Energy storage battery automatic charging and discharging system and method based on load tracking
Technical Field
The invention relates to the technical field of load tracking, in particular to an energy storage battery automatic charging and discharging system and method based on load tracking.
Background
At present, domestic power and electricity demand is increasingly large, however, domestic power demand management is still in a relatively low-level stage, in order to achieve scientific power utilization, ordered power utilization and electricity saving and improve the efficiency of power and electricity utilization so as to achieve the purposes of energy conservation and environmental protection, a national power department introduces a step electricity price strategy and adjusts the power utilization behavior of power consumers by adopting peak-valley electricity prices. The direct current power supply system consists of an alternating current input part, a switching power supply system, a storage battery and a load, wherein the storage battery is used as a backup battery, and when the alternating current input is power-off, the storage battery can ensure the normal operation of the load. Because the power supply condition of the power grid in China is good, the alternating current is very stable, the storage battery works in a floating charge state for a long time, and a part of electric power is wasted. On the premise of ensuring reliable power supply of a load, in order to fully exert the value of a storage battery, a plurality of power consumers, battery manufacturers and power equipment manufacturers develop a peak clipping and valley filling energy storage control system based on a switching power supply system, and the method mainly changes the original floating and energy consumption mode of the storage battery into a peak clipping and valley filling circulating operation mode by indirectly controlling the operation mode of the switching power supply through a newly added peak clipping and valley filling controller or adding a peak clipping and valley filling control function in a controller in the switching power supply system, so that the use value of the battery is improved, the power resource is reasonably utilized, and the economic benefit is brought. The conventional storage battery and power grid connected power supply lacks a mode of tracking the power of a load, the charging and discharging power of the storage battery cannot be fed back and adjusted according to the power consumption of the load, the storage battery is overcharged or overdischarged, the service life of the storage battery is damaged, and meanwhile, the storage battery and power grid connected power supply is switched without power interruption, and the service life of the load is reduced due to excessive instantaneous voltage change times.
Disclosure of Invention
The invention solves the problems of over-discharge or over-charge of grid-connected power supply of a storage battery and reduction of the service life of a load caused by switching power supply in the prior art, and provides an energy storage battery automatic charge-discharge system and method based on load tracking. And in the charging process of the energy storage system, the difference value of the power which can be provided according to the energy storage items and the power required by the load can be used for charging the energy storage system. The power required by the load is monitored through a tracking algorithm in the discharging process of the energy storage system, and the energy storage system can provide corresponding power to meet the requirement of load work, so that the energy utilization rate of the energy storage system is improved.
In order to realize the purpose, the following technical scheme is provided:
an energy storage battery automatic charging and discharging system based on load tracking comprises:
the mains supply input end is electrically connected with the energy storage converter PCS and the power supply switching device, and is used for charging the energy storage unit BAT through the energy storage converter PCS and supplying power to a load through the power supply switching device;
the matrix digital energy management control system DMMES is electrically connected with the main control module, the energy storage converter PCS and the energy storage unit BAT and is used for receiving control signals of the main control module, controlling the input and output power of the energy storage converter PCS and controlling the charging and discharging of the energy storage unit BAT;
the main control module is electrically connected with a signal acquisition end, the signal acquisition end is used for acquiring the real-time power of a load, the main control module is used for acquiring electricity price information, and generating a control signal to a matrix digital energy management control system DMMES according to the electricity price information and the real-time power of the load;
the energy storage unit BAT is used for storing electric energy and supplying power to a load;
and the power supply switching device is used for switching the power supply mode under the condition that the load is not powered off.
