CN210839049U

CN210839049U - Intelligent photovoltaic power supply system adaptive to extreme illumination conditions

Info

Publication number: CN210839049U
Application number: CN201922016160.6U
Authority: CN
Inventors: 麦文; 秦爽
Original assignee: Sichuan Normal University
Current assignee: Sichuan Normal University
Priority date: 2019-11-20
Filing date: 2019-11-20
Publication date: 2020-06-23
Anticipated expiration: 2029-11-20

Abstract

The utility model discloses an intelligent photovoltaic electrical power generating system who adapts to extreme illumination condition, including photovoltaic cell module, power output and the communication interface module, battery module, low-power consumption controller module, modBus communication module, the variable current battery module of charging, the variable voltage super capacitor module of charging, low-power consumption switching power supply module, current-voltage detection and the switch module that has the temperature to detect the function. The system integrates the functions of battery charging management, standby protection under long-time rainy weather conditions, electric energy collection under extremely-weak illumination environment and communication. The power supply problem in the current application of the Internet of things is well solved. Especially, outdoor application can solve the problem of large-quantity, low-power and low-cost power supply deployment under the condition of no special power supply line. The system has the characteristics of maintenance free, convenient installation and long-time and high reliability work, and the later maintenance cost is saved to the greatest extent.

Description

Intelligent photovoltaic power supply system adaptive to extreme illumination conditions

Technical Field

The utility model relates to an intelligence power supply unit field especially can collect the electric energy in the extreme light environment, adapts to the intelligent photovoltaic electrical power generating system of extreme illumination condition work such as long-time overcast and rainy weather.

Background

At present, the application scale of the Internet of things is rapidly increased, and the market demand of the Internet of things is huge. The internet of things equipment generally has the characteristics of low power consumption, low cost, networking capability and the like. In a plurality of application scenes of the internet of things, the outdoor internet of things system has important application value and is an important part in the application of the internet of things. However, many outdoor internet of things devices are difficult to supply power by using a wired power supply because the outdoor internet of things devices do not have environmental conditions. Such as on hills, in lakes, in agricultural fields, on remote highways, etc. In these environments, although it is possible to select a power generation system such as wind power, water power, etc. to supply power, it is limited that the local environment generally does not have widely applied conditions. Solar energy is the most desirable solution as a clean energy source. The solar power supply is generally not affected by regions, and can be used in places irradiated by sunlight. The internet of things equipment is low in power consumption, the required solar power supply is usually enough in a few watts, and therefore the solar power supply is very high in market value.

However, the solar power supply has uncertainty, the solar illumination intensity is different in different regional environments, even if the weather changes in the same region, the influence on the solar power supply is huge, and the influence on the solar power supply is also huge due to the changes of day and night in the same place within one day. At present, the maximum photovoltaic power generation efficiency is about 20%, the efficiency is low, and full utilization is very necessary. The maximum power output of the photovoltaic cell is greatly influenced by the illumination intensity and the temperature no matter the photovoltaic cell is made of a monocrystalline silicon material or a polycrystalline silicon material, so that the solar energy is fully utilized and a special control system is required. For the internet of things equipment, the main dilemma of the photovoltaic power supply is as follows: how to achieve ultra-low power consumption and miniaturization? How to do low-cost scalable applications? How to deal with extreme weather conditions, such as prolonged overcast and rainy weather? Without solving these problems, the practical application of photovoltaic power sources is of no significance. The utility model provides a new implementation scheme has characteristics of miniwatt, low cost, miniaturization, high efficiency, ability overlength standby to the thing networking power, uses innovative circuit and software algorithm to solve this a series of problems, and this scheme is really also fit for more powerful photovoltaic power realization certainly.

A lot of research has been conducted by researchers at home and abroad on photovoltaic power systems, and some of the research have been conducted on an optimal mathematical model for characteristics of a photovoltaic cell, some have been conducted on a power tracking technique for maximum output power of a photovoltaic cell, some have been conducted on a switching power supply conversion method for photovoltaic cell adaptation for optimal conversion efficiency, and the like. These studies are currently being carried out in a very large number of ways. However, these studies have a drawback that the overall solution is not discussed for specific application scenarios. They pursue the best solution to a certain problem, neglecting the relevance of system implementation, individual optimality is not the overall optimality. For example, in discussing a DC-DC converter in a photovoltaic power supply, researchers usually only focus on the implementation of a mathematical model and an algorithm of the converter, and the research is completed with certain achievements. These results are not actually suitable for use in the internet of things. Because researchers only pursue methods to realize the problems of neglecting the cost, volume, power consumption and the like of the used equipment. Some researchers have discussed the problem of realizing the maximum power point MPPT of the photovoltaic cell by using an analog circuit method, and although the method is low in cost and very effective, the researchers neglect that the discreteness of analog devices is too large, and the individual success does not represent the success of mass production, and neglect the fact that the equipment of the Internet of things is applied in a large scale. Other researchers mainly research various algorithms aiming at the MPPT algorithm, even introduce many intelligent algorithms, although the algorithms are good, the algorithms neglect the high real-time performance required by the algorithms and require a high-performance processing platform to be realized, and the problems brought by the algorithms are that the cost, the power consumption and the system volume are increased, and the algorithms are not suitable for being used in the Internet of things. In addition, the power supply of the internet of things is not only used for supplying electric energy, but also used for long-time unattended equipment, and self-monitoring, self-maintenance, timely communication and the like are needed. For example, in a continuous rainy weather for a long time, the photovoltaic cell actually cannot provide enough electric energy to ensure that the system works normally, how can the energy storage battery pack achieve ultra-low power consumption standby at this time? How does the system enter self-protection, wait for the opportunity to recover again? How is the last warning message sent? How to fully utilize the conditions of overcast and rainy and weak light for power generation? And the like, the problems need to be comprehensively solved to really make the power supply equipment of the internet of things have practical significance.

