmission of the control signal, which proposes a specific and feasible solution for how to achieve the balance of power supply and demand in the power grid. When the system works in the peak-shaving and valley-filling mode, it learns in advance how many power generation and power consumption equipment can be started to balance the power supply and load of the power grid, and stores this knowledge in the storage unit of the FPGA chip; when the system works, it converts the detected grid voltage into addressing data, reads the learned knowledge to intelligently remotely control the start or stop of the generator and pump in the pumped storage power station, so as to achieve the purpose of peak-shaving and valley-filling of the power grid, and enables the power generation equipment in the power grid to generate as much as possible, reduce power abandonment, and improve power generation efficiency. When the system works in profit mode, the power price difference between day and night in the power grid is used to control the pumped storage power station to generate electricity or pump water to make profits, and wind power / photovoltaic power can be generated as much as possible to make profits, and it may also be used for pumping. The device and method of this embodiment have practical application value.

[0071] Example 2

[0072] The difference between this embodiment and embodiment 1 is that the voltage detection intelligent control peak-shaving and valley-filling device based on FPGA is set in the urban power grid management center far away from the pumped storage power station. Figure 5 As shown, the control signals pwctr[7:0], puctr[7:0], windctr[7:0], solarctr[7:0] output by the power generation control module U5, the pumping control module U6, and the wind power / photovoltaic power generation control module U7 for the generator, pump, wind power generator / photovoltaic generator group are at a certain distance from the power generation and pumping equipment, and the existing wired / wireless transmission technology is used for data transmission to convey it to the controlled equipment.

[0073] The long / short-range transmission module in this embodiment is a long-distance transmission module, specifically one of the wired transmission communication systems such as optical fiber communication, Internet communication, and power communication, or one of the wireless communication systems such as 4G / 5G and microwave. It can unify the pumped-storage power stations and wind power / photovoltaic power generation groups distributed in various places in a wide area for centralized management, so as to better realize intelligent peak shaving and valley filling to balance the power supply and load of the power grid.

[0074] Obviously, the above embodiments are merely examples for clearly explaining the technical solutions of the present invention, and are not intended to limit the implementation methods of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the protection of the claims of the present invention.

Claims

1. A voltage detection intelligent control peak shaving and valley filling device based on FPGA, characterized in that: The peak shaving and valley filling device comprises a clock control module, a step-down rectifier and filter module, an analog / digital conversion module, an analog / digital conversion control module, a power grid voltage detection module, a voltage transformation control learning module, a hydropower control module, a pumping control module, and a wind power / photovoltaic power control module; the clock control module, the analog / digital conversion control module, the power grid voltage detection module, the voltage transformation control learning module, the wind power / photovoltaic power control module, the hydropower control module, and the pumping control module are integrated in an FPGA chip; the FPGA chip is connected to an external clock; The step-down rectification and filtering module steps down the grid voltage to a low voltage and performs full-wave rectification, and then performs filtering and voltage division to obtain a DC voltage signal to be measured; The analog / digital conversion module converts the DC voltage signal to be measured into a digital signal; the analog / digital conversion control module provides a chip selection signal and a conversion driving external clock for the analog / digital conversion module; and also converts the digital signal serially output by the analog / digital conversion module into a parallel output signal, which is used as input data for the grid voltage detection module; The grid voltage detection module converts the detected grid voltage data into an addressing address of a memory in the voltage transformation control learning module; The transformer control learning module is used to learn the control data of the amount of power generation equipment that needs to be started when the grid voltage drops to balance the increased load of the grid, the control data of the amount of pumped storage equipment that needs to be started when the grid voltage rises to balance the reduced load of the grid, how to select wind power generators and photovoltaic generators that are in normal power generation state and / or in shutdown maintenance state, and the start and stop time of generators and pumps when in profit mode; The learning data is stored in the storage unit of the FPGA, and the hydropower control module, wind power / photovoltaic power control module, and pumping control module are controlled according to the learning content to realize hydropower generation control, wind power generation control, photovoltaic power generation control, and pumping control respectively; The step-down rectification and filtering module comprises a transformer, a bridge rectifier circuit, a π-type filter circuit and a variable resistor voltage divider unit connected in sequence, and the voltage across the variable resistor voltage divider unit is used as a DC voltage to be measured; The peak shaving and valley filling device also includes a long / short-range transmission module; the hydropower control module, wind power / photovoltaic power control module, and pumping control module respectively realize hydropower generation control, wind power generation control, photovoltaic power generation control, and pumping machine control through the long / short-range transmission module; when the peak shaving and valley filling device is arranged in a pumped-storage power station, the control signals for starting and stopping the generator and the pump are transmitted in a short-range transmission manner; when the peak shaving and valley filling device is arranged in an urban power grid management center far away from the pumped-storage power station, the control signals for starting and stopping the generator and the pump are transmitted in a long-range transmission manner.

2. The voltage detection intelligent control peak shaving and valley filling device based on FPGA according to claim 1 is characterized in that: The analog / digital conversion module is a TLC549 chip.