According to the electricity price information, when the electricity price is low, the commercial power is used for supplying power to the load, meanwhile, the storage battery is charged according to the electric quantity of the storage battery, the storage battery is prevented from being overcharged, the storage battery refers to the energy storage unit BAT, when the electricity price is high, the storage battery is used for supplying power to the load, the storage battery is prevented from being overdischarged according to the electric quantity information of the storage battery, the charging power and the discharging power are dynamically adjusted according to the load tracking algorithm in the discharging process of the storage battery charging, the electricity consumption is always smaller than the electric quantity required by the load, the backflow phenomenon is avoided, the maximum power is guaranteed to be dynamically output, and the energy storage system is guaranteed to obtain the maximum benefit. The invention also provides a power supply switching device for switching the power supply mode of the load without power failure.
Preferably, the power supply switching device comprises a mains supply access branch, an energy storage converter PCS access branch, a main power supply path and a power supply bypass, the mains supply access branch and the energy storage converter PCS access branch are connected in parallel with one end of the main power supply path, the other end of the main power supply path is connected with a load, one end of the power supply bypass is connected in parallel with the mains supply access branch and the energy storage converter PCS access branch, the other end of the power supply bypass is connected with the load, only one of the mains supply access branch and the energy storage converter PCS access branch is switched on, and the power supply bypass is used for storing electric energy and supplying power to the load when power supply is switched.
Preferably, the mains supply access branch comprises a first relay K1 and a commutation diode D1, the energy storage converter PCS access branch comprises a second relay K2 and a backflow prevention diode D2, the main power supply circuit comprises a voltage stabilizing diode D3, the power supply bypass comprises a capacitor, one end of the first relay K1 is connected with the mains supply input end, the other end of the first relay K1 is connected with one end of the commutation diode D1, the other end of the commutation diode D1, one end of the voltage stabilizing diode D3, one end of the backflow prevention diode D2 and one end of the capacitor are connected in parallel, the other end of the backflow prevention diode D2 is connected with one end of the second relay K2, the other end of the second relay K2 is connected with the energy storage converter PCS, the other end of the voltage stabilizing diode D3 and the other end of the capacitor are connected in parallel to a load, and the first relay K1 and the second relay K2 are electrically connected with the main control module.
The energy storage battery automatic charging and discharging method based on load tracking adopts the energy storage battery automatic charging and discharging system based on load tracking, and specifically comprises the following steps:
s1, acquiring a real-time electricity price a, judging whether the real-time electricity price a is smaller than or equal to a first threshold value b, if not, entering S2, if so, supplying power to a load through a mains supply input end, judging whether an energy storage unit BAT is larger than or equal to c% of the total capacity, if so, not operating, otherwise, charging the energy storage unit BAT through the mains supply input end, and obtaining the input power of an energy storage converter PCS based on a load tracking algorithm until the electric energy stored in the energy storage unit BAT reaches d% of the total capacity;
s2, judging whether the electric energy stored by the energy storage unit BAT is larger than or equal to e% of the total capacity, if so, supplying power to a load by the energy storage unit BAT, obtaining the output power of the energy storage converter PCS based on a load tracking algorithm, and if not, entering S3;
s3, judging whether the real-time electricity price a is smaller than or equal to a second threshold value f, if so, supplying power to the load through a mains supply input end, and if not, entering S4;
s4, judging whether the electric energy stored by the energy storage unit BAT is larger than or equal to g% of the total capacity, if so, supplying power to the load by the energy storage unit BAT, obtaining the output power of the energy storage converter PCS based on a load tracking algorithm, and if not, supplying power to the load by a mains supply input end;
wherein the numerical relationship is as follows: b is less than f, c% is less than d%, e% is less than c%, and g% is less than e%.
Preferably, the process of obtaining the input power of the energy storage converter PCS based on the load tracking algorithm is as follows:
within a certain time period T, judging whether the average power P1 of the load is less than or equal to the power P3 of the commercial power input end, if not, not operating; and if so, calculating the charging power of the energy storage unit BAT to be P2= P3-P1, and adjusting the input power of the energy storage converter PCS to be P2.