From this point of view, thing networking power is an intelligent power supply unit in nature. It must adapt to environmental changes, especially to the great drop change of illumination intensity, and utilize light energy to generate electricity to the maximum extent. It can also make sleep protection under extreme conditions, and it can wake up and recover itself under timely conditions. The power supply of the internet of things solves all the problems by never solving a certain MPPT algorithm. For these reasons, most of the current research results cannot be directly applied to the internet of things in practice. No matter how advanced and powerful an outdoor internet of things device is, if an adaptive power supply is not available, the maintenance cost and the labor consumption of the outdoor internet of things device are huge.

SUMMERY OF THE UTILITY MODEL

The utility model aims to solve the technical problem that an intelligent photovoltaic electrical power generating system who adapts to extreme illumination condition is provided. The system integrates a storage battery management function, an electric energy collection function in an extremely weak illumination environment, a sleep protection function in a long-time rainy weather condition and a network communication function. The problem of power supply bottleneck in the current application of the Internet of things is well solved. Especially, outdoor application can solve the problem of large-quantity, low-power and low-cost power supply deployment under the condition of no special power supply line. The system has the characteristics of maintenance free, convenient installation and long-time and high reliability work, and the later maintenance cost is saved to the greatest extent.

The utility model adopts the technical proposal that:

an intelligent photovoltaic power system adapting to extreme illumination conditions comprises a photovoltaic cell module with a temperature sensor, a power supply output and communication interface module, a variable-current storage battery charging module, a low-power-consumption processor module, a ModBus communication module, a storage battery charging module, a variable-voltage super capacitor charging module, a low-power-consumption voltage reduction switch power module and a current-voltage detection and switch control module;

the photovoltaic cell module with a temperature sensor: converting the light energy into electric energy which is used as electric energy input and can detect the working temperature of the photovoltaic cell;

the power supply output and communication interface module: providing power output and photoelectric isolation type RS485 serial port communication, and inputting and outputting alarm information and control information of the photovoltaic power supply through the port;

the battery module is characterized in that: under the condition of sufficient illumination, the electric energy is stored in the storage battery, and under the condition of insufficient illumination, the storage battery discharges electricity for the load to use;

the low-power processor module: the low-power-consumption single-chip microcomputer and the low-static-current switching power supply are adopted, and a Buck voltage reduction converter is adopted to provide power for the single-chip microcomputer;

the ModBus communication module: the ModBus RTU communication protocol is used for communicating with peripheral equipment and providing power output, and the power output and the communication input and output form a bus structure capable of providing electric energy;

the variable-current storage battery charging module comprises: the battery charging system is realized by a Buck voltage reduction type battery charging integrated circuit U5, a U5 is provided with a maximum power point tracking MPPT control end, and the control end is connected with a single chip microcomputer U1 of a low-power-consumption processor module through an IO _ BUS 2; the single chip microcomputer U1 controls the MPPT port to realize the control of the magnitude of the charging current; the variable-current storage battery charging module is based on a switch type charging integrated circuit with an external resistor and an adjustable working current mode, and comprises a storage battery negative temperature coefficient NTC detection resistor R10, an energy storage inductor L2, a charging current detection resistor RS3 and a charging current adjusting resistor Rx; single or a plurality of digital potentiometers U10 are adopted to be cascaded to replace a current adjusting resistor Rx, so that variable charging current control is realized;

the variable voltage super capacitor charging module: the Buck-Boost type switching power supply consisting of a single chip microcomputer U1 and a U6 is realized, and an MPPT port of U6 is connected with the single chip microcomputer U1 in the low-power-consumption processor module through an IO _ BUS 1; the single chip microcomputer U1 controls the single chip microcomputer U6 unit to output different voltages through an IO _ BUS 1; the variable-voltage super-capacitor charging module consists of a Buck-Boost Buck-Boost conversion circuit, and comprises an energy storage inductor L3, a Boost capacitor C8 and a Boost capacitor C9, wherein 4 MOS (metal oxide semiconductor) tubes are arranged in internal logic to drive the energy storage inductor L3; also comprises an output feedback resistor R13,A resistor R16; an operational amplifier U9 is introduced between the voltage fed back by the resistors R13 and R16 and the integrated circuit U6, so that the resistor R13, the resistor R16, the resistor R19, the resistor R20 and the operational amplifier U9 form a subtraction operator; the single chip microcomputer is used for controlling the DAC to output different voltages, the operational amplifier U10 is used for level conversion, the DAC voltage is converted into suitable voltage for driving the U9 reverse end, and therefore the output voltage V of the Buck-Boost converter is changed_outThe output voltage is controllable;

the low-power step-down switching power supply module: the device comprises a U7, wherein a power supply input is provided by a solar cell, and an output voltage is provided for U3 and U4; u3 and U4 are both low-voltage low-power input current detection amplifiers;

the current and voltage detection and switch control module: the method comprises current detection, voltage detection and switch control; the photoelectric cell output current measuring circuit is provided with a photoelectric cell output current detection resistor RS1, a photoelectric cell output current measuring circuit is formed by a resistor RS1 and a low-voltage low-power-consumption input current detection amplifier U3, and the output of U3 is connected with an analog-to-digital interface AD1 of a singlechip U1; the device is provided with a series voltage-dividing type photocell voltage detector, and consists of a resistor R7 and a resistor R6, wherein the voltage-dividing output is connected with an AD2 of a singlechip U1; the current measurement and the voltage measurement formed by the AD1 and the AD2 can detect the output power of the photocell; an output current detection resistor RS2 is arranged, an output current measurement circuit is formed by the resistor RS2 and a low-voltage low-power-consumption output current detection amplifier U4, and the output of U4 is connected with the AD5 of the singlechip U1; the voltage divider circuit is arranged in series, consists of a resistor R2 and a resistor R3 and is used for detecting the magnitude of output voltage; the AD5 and the AD6 form output voltage and output current measurement, and can measure the output power of the power supply.

Furthermore, an MOS tube can be used for replacing a charging current adjusting resistor Rx, the method is that a resistor between a source S and a drain D of the MOS tube is used as the current adjusting resistor Rx, different voltages are applied to a grid G of the MOS tube, and the change of the on-resistance between the source S and the drain D of the MOS tube is realized; the variable voltage is applied in a mode that the singlechip generates a PWM wave, and the PWM voltage wave generated by the singlechip is realized by a low-pass filter; when the duty ratio of the PWM signal output by the singlechip is changed, the voltage output by the low-pass filter is changed along with the change, so that the control of the grid voltage of the MOS tube is realized.