3. A voltage detection intelligent control peak shaving and valley filling method based on FPGA, characterized in that: The FPGA-based voltage detection intelligent control peak shaving and valley filling device according to any one of claims 1 to 2 is applied, and the method comprises the steps of: S1. The grid voltage is stepped down to a low voltage and full-wave rectified, and then filtered and divided, so that the output at this time is a preset standard DC voltage when the grid voltage is 220V; S2. Learn the control data of the amount of power generation equipment that needs to be started when the grid voltage drops to balance the increased load of the grid, and learn the control data of the amount of pumped storage equipment that needs to be started when the grid voltage rises to balance the reduced load of the grid; learn how to select wind power generators and photovoltaic generators that are in normal power generation state and / or in shutdown maintenance state; Learn the start and stop times of generators and pumps when in profit mode; Store the learning data in the storage unit of FPGA; S3. Perform peak-shaving and valley-filling control: collect the grid voltage at preset time intervals; convert the collected grid voltage into addressing data of the storage unit of the FPGA; control the start and stop status of the generator and pump in the pumped-storage power station according to the range of the grid voltage by reading the corresponding storage unit; and the generator and pump do not work at the same time; S4. Control the profit model: Use the learning data written in the storage unit of FPGA to control the pumped storage power station to pump water and store energy when the learned grid price is low, and release water to generate electricity at the specified time when the learned grid price is high; the wind power / photovoltaic motor power generation period is not restricted by time, and the generated electricity can be used for both electricity sales and pumping. The number of equipment started and stopped is controlled by the learning data, which is convenient for shutting down the motor when repairing the equipment.

4. The voltage detection intelligent control peak shaving and valley filling method based on FPGA according to claim 3 is characterized in that: The method for obtaining the preset standard DC voltage in step S1 specifically includes: the 220V AC voltage input from the power grid is converted into a low-voltage AC signal of 18V to 9V by a transformer, converted into a pulsating DC signal by a bridge full-wave rectifier, and then filtered to obtain a stable DC signal, and finally a 2.87V stable DC voltage is obtained by a voltage divider resistor and output to an analog / digital conversion module for analog-to-digital conversion; the detection accuracy of the DC voltage to be detected output by the voltage divider resistor is 0.013V, and the corresponding AC detection accuracy is 1V.

5. The voltage detection intelligent control peak shaving and valley filling method based on FPGA according to claim 3 is characterized in that: The method for learning the start and stop time of the generator and the pump in step S2 when in the profit mode specifically includes: By building multiple memory units in the FPGA chip, each unit has a multi-bit operand, and there are multiple different combinations for controlling the generator and pumping unit when fully decoded; the address of the memory is manually set in the learning state, and the control data is input; When the peak-shaving and valley-filling control is performed, the address converted from the detected grid voltage is used for addressing, and the data in the address corresponding to the external voltage is read out to control how many generators or pumping units should be started or stopped when the grid is at that voltage; When the peak shaving and valley filling control is performed, the data of the unit in the storage address corresponding to the 220V grid standard voltage is 0, indicating that the generator or the pumping unit is not working; when the grid voltage is lower than the 220V grid standard voltage, the address is reduced by 1 from the unit in the storage address corresponding to the 220V grid standard voltage for each 1V decrease, and a group of generators are started to work, so that when the voltage drops to equal to or lower than the preset minimum voltage, the address is 0, and all generators are started to work; when the grid voltage is higher than the 220V grid standard voltage, the address is increased by 1 from the unit in the storage address corresponding to the 220V grid standard voltage for each 1V increase, and a group of pumps are started to work, and when the address number reaches the address corresponding to the preset maximum voltage, all pumps are started to work; When profit control is performed, the number of wind generators controlled and the number of photovoltaic generators controlled are stored in an address unit that is 1 more than the address corresponding to the preset maximum voltage; the number of hours when the generator starts working when the profit mode is specified is stored in the next address unit, and the number of hours when the generator ends working when the profit module is stored in the address unit that is 1 more than the address corresponding to the preset maximum voltage.

6. The voltage detection intelligent control peak shaving and valley filling method based on FPGA according to claim 3 is characterized in that: The grid voltage collected in step S3 passes through step S1, outputs a digital signal through analog / digital conversion, and then converts it into a memory address; the grid voltage load at this time is judged according to the analog-to-digital conversion output value range corresponding to the DC voltage variation range to be detected, and the start and stop status of the generator and the pump are controlled according to the learning data.

7. The voltage detection intelligent control peak shaving and valley filling method based on FPGA according to claim 6 is characterized in that: When the grid voltage changes in the range of 249V to 190V, the corresponding analog / digital conversion value is rev=251~191, and the corresponding memory address is vramadr=30±30=60~0; when the analog / digital conversion output detection voltage data is rev=251~222, it means that the grid voltage increases and the load is light, and the load should be increased, that is, pumped storage work is carried out; when the analog / digital conversion output detection voltage data is rev=220~191, it means that the grid voltage decreases and the load is heavy, and the power supply should be increased, that is, water discharge and power generation work is carried out; when the voltage detection output signal rev=221, it means that the grid voltage reaches the standard value, the grid power generation and power consumption are balanced, and the generator and pump are not working; there are 30 gears to choose from for grid overvoltage and undervoltage, and each memory unit has 8 bits, which can form 256 combinations of motors.

8. The voltage detection intelligent control peak shaving and valley filling method based on FPGA according to any one of claims 3 to 7, characterized in that: The learning of the transformer control learning module is achieved by providing the address of the memory unit through an external input learning address signal, and writing the control data of the generator set or pumping unit that should be started or shut down when the grid voltage changes by 1V into the memory unit through the learning data bus.

Images