Preferably, the P2 further comprises an overload adjusting step, specifically comprising the following steps:
sa, acquiring real-time power of a load at certain time intervals within a certain time period T, grouping the acquired real-time power according to a time sequence, and taking all real-time power before the current acquisition time as historical real-time power;
sb, calculating a real-time power change value according to the historical real-time power;
taking a group of the most recent real-time power data before the acquisition moment to perform mean value calculation to obtain each group of variation values E [ X (n) ]:
E[X(n)]=X(n)-An
wherein: x (n) is the nth real-time power data of the group, and An is the average value of the group of data; an is calculated as:
sc, judging whether the fluctuation value of the real-time power change value is smaller than a set allowable fluctuation threshold value, if so, returning to Sc, and if not, performing Sd;
and Sd, adjusting the input power P2 of the energy storage converter PCS by using the mean value of the historical real-time power, and returning to Sa.
Preferably, the Sc specifically comprises the following steps:
calculating the variance S of a group of the latest real-time power data before the acquisition time, judging whether the variance S is greater than a variance threshold, if so, returning to the Sa, and if not, judging whether the fluctuation value of the real-time power change value is less than a set allowable fluctuation threshold, namely, judging whether | Emax-Emin | is less than L, wherein: emax is the maximum variation value, emin is the minimum variation value, and L is the allowable fluctuation threshold; if yes, the process is not carried out, the Sa is returned, and if not, the Sd is carried out.
Preferably, the input power P2 of the energy storage converter PCS is adjusted by using the average value An of the historical real-time power, and then the input power P2 of the energy storage converter PCS is adjusted by returning to Sa, and the formula is as follows:
P2=KAn+B
wherein: k is a conversion coefficient, B is a deviation value, and K and B are obtained by solving by using historical data as a training set and using P2+ P1 not more than P3 as an objective function through a BP neural network algorithm.
Preferably, the process of obtaining the output power of the energy storage converter PCS based on the load tracking algorithm is as follows:
and acquiring the average power P4 of the load in a set period T before the moment of the switching time node T1, and calculating the output power P5= P4-X of the energy storage converter PCS, wherein X >0,X is a backflow prevention constant.
The beneficial effects of the invention are: according to the electricity price information, when the electricity price is low, the commercial power is used for supplying power to the load, meanwhile, the storage battery is charged according to the electric quantity of the storage battery, the storage battery is prevented from being overcharged, the storage battery refers to the energy storage unit BAT, when the electricity price is high, the storage battery is used for supplying power to the load, the storage battery is prevented from being overdischarged according to the electric quantity information of the storage battery, the charging power and the discharging power are dynamically adjusted according to the load tracking algorithm in the discharging process of the storage battery charging, the electricity consumption is always smaller than the electric quantity required by the load, the backflow phenomenon is avoided, the maximum power is guaranteed to be dynamically output, and the energy storage system is guaranteed to obtain the maximum benefit. The invention also provides a power supply switching device for switching the power supply mode of the load without power failure.
Drawings
FIG. 1 is a system configuration diagram of the embodiment;
FIG. 2 is a configuration diagram of a power supply switching device of the embodiment;
fig. 3 is a flow chart of a method of an embodiment.