Furthermore, a digital potentiometer can be used for replacing DAC output, the input voltage of the U10 equidirectional end is replaced by digital potentiometer partial voltage output, and variable voltage output can also be realized.

Compared with the prior art, the beneficial effects of the utility model are that:

1) the ultra-long standby characteristic is achieved, and the problem that the battery is easily damaged when the power supply of the photovoltaic cell is insufficient in the case of unmanned maintenance when the power supply of the Internet of things encounters rainy days for a long time is solved. The utility model discloses only consume 2 hundred microamperes electric current or even lower under the standby condition, the battery of 1-2 ampere hours can realize the overlength standby of several months. The system can cope with extreme overcast and rainy weather, can automatically recover to work normally under the condition of weather recovery, and is not available for various types of photovoltaic power supplies in the current market.

2) The system has the advantages of low cost, low power consumption, high efficiency and miniaturization, and meets the application requirements of the Internet of things. The photovoltaic power generation system is realized by innovating and improving a conventional special integrated circuit, and double optimization of low cost and high efficiency is achieved. It is more cost-effective than using expensive specialized photovoltaic power integrated circuits. The current special photovoltaic power supply integrated chip is mainly a foreign product, is expensive in price and single in function, and does not have the management function and the communication function of a power supply system. The system is realized by using domestic devices, the cost is one tenth of that of the foreign devices, and the system is combined with a low-power-consumption processing system, so that the functions are powerful and perfect. Compared with the DSP or ARM proposal proposed by a plurality of researchers, the method has the advantages of simplicity, easy realization, low cost, miniaturization and extremely low power consumption, and is more suitable for a battery power supply system.

3) An MPPT tracking algorithm based on state memory is provided and realized. The algorithm is combined with a specially designed circuit, the MPPT tracking problem under the conditions of strong light irradiation, weak light irradiation and extremely weak light irradiation is solved, solar energy collection and power generation under multiple conditions are realized, and the utilization rate of a photovoltaic cell is improved. The scheme also overcomes the defect that the photovoltaic cell on the current market can only generate electricity under the condition of strong light. Compared with the MPPT tracking algorithm proposed by numerous researchers, the MPPT tracking algorithm is simpler, does not need high calculation performance, and solves the problem of quick response of a low-speed and low-power-consumption processing system.

4) The intelligent degree is high, and the general conditions that a conventional photovoltaic power generation system is not suitable for application of the Internet of things, such as no communication networking, no self-monitoring and the like, are solved. The power supply system has the functions of charge management, self detection, self recovery, extreme condition protection, communication alarm and the like, and realizes high intellectualization. The intelligent charging power supply integrates intelligent management and intelligent charging, and can meet the application requirements in various complex scenes.

Drawings

Fig. 1 is a block diagram of the overall composition of a photovoltaic power supply.

Fig. 2 is a photovoltaic power supply implementation schematic.

Fig. 3 is a general schematic diagram of a battery charge management integrated circuit.

Fig. 4 is a circuit diagram of a modified variable power battery charging circuit.

Fig. 5 is a general working principle diagram of the Buck-Boost converter.

Fig. 6 is a schematic diagram of an implementation of the adjustable output voltage Buck-Boost converter.

FIG. 7 is a graph of photovoltaic cell I-V characteristics under different lighting conditions.

FIG. 8 is a graph of photovoltaic cell P-V characteristics under different illumination conditions.

Figure 9 is a MPPT implementation flow diagram.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

As shown in fig. 1 and 2, the power supply system is composed of 9 parts, each enclosed by a box in fig. 2. Each unit is detailed below:

i unit (photovoltaic cell module): a photovoltaic cell having a temperature detection function. Q1 is a temperature sensor for detecting the operating temperature of the photovoltaic cell.

Unit II (power supply output and communication interface module): a power supply and communication interface; the port provides power output and serial port communication, and the photovoltaic power supply alarm information and the control information are input and output through the port.

Cell III (battery module): energy storage battery of photovoltaic power supply. R10 is a negative temperature coefficient resistance for detecting the battery temperature. Under the condition of sufficient illumination, the photovoltaic power supply stores electric energy into a storage battery; under the condition of insufficient illumination intensity, the storage battery discharges for the load to use.

IV unit (low power consumption control module): a low power consumption controller and a power supply part. The photovoltaic power supply control part is a control core of the whole photovoltaic power supply and is composed of a low-power-consumption single chip microcomputer and a low-power-consumption switching power supply. The part has extremely low current consumption during normal operation, and the current consumption is usually less than 200uA, which is the key for realizing the ultra-long standby. The DC-DC-1 is a Buck Buck converter and provides power for the single chip microcomputer, and the quiescent current of the power converter is extremely low and is usually less than 50 uA. U1 is an ultra-low power consumption singlechip, and its current is below 200uA when normal work. The input power supply of the DC-DC-1 can be a photocell or a storage battery, and the core single chip microcomputer U1 is always kept to work uninterruptedly. In the dormant state, the low-power controller further reduces the working frequency and the power consumption.

V unit (ModBus communication module): a photoelectrically isolated ModBus communication unit is used. The part communicates with the peripheral equipment by using a ModBus-RTU communication protocol and provides power output. The isolation module U8 is connected with the U1 of the IV unit through serial port lines Tx and Rx handshake, so as to realize the transmission of information.

VI unit (variable current battery charging module): the part is realized by a Buck Buck battery charging integrated circuit U5, a U5 has a maximum power point tracking MPPT control end, and the port is connected with a U1 single chip microcomputer of an IV unit through an IO _ BUS 2. The single chip microcomputer can control the MPPT port to realize the control of the magnitude of the charging current. The variable power charging implementation principle is shown in fig. 3 and 4, and the implementation principle will be explained below.

VII unit (variable voltage super capacitor charging module): and a variable voltage super capacitor charging unit. The unit is realized by a U6Buck-Boost voltage-type switching power supply, and a U6 has an MPPT control port and is connected with a U1 single chip microcomputer in an IV unit through an IO _ BUS 1. The single chip microcomputer can control the U6 unit to output different voltages through the IO _ BUS 1. The output voltage of U6 charges a series super capacitor composed of S1-Sn through an isolation diode D4. The function of D4 is to prevent the super capacitor current from flowing backward into U6, and D4 is also the current limiting resistor. The diode D4 has internal resistance when conducting in forward direction, and the charging current of the whole super capacitor can be changed by adjusting the output voltage of the U6. The variable power super capacitor charging implementation principle is shown in fig. 5 and 6, and the implementation principle will be explained below.

VIII unit (low power switching power supply module): the low-static-current Buck voltage-reducing switching power supply is formed. The part is mainly composed of U7 and related elements, and the quiescent current is usually required to be lower than 50 uA. The input of the partial power supply is provided by a solar battery, and the output voltage is provided to U3 and U4. U3 and U4 are low voltage low power current sense amplifiers. The DC-DC-2 composed of U7 can ensure the normal work of U3 and U4 only by extremely low current output, and the current is less than 500 uA.

IX unit (detection and switch control module): this portion is the portion remaining after each unit division in fig. 2. The part is mainly current detection, voltage detection, switch control and the like. The resistor RS1 is used for detecting the output current of the photocell, the RS1 and the U3 form a photocell output current measuring circuit, and the output of the U3 is connected with the U1 analog-to-digital interface AD 1. The voltage of the series voltage division type photocell is detected by the R7 and the R6, and the magnitude of the divided voltage is detected by the AD2 of U1. The AD1 and the AD2 form input current and voltage measurement, and the output power of the photocell can be measured. The resistor RS2 is used for detecting output current, the RS2 and the U4 form an output current measuring circuit, and the output of the U4 is connected with the AD5 of the U1. R2 and R3 are series voltage dividing circuits, and the divided voltage is detected by the AD6 of U1, and is used for detecting the output voltage. The AD5 and the AD6 form output voltage and output current measurement, and the power output can be measured. Further, R14 and R17 constitute a supercapacitor voltage measuring circuit, and are detected by the AD3 of U1. R11 and R12 constitute a battery voltage measuring circuit, and are detected by the AD4 of U1. K1, K2, K3, K4, and K5 are electronic switches for switching the operating state. D1 and D3 form a photovoltaic cell protection circuit, and D1 prevents current from reversely entering the photovoltaic cell to damage the photovoltaic cell. D3 is a transient absorption diode that prevents the input from being damaged by static electricity. The D2 forms a photocell and storage battery automatic switching circuit, and ensures that a switching power supply DC-DC-1 composed of U2 and the like is supplied with power under any state, and the U1 is never stopped.

Charging circuit principle of current-variable storage battery

This part is the key to achieving maximum power transfer of the photovoltaic cell. As previously mentioned, the present invention is implemented using a universal battery charging integrated circuit modification. Different from the scheme provided by a common researcher, the common researcher uses an ARM processor or a DSP with high power to realize the purpose, the main purpose is to verify the control principle, and the result is not suitable for being directly used in the power supply of the Internet of things. The utility model discloses use general battery charging integrated circuit to realize, but general integrated circuit output is all fixed, can not adjust the output current size that charges. However, the universal battery charging integrated circuit has high cost performance, and the conversion efficiency, the cost and the static power consumption are all excellent. One class of switch-mode charging integrated circuits is designed in an external adjustable current mode, and a user can adjust the charging current through an external resistor. The operating principle of such a charge management integrated circuit can be described by the structure shown in fig. 3. In fig. 3, R10 is a battery temperature detection resistor, L2 is a storage inductor of the switching circuit, RS3 is a charging current detection resistor, and Rx is a charging current adjustment resistor. The preset threshold current is biased by the preset Rx inside the integrated circuit, and then the MOS tube driven by the driving logic is controlled to realize the change of the turn-on time, so that the adjustment of the charging current is realized. At present, such integrated circuits have many models, and domestic chip companies adopt the scheme. The domestic integrated circuit has very low price and excellent performance, and completely meets the design requirements.

In view of the operation principle described in fig. 3, the programming of the charging current can be realized by only designing a variable resistance circuit. The utility model discloses in consider low-cost and higher control accuracy, use the digital potentiometer to realize. Improved circuit as shown in fig. 4, the digital potentiometer is a multi-tap circuit, and very high control precision can be realized by using single or a plurality of digital potentiometer cascades. In fig. 4, U10 is a substitute for Rx in fig. 3. Through improvement, the singlechip can control a digital potentiometer tap to realize variable resistance control, and further realize charging current control. The common digital potentiometer has 512-level control, very high control precision can be realized by 2-3 cascades, the control can realize charging current control of 10 milliamperes to several amperes, and the MPPT tracking has high enough precision.

1. In addition, other ways to implement such variable resistance control may be used instead of Rx. For example, the MOS transistor itself has a resistance characteristic, and the on-resistance between the source S and the drain D of the MOS transistor can be changed by applying different voltages to the gate G of the MOS transistor. The variable voltage can be applied by generating a PWM wave (Pulse width modulated wave) by the single chip, and passing the PWM wave through a low pass filter, such as an RC low pass filter.

The voltage output by the PWM wave duty ratio low-pass filter of the control singlechip is changed along with the change of the voltage of the grid G of the MOS tube, so that the on-resistance between the source S and the drain D of the MOS tube is changed, and the control effect similar to that of a digital potentiometer is achieved.

The variable current charging circuit is used for charging the storage battery under the condition of sufficient illumination. The variable current charging circuit is a part for realizing MPPT, and the variable current charging circuit also needs to be formed together with a singlechip system of the IV unit and an input voltage and current detection circuit of the IX unit, and can be realized by matching with a tracking algorithm. The MPPT implementation is set forth in software design.