Detailed Description
Example (b):
this embodiment provides an energy storage battery automatic charging and discharging system based on load tracking, refer to fig. 1, including: the mains supply input end is electrically connected with the energy storage converter PCS and the power supply switching device, and is used for charging the energy storage unit BAT through the energy storage converter PCS and supplying power to a load through the power supply switching device; the matrix digital energy management control system DMMES is electrically connected with the main control module, the energy storage converter PCS and the energy storage unit BAT and used for receiving control signals of the main control module, controlling the input and output power of the energy storage converter PCS and controlling the charging and discharging of the energy storage unit BAT; the main control module is electrically connected with a signal acquisition end, the signal acquisition end is used for acquiring real-time power of a load, the main control module is used for acquiring electricity price information, and generating a control signal to the matrix type digital energy management control system DMMES according to the electricity price information and the real-time power of the load; the energy storage unit BAT is used for storing electric energy and supplying power to a load; and the power supply switching device is used for switching the power supply mode under the condition that the load is not powered off. The power supply switching device comprises a mains supply access branch, an energy storage converter PCS access branch, a main power supply path and a power supply bypass, the mains supply access branch and the energy storage converter PCS access branch are connected in parallel with one end of the main power supply path, the other end of the main power supply path is connected with a load, one end of the power supply bypass is connected with the mains supply access branch and the energy storage converter PCS access branch in parallel, the other end of the power supply bypass is connected with the load, only one of the mains supply access branch and the energy storage converter PCS access branch is connected with the power supply bypass, the power supply bypass is used for storing electric energy, and the power supply bypass supplies power for the load when power supply is switched. Referring to fig. 2, the commercial power access branch includes first relay K1 and commutation diode D1, energy storage converter PCS access branch includes second relay K2 and prevents flowing backward diode D2, the main power supply way includes zener diode D3, the power supply bypass includes electric capacity, first relay K1's one end is connected with mains input, first relay K1's the other end is connected with commutation diode D1's one end, commutation diode D1's the other end, zener diode D3's one end, prevent flowing backward diode D2's one end and the one end of electric capacity and connect in parallel, prevent flowing backward diode D2's the other end and be connected with second relay K2's one end, second relay K2's the other end is connected with energy storage converter PCS, zener diode D3's the other end and the other end of electric capacity connect in parallel in the load, first relay K1 and second relay K2 are connected with the main control module electricity.
The embodiment also provides an energy storage battery automatic charging and discharging method based on load tracking, and the energy storage battery automatic charging and discharging system based on load tracking, referring to fig. 2, specifically includes the following steps:
s1, acquiring a real-time electricity price a, judging whether the real-time electricity price a is smaller than or equal to a first threshold value b, if not, entering S2, if so, supplying power to a load through a mains supply input end, judging whether an energy storage unit BAT is larger than or equal to c% of the total capacity, if so, not operating, otherwise, charging the energy storage unit BAT through the mains supply input end, and obtaining the input power of an energy storage converter PCS based on a load tracking algorithm until the electric energy stored in the energy storage unit BAT reaches d% of the total capacity;
s2, judging whether the electric energy stored by the energy storage unit BAT is larger than or equal to e% of the total capacity, if so, supplying power to a load by the energy storage unit BAT, obtaining the output power of the energy storage converter PCS based on a load tracking algorithm, and if not, entering S3;
the process of obtaining the input power of the energy storage converter PCS based on the load tracking algorithm is as follows:
within a certain time period T, judging whether the average power P1 of the load is less than or equal to the power P3 of the commercial power input end, if not, not operating; and if so, calculating the charging power of the energy storage unit BAT to be P2= P3-P1, and adjusting the input power of the energy storage converter PCS to be P2.
P2 further comprises an overload adjustment step, specifically comprising the steps of:
sa, acquiring real-time power of a load at certain time intervals within a certain time period T, grouping the acquired real-time power according to a time sequence, and taking all real-time power before the current acquisition time as historical real-time power;
sb, calculating a real-time power change value according to historical real-time power;
taking a group of real-time power data which is the nearest before the acquisition moment to perform mean value calculation to obtain each group of change values E [ X (n) ]:
E[X(n)]=X(n)-An
wherein: x (n) is the nth real-time power data of the group, and An is the mean value of the group of data; an is calculated as:
sc, judging whether the fluctuation value of the real-time power change value is smaller than a set allowable fluctuation threshold value, if so, returning to Sc, and if not, performing Sd;
the Sc specifically comprises the following steps:
calculating the variance S of a group of the nearest real-time power data before the acquisition time, judging whether the variance S is greater than a variance threshold value, if so, returning to the Sa, and if not, judging whether the fluctuation value of the real-time power change value is less than a set allowable fluctuation threshold value, namely, judging whether | Emax-Emin | is less than L, wherein: emax is the maximum variation value, emin is the minimum variation value, and L is the allowable fluctuation threshold; if yes, the process is not carried out, the Sa is returned, and if not, the Sd is carried out.