Charging circuit principle of variable-voltage super capacitor

A variable voltage super capacitor charging circuit is used for collecting electric energy under the condition that the illumination of a photovoltaic cell is insufficient and even weak. The part of the circuit greatly improves the use efficiency of the photovoltaic cell. The traditional photovoltaic cell mainly works under the condition of sufficient illumination, and the electric energy generated under the condition of insufficient illumination cannot be utilized. The main working principle of the part is to store energy by using a super capacitor. Under the condition of insufficient illumination, the output voltage of the photovoltaic cell is very low, the current is very small, and the variable-current charging system of the VI unit cannot work normally. However, the Buck-Boost conversion circuit in the VII can store electric energy output by the photovoltaic cell into the super capacitor, and when enough electric energy is stored, the stored energy of the super capacitor is switched to the VI unit to convert the electric energy into the storage battery, so that the energy collection work under the condition of insufficient illumination is completed.

As shown in fig. 5, the general principle of the Buck-Boost conversion circuit is shown. The structure L3 is an energy storage inductor, C8 and C9 are Boost capacitors, and 4 MOS transistors in internal logic drive L3. R13 and R16 are output feedback resistors. When the input voltage is higher than the set output voltage, the internal logic drives 4 MOS tubes to perform voltage reduction conversion, and the stable output voltage is unchanged; when the input voltage is lower than the set output voltage, the internal logic drive 4 only performs the boost conversion on the MOS, thereby stabilizing the output voltage. The output voltage is determined by R13, R16, and is estimated as Vout — Vref (R13+ R16)/R16 where Vref is the chip internal reference voltage. The conventional Buck-Boost Buck-Boost conversion integrated circuit is fixed in output voltage and cannot realize variable voltage output.

Most of the Buck-Boost Buck-Boost conversion integrated circuits in the market adopt the working principle shown in FIG. 5. In order to achieve the effect that the output voltage can be arbitrarily adjusted to realize the MPPT function, a modified circuit is shown in fig. 6. Fig. 6 is a modification of fig. 5, in which an operational amplifier U9 is introduced between the feedback voltages of R13 and R16 and the integrated circuit U6. R13, R16, R19, R20, and U9 actually constitute a single subtractor. If the U10 output voltage Vdac selects R13-R19-R and R16-R20-Rf, the basic principle of adjustable output voltage is that the U9 output voltage VF (Rf/R) × (Vout-Vdac) ═ Vref. Here Vref is the U6 chip internal reference voltage. The expression is further arranged to obtain: vout is Vref (R/Rf) + Vdac, which is the modified Buck-Boost converter output voltage expression. Likewise, where Vref is the U6 chip internal reference voltage, Vref is typically within 1V. Selecting R, Rf the appropriate resistance value completes the variable voltage range control. The single chip microcomputer is used for controlling the digital-to-analog converter DAC to output different voltages, and the U10 operational amplifier amplifies the voltages and outputs the amplified voltages to the U9 reverse end. Thereby changing the output voltage Vout of the Buck-Boost converter. According to the expression that Vout is Vref (R/Rf) + Vdac, the higher the voltage output by the DAC, the higher the output voltage of the Buck-Boost switching power supply, and the purpose of controlling the output voltage is realized.

Fig. 6 shows only the implementation of the variable voltage supercapacitor charge circuit, which is only a part of the MPPT function. To realize the MPPT function, an input current detection unit composed of a single chip microcomputer U1, RS1 and U3, an input voltage detection unit composed of R7 and R6, and an MPPT algorithm are required.

Another way to achieve a variable voltage is to use a digital potentiometer. In fig. 5, R16 is replaced with a digital potentiometer to realize a variable voltage output. However, the operating frequency of the switching power supply is generally high, while the operating frequency of the digital potentiometer is low, generally not exceeding 100KHz, so that the scheme is not suitable for being adopted. It is also feasible if high-speed digital potentiometers are used, but this adds no small cost. The high-speed digital potentiometer is expensive, the controllable number of taps under the current technical condition is usually not more than 64, and the precise control is difficult to realize, so the high-speed digital potentiometer is not suitable for use. While the variable resistor Rx function of fig. 4 can use a digital potentiometer because the operating frequency of fig. 4 is very low, which is a dc circuit. But the DAC part in fig. 6 can be implemented by using a digital potentiometer, and the voltage output by the tap of the digital potentiometer is used to replace the DAC output, so that the cost can be reduced by adopting the scheme.

Photovoltaic Power Process Explanation

The structure composition of the photovoltaic power supply is explained, the optimal work can be completed only by controlling corresponding parts to change in real time by the singlechip according to different state conditions, and the optimal work can be divided into a plurality of control strategies and processes under different illumination conditions.

1) Working process under sufficient illumination

Sufficient illumination means that the photovoltaic cell can reach a rated output power state. At this time, the output voltage of the photocell is the highest, and the output current is the largest. The single chip microcomputer control strategy is to directly provide part of the electric energy generated by the photocell for the load to work, so that the efficiency is highest. Any more conversion reduces the efficiency of the use of electrical energy. And the redundant electric energy generated by the photovoltaic cell is completely converted and stored into the storage battery. Through detecting the output power of the photocell, the charging current of the storage battery is adjusted in real time to ensure that the photocell works at the maximum output power point, namely the MPPT tracking function is realized.

The specific control operation is as follows: in fig. 2, a U1 single chip microcomputer controls electronic switches K1, K2 and K3 to be switched on, and K4 and K5 to be switched off. At this time, U5 operates normally and the battery is in a charged state. The voltage + V-SUN transmitted by the photocell is actually far higher than the voltage of the storage battery, so that the D2 diode is reversely biased, the current and the voltage on the load are directly provided by the photocell, the storage battery is isolated from the output, and the storage battery is in a large-current charging state. The single chip microcomputer calculates the output power of the photocell through data acquired by AD1 and AD2, and controls the charging current of the U5 through IO _ BUS2, so that the photocell is always in the maximum power output state, and MPPT tracking is completed.

2) Working process under insufficient illumination

The underillumination state means that the photovoltaic cell is not in an optimal state, the output power cannot reach a rated value, but the load operation can be supported, and redundant electric energy can be output. At the moment, the singlechip microcomputer control strategy is to start a low-current charging mode of the storage battery and store redundant electric energy. If the output of the photocell just meets the load requirement, the storage battery is closed to charge, and the electric energy generated by the photocell is directly supplied to the load for use.

The specific control operation is as follows: in fig. 2, a U1 singlechip controls electronic switches K1 and K2 to be switched on, and K4 and K5 to be switched off. The single chip microcomputer calculates the output power of the photocell through data acquired by AD1 and AD2 to judge the position of the maximum power point. If the output current needs to be increased, K3 is started to make the accumulator in low current charging state and control the photocell in maximum power output state. The K3 is turned off if the photovoltaic cell has been brought to near its maximum power point by the power supplied directly to the load, allowing the power from the photovoltaic cell to be supplied directly to the load.

3) Working process under extremely weak illumination state

The extremely weak illumination state refers to a state that the photocell has output, but is weak and cannot directly drive a load or drive a charging circuit of the storage battery. The control strategy at this time is: and the storage battery charging circuit is closed, and the load directly uses the storage battery for power supply. And starting the super capacitor energy storage circuit of the VII unit. The single chip microcomputer controls the output voltage of a Buck-Boost converter of the U6 to enable the voltage on the super capacitor to gradually increase. When enough electric energy is available, the storage battery charging circuit is started, and the electric energy of the super capacitor is quickly transferred to the storage battery. Then, the storage battery charging circuit is closed again, and the energy storage process of the second period is started. The Buck-Boost converter can work in a voltage boosting and reducing mode, so that the working voltage range is wide, and weak electric energy generated by the photovoltaic cell can be fully utilized. The single chip can control the voltage difference between the output voltage of the U6 and the super capacitor, the isolation diode D4 has internal resistance, the dual functions of circuit isolation and charging current limitation can be achieved, the charging power can be controlled by controlling the charging voltage difference on the D4, and therefore the MPPT tracking function of the photocell under the condition of extremely weak light is achieved.

The specific control is as follows: the singlechip controls K2 and K3 to be closed, and K1 and K4 to be opened. The single chip microcomputer detects the voltage on the super capacitor through the AD3, and detects the output power of the optical cell through the AD1 and the AD2 to find a maximum power output point. And the voltage on the super capacitor continuously rises in the charging process, and when the voltage reaches a preset voltage, K5 is started to transfer the electric energy of the super capacitor to the storage battery. Then K5 is turned off to control U6 to enter the next super capacitor electrical energy storage cycle.

4) Working process in non-illumination state

A non-illuminated state is a state where the photovoltaic cell has no output or only a little weak output and cannot be effectively utilized. The control strategy at this time is to close the charging loop to minimize the ineffective power consumption. The method comprises two stages: firstly, when the storage battery has enough electric energy, the storage battery continuously provides electric energy for the load; and secondly, when the electric energy of the storage battery is reduced to a preset value, the load power supply is turned off to stop the load, all invalid electric energy consumption is turned off, the controller switches the working frequency to a lower mode, and the whole power supply enters an extremely low power consumption dormant state. The system enters a process of waiting for resuscitation. The design is to protect the battery pack from damage caused by over-discharge. The design considers that the storage battery can still last for weeks or months, so that the storage battery can be used outdoors even when encountering severe weather. In addition, when the photocell is damaged and no electric energy is output, the battery pack can wait for maintenance for a long time, and the battery pack is not completely damaged.

The specific control is as follows: the singlechip controls K2, K3, K4 and K5 to be closed. When the electric energy of the storage battery is sufficient, K1 is started to ensure that the load can work normally. When the situation that the electric energy of the storage battery is insufficient is detected, the K1 is closed to cut off the load power supply, and meanwhile, the single chip microcomputer closes the communication function of the U8 unit through the IO6, so that the electric energy is saved to the maximum extent. At the moment, the storage battery power supply is only supplied to the single chip microcomputer to work, other circuits do not obtain electric energy any more, and the system is in a low power consumption state. The singlechip unit IV is an ultra-low power consumption system, so that the battery pack can be maintained for a long period of working time. As long as there is a normal power supply state in this period, the system can be recovered again, and the reliability of the Internet of things equipment is improved.

5) Information communication working process

The traditional power supply usually works independently, and the state of the power supply cannot be known. The power supply of the internet of things is also an internet of things device in nature, and under the unattended condition, the working state of the power supply needs to be known. In the design, a ModBus-RTU communication mode is used for transmitting various parameters of the power supply to a remote end. The sensor and other equipment can be networked by using ModBus communication. In the scheme, the power supply is a ModBus slave device, and the remote master device returns the parameters when inquiring. The master device can realize the state inquiry of a plurality of power supplies through a network, so that the fault can be found in time or the process tracking can be realized. U8 in FIG. 2 is a communication circuit for realizing photoelectric isolation, and the harm caused by outdoor strong interference can be eliminated by using differential lines. Such as lightning, industrial impulse electromagnetic interference, etc.

MPPT maximum power point tracking method based on state memory

The MPPT algorithm is an effective method for improving the utilization efficiency of the photovoltaic cell, and various mature methods such as a disturbance observation method, a conductance increment method and the like exist at present, and in addition, various improved algorithms exist. A common feature can be found by summarizing the conditions under which these algorithms are applied. They are all generic algorithms that are considered based on the unknown state of the photovoltaic cell. In fact, the internet of things power supply has certain specificity. The power supply of the internet of things needs to integrate functions of intelligent management, networking communication and the like. This is completely different from the pure MPPT algorithm design. There is also one of the most important reasons that the power supply of the internet of things must use a low power consumption scheme. Generally, low power consumption and high performance are contradictory, a single chip computer system cannot perform high-performance calculation processing in a low power consumption state, and the mature algorithms cannot be used, so that the system can solve the problem only by adopting the algorithm suitable for low power consumption.