And Sd, adjusting the input power P2 of the energy storage converter PCS by using the mean value of the historical real-time power, and returning to Sa. And adjusting the input power P2 of the energy storage converter PCS by using the average value An of the historical real-time power, returning to Sa, and adjusting the input power P2 of the energy storage converter PCS according to the following formula:
P2=KAn+B
wherein: k is a conversion coefficient, B is an offset value, and K and B are obtained by using historical data as a training set and solving by using a BP neural network algorithm and using P2+ P1-P3 as a target output u (K). The method comprises the following steps:
1) Initializing the structure of a selected BP neural network controller, selecting the number i of nodes of an input layer, the number j of nodes of a hidden layer and the number n of nodes of an output layer, giving random values of weight coefficients v [ i ] [ j ] of the hidden layer and w [ j ] [ n ] of the output layer within the range of-1 to 1, and selecting a learning rate a and a smoothing factor b;
2) Sampling to obtain estimated power r (k) and actual output y (k), and calculating an error e (k) = r (k) -y (k);
3) Normalizing the estimated power r (k), the actual output y (k), the target output u (k) and the error e (k) to be used as the input of a neural network;
4) Calculating input values and output values HO2[ n ] of neurons in each layer of the BP neural network controller, wherein the output values HO2[ n ] are reversely normalized to be K and B, and sending target output u (K) to a controlled object to participate in control and calculation;
5) Through parameter comparison of an input value and an output value, a learning rate a and weight coefficients v [ i ] [ j ] and w [ j ] [ n ] are adjusted by a steepest descent method, the weight coefficients of the BP neural network are corrected by the steepest descent method, the adjustment direction is the negative direction of E (k), an inertia quantity which enables search to accelerate convergence of a global minimum value is added, and for optimization of the steepest descent method, the adjustment formula is as follows:
wherein, P is the step length of one adjustment, gamma 1 and gamma 2 are the variation coefficients of the learning rate and the weight coefficient variation respectively, and theta is a threshold value; when e (k) is greater than e (k-1), 0 < gamma 1 < 1, gamma 2=1; when Δ e (k) is less than Δ e (k-1), γ 1 is greater than 1, and when | e (k) | is greater than θ, γ 2 is greater than 1;
adjusting the step length P by adjusting the value of gamma, and adjusting the weight coefficients v [ i ] [ j ] and w [ j ] [ n ] by adjusting the value of gamma and the adjusted step length P to enable the steepest descent method function curve to be close to the global optimal solution;
6) And (3) sending the weight coefficient w [ j ] [ n ] of the optimized output layer, the weight coefficient v [ i ] [ j ] of the hidden layer and the learning rate a to the BP neural network controller, wherein 2 output values of the BP neural network controller output layer are K and B.
S3, judging whether the real-time electricity price a is smaller than or equal to a second threshold value f, if so, supplying power to the load through a mains supply input end, and if not, entering S4;
s4, judging whether the electric energy stored by the energy storage unit BAT is larger than or equal to g% of the total capacity, if so, supplying power to a load by the energy storage unit BAT, obtaining the output power of the energy storage converter PCS based on a load tracking algorithm, and if not, supplying power to the load by a mains supply input end; the process of obtaining the output power of the energy storage converter PCS based on the load tracking algorithm is as follows:
and acquiring the average power P4 of the load in a set period T before the moment of the switching time node T1, and calculating the output power P5= P4-X of the energy storage converter PCS, wherein X >0,X is a backflow prevention constant. The power consumption is always less than the power required by the load, the backflow phenomenon is avoided, the maximum power of dynamic output is ensured, and the maximum benefit of the energy storage system is ensured.
Wherein the numerical relationship is as follows: b is less than f, c% is less than d%, e% is less than c%, and g% is less than e%.