The fact that the general application scene of the internet of things equipment is inspected can be found, and the internet of things equipment is usually fixed in a certain place for long-term use. Such as farm electricity, forest, hillside, etc. In the same place, sunlight, temperature and the like have certain periodicity, and a basis for fast MPPT tracking is objectively provided. Although the utility model discloses used the low-power consumption singlechip, singlechip operating speed is slow, but can utilize the inside accumulator of singlechip, get off the tracking speed who promotes the algorithm with photocell maximum power point state memory, also can accomplish quick response. Both theoretical and experimental studies indicate that the maximum power point of the photovoltaic cell is mainly influenced by illumination intensity and temperature, and a state memory algorithm can be realized if the two parameters can be obtained. The utility model discloses a MPPT algorithm thought is just with the temperature range segmentation of photocell work, realizes the illumination intensity segmentation that the photocell received again. The state of the photovoltaic cell falls within a two-dimensional spatial lattice formed by the interweaving of the two segmented regions. As long as the data corresponding to each two-dimensional space lattice point is recorded as the optimal power point parameter in the state, the position of the optimal power point can be quickly found by using the state recording table after the environment is changed. The memory algorithm has rapidity, and the adjusting speed of the memory algorithm is greatly higher than the tracking speed realized by the conventional algorithm. However, the method has certain defects, and is mainly limited by the capacity of the storage space of the singlechip. The state data partition can not achieve any precision, and can only achieve a fuzzy area range. However, the maximum power point value range can be greatly shortened by the state memory parameters, and the system can be adjusted in real time, so that the rapid tracking can be realized. Currently, a single-chip memory generally has several hundred KB, and can be divided into thousands or tens of thousands of state partitions, so that the single-chip memory has high enough precision.

One problem is that the first time the system tracks the MPPT location, there is no optimal parameter information on the memory, how to look up the table? The system mainly depends on the automatic calculation function to find the maximum power point, and then writes the optimal parameters into the system partition. After the system traverses various states, the function of large-range memory tracking can be automatically realized. As mentioned above, the same region has a certain periodicity regardless of the illumination variation range or the temperature variation range, which is why the algorithm is effective. The MPPT algorithm implementation principle based on state memory is described as follows, and can be divided into 2 steps. Firstly, planning available storage space, and secondly, implementing a corresponding MPPT tracking algorithm according to the current detected state.

Storage space planning and parameter acquisition

Taking 64KB as an example, the storage space available to a single chip computer is partitioned into two-dimensional lookup tables. The general single chip microcomputer has available space above 64KB, and the 64K address uses 16bit addressing. The temperature range is divided into 8bit intervals, and then the range of the illumination intensity value is divided into 8bit intervals, as shown in table 1.

Table 1 lookup table data space partitioning

The abscissa value is a temperature expression range, the ordinate is an illumination expression range, the abscissa and the ordinate form a 16-bit address number, and the 16-bit address number is the address of a 64KB storage space. Each address can store 8-bit data, and the 8-bit data is the optimal parameter found when the range works. For example, the temperature range is quantized using a 10bit number, then taking the upper 8 bits realizes the temperature zone division. If the illumination intensity range is quantized using 12-bit data, then taking the upper 8 bits realizes illumination value area division. These two 8-bit numbers may constitute a 16-bit address addressing space. If the addressing address has an offset, then the use of any offset range can be realized by adding the offset address amount when the table is looked up. The data stored in each address in table 1 is the best parameter found by a certain MPPT search. The 16-bit addressing space allows 65536 optimal value storage, which means that even a conventional single chip can achieve very accurate zone control.

Obtaining an illumination intensity parameter: a large number of theories and experiments show that under the condition of the same temperature and different illumination intensities, the no-load voltage of the photovoltaic cell is different. A typical 18V rated output photovoltaic cell I-V characteristic is shown in figure 7. As can be seen from fig. 7, the no-load voltage of the photovoltaic cell is completely different under different illumination conditions, so that the no-load voltage can be used as the illumination intensity information.

On the other hand, theories and experiments show that the no-load voltage of the photovoltaic cell is mainly concentrated in a small range. In fig. 7, the intersections of the curves with the horizontal axis, which are mainly at the right end of the horizontal axis, exhibit a pronounced non-uniform distribution characteristic. The illumination variation range in the figure is from 1000W/m²To 30W/m²In the change, the no-load voltage only changes from 22v to 16v, and the change range is not large. Then the range can be focused on during quantization, and the data range is compressed, thereby improving the quantization precision. The whole principle of the utility model is shown in figure 1, and K2, K3 and K4 are cut off, and the photocell is cut off from the load. The main load of the photocell is a current and voltage detection circuit. Because the detection circuit has large impedance, a more accurate no-load voltage measurement value can be obtained.

Acquiring temperature parameters: the temperature parameter refers to the temperature of the photovoltaic cell. A group of temperature sensors are arranged on the photovoltaic cell, Q1 in figure 2 is the temperature sensor, and the low-cost realization method is to use a temperature sensitive resistor. The temperature-sensitive resistor has the advantages of low cost, simple structure and good repeatability. In the system, the absolute accuracy of the temperature is not required, but the relative accuracy of the temperature is required, so that the system can be used as long as the temperature is repeated with enough accuracy. Therefore, a temperature sensitive resistor, which is a low cost device, is sufficient for temperature detection. If the single chip microcomputer has a larger memory space, the temperature sensor with better performance can be used for realizing the temperature sensor.

MPPT maximum power point tracking method

In the MPPT tracking algorithm, no matter a storage battery is used for charging or a super capacitor is used for charging, the control and detection modes are the same, so that the algorithm implementation difficulty is simplified, but two conditions are respectively explained below.

In the first case, the look-up table data has been established. The system has undergone one MPPT tracking, obtaining the optimum parameters, which have approached the optimum operating point in the current state. Thus, starting from this position, the system only needs to be properly fine-tuned to enter the optimal state.

The single chip microcomputer adjusts the charging state of the system according to the minimum step length, when the maximum power output is detected, the state is that delta P/delta V is 0, namely the position of a triangle on the curve in the graph in fig. 8, and the delta P/delta V is the slope of the P-V curve. The minimum step length refers to the minimum adjustment range in the system, the system adopts a digital potentiometer to adjust the charging current, and the minimum step length is 1 scale adjustment. Under the condition, the system tracking process has stability, is not easy to generate large-range oscillation, and is relatively quick. Both theory and experiment show that the photovoltaic cell has a unique maximum output power point under specific conditions, and the characteristics of the photovoltaic cell are similar to the curve shown in fig. 8, which is an important basis for the implementation of the algorithm. The P-V diagram of the photovoltaic cell shown in fig. 8 corresponds to the I-V diagram shown in fig. 7, and is a diagram of the relationship between the output power and the output voltage of the photovoltaic cell under the same temperature and different illumination intensities.

In the second case, no look-up table data has been established. The system has not yet built the best state look-up table shown in table 1. At this time, the system needs to approach the maximum power point step by step from a starting point. It can be seen from fig. 8 that each curve represents the power-voltage variation curve at different illumination intensities. It is obvious that gradually increasing the output power from a point to the right can track to the maximum power point position. Compared with the curve starting from a certain point on the left side, the curve starting from the right side has larger change of the steep curve delta P/delta V, and the detection data are easier to judge. Yet another reason is when the illuminated area of the photovoltaic cell is partially obscured by a small area, such as a shade of tree. The P-V diagram of the photovoltaic cell then appears bimodal or multimodal. However, the change trend of the P-V curve is unchanged due to only a small part of the shielding, and the main peak position is still at the rightmost position. Therefore, tracking from the right can also avoid finding the wrong peak point. The algorithmic process in the second case is described below.

Setting the charging state to a certain low current state, and detecting the output voltage V of the photocell₁And current I₁Recording the output power P₁. Then, a step control is added to detect the output voltage V of the photocell₂And current I₂Recording the output power P₂. Calculating Δ P ═ P₂-P₁And Δ V ═ V₂-V₁Further calculate the P-V curve at V₁、 V₂The average slope K of this segment is Δ P/Δ V.

And selecting a proportionality coefficient M, calculating a product value of M and K, and taking the value of M and K as a positive integer, wherein T is assumed. The system takes T as a charging step control value. The singlechip adjusts the step length every time according to the calculated T value, and determines the adjusting direction according to the slope K. The relative position of the current working point and the maximum power point can be determined according to the slope K, the power adjustment direction is further determined, and the system working point can be automatically pulled back even if entering the left side of the maximum working point.

According to the above process, when the current working point is far away from the maximum working point, the curve K value is large, so the adjustment step length is also large. When the maximum power point is approached, the K value tends to 0, the step length is automatically reduced, so that the precise tracking can be realized, and the step length changing mode can improve the tracking speed and precision. And a proper M value is selected, the system can realize the step length changing process in the previous stage, and the system can fall in the 1 step length adjusting range in the final stage, so that the system oscillation is avoided.

The maximum power point judgment is that K is 0, which cannot be found exactly in practice, but between two adjacent control steps, the K value swings between positive and negative, and whether the peak value point is reached can be determined according to the change of the K value. After the peak point is reached, the parameters are recorded, and the current parameters are written into the corresponding positions in the table 1 by the single chip microcomputer, so that one-time tracking is completed. The complete MPPT tracking procedure is described as shown in fig. 9.

The flow in fig. 9 is divided into two loop paths, a flow with look-up table data and a flow without look-up table data. If the system enters a small step state, i.e. a process with a lookup table, the system will find the maximum power point in the small step state. When updating the parameters of the lookup table, whether the parameter change is larger or not is confirmed, if the parameter change exceeds the expected range, the parameter is updated, and if not, the parameter change is not updated. The processing can reduce the times of writing the memory, prolong the service life and update the influence caused by slow aging of the system in time. If the system has no lookup table data, namely the system is in a state which has not been traversed, the system enters a variable step tracking state, and after finding the optimal power point, the data is written into a memory and stored. The system enters the current state detection no matter which state is entered, and finally returns, so that the quick response to the influence of the environmental sudden change can be realized. The current state detection is also the switching detection of the whole system entering other states, such as a sleep maintenance state, another charging state and the like, the MPPT algorithm flow is mainly highlighted in fig. 9, and the power system management part is simplified.

Claims

1. An intelligent photovoltaic power system adaptive to extreme illumination conditions is characterized by comprising a photovoltaic cell module with a temperature sensor, a power supply output and communication interface module, a storage battery module, a low-power-consumption processor module, a ModBus communication module, a variable-current storage battery charging module, a variable-voltage super-capacitor charging module, a low-power-consumption voltage reduction switch power module and a current-voltage detection and switch control module;

the variable voltage super capacitor charging module: the Buck-Boost type switching power supply consisting of a single chip microcomputer U1 and a U6 is realized, and an MPPT port of U6 is connected with the single chip microcomputer U1 in the low-power-consumption processor module through an IO _ BUS 1; the single chip microcomputer U1 controls the single chip microcomputer U6 unit to output different voltages through an IO _ BUS 1; the variable-voltage super-capacitor charging module consists of a Buck-Boost Buck-Boost conversion circuit, and comprises an energy storage inductor L3, a Boost capacitor C8 and a Boost capacitor C9, wherein 4 MOS (metal oxide semiconductor) tubes are arranged in internal logic to drive the energy storage inductor L3; the circuit also comprises an output feedback resistor R13 and a resistor R16; an operational amplifier U9 is introduced between the voltage fed back by the resistors R13 and R16 and the integrated circuit U6, so that the resistor R13, the resistor R16, the resistor R19, the resistor R20 and the operational amplifier U9 form a subtraction operator; the single chip microcomputer is used for controlling the DAC to output different voltages, the operational amplifier U10 is used for level conversion, the DAC voltage is converted into suitable voltage for driving the U9 reverse end, and therefore the output voltage V of the Buck-Boost converter is changed_outThe output voltage is controllable;

2. The intelligent photovoltaic power system adapting to the extreme illumination condition of claim 1, wherein a MOS transistor is further adopted to replace the charging current adjusting resistor Rx, by using a resistor between the source S and the drain D of the MOS transistor as the current adjusting resistor Rx, and applying different voltages to the gate G of the MOS transistor, the change of the on-resistance between the source S and the drain D of the MOS transistor is realized; the variable voltage is applied in a mode that the singlechip generates a PWM wave, and the PWM voltage wave generated by the singlechip is realized by a low-pass filter; when the duty ratio of the PWM signal output by the singlechip is changed, the voltage output by the low-pass filter is changed along with the change, so that the control of the grid voltage of the MOS tube is realized.

3. The intelligent photovoltaic power system adapting to the extreme lighting conditions, according to claim 1, wherein a digital potentiometer is further adopted to replace the DAC output, and the input voltage of the U10 common-direction end is replaced by a digital potentiometer divided output, so that a variable voltage output can be realized.