CN110034569B - Combined system hybrid model prediction control method containing variable-speed seawater pumping and storage unit and chemical energy storage - Google Patents

Abstract

The invention provides a hybrid model prediction control method of a combined system comprising a variable-speed seawater pumping and storage unit and chemical energy storage. The real-time operation simulates the actual operation condition of the variable-speed seawater pumped storage unit and the energy storage battery in the power grid, which is interfered by the random prediction error of the renewable energy source; by adopting time-reduction-domain rolling optimization, the accumulated influence of small random prediction errors of the output of renewable energy sources on the SOC of the energy storage battery is solved, and the running economy of the variable-speed seawater pumped storage unit is considered; the accumulated influence of a large random prediction error of the output of the renewable energy on the SOC of the energy storage battery is dealt with by adopting heuristic control; and through feedback correction, the running states of the variable-speed seawater pumped storage unit and the energy storage battery are changed in time.

Description

Combined system hybrid model prediction control method containing variable-speed seawater pumping and storage unit and chemical energy storage

Technical Field

The invention belongs to the field of micro-grid operation control, and particularly relates to a Hybrid-MPC (Hybrid-MPC) method for online closed-loop Hybrid model predictive control of a composite energy storage combined system (island micro-grid operating in an isolated network) comprising a variable-speed seawater pumped-storage unit and chemical energy storage.

Background

The island micro-grid which operates in an isolated network and comprises a variable-speed seawater pumped storage unit, a lead-acid energy storage battery and renewable energy adopts a three-layer control structure in order to ensure that the micro-grid system can operate economically, safely and reliably. And the top layer (energy management) adopts an optimization algorithm to solve the objective function according to the day-ahead prediction information of the output of the renewable energy sources and the load demand, the unit operation constraint condition, the energy storage battery operation constraint condition, the objective function and the like, optimizes and arranges the day-ahead operation plan of the unit and the energy storage battery, and improves the system operation economy. The bottom layer adopts droop control, so that the power of the unit and the energy storage battery is adjusted in real time, and the energy supply-demand real-time balance relation of the micro-grid is met. The intermediate layer is subjected to secondary frequency modulation and voltage regulation control, and the requirement of the system on the electric energy quality is met.

However, the output prediction error of renewable energy is often large, so that the current operation plan of the unit and the energy storage battery established by the energy management system deviates from the actual operation condition. The unit and the energy storage battery adopt droop control to participate in primary frequency modulation of the micro-grid, and both the actual power values of the unit and the energy storage battery and the respective operation plan power set values can generate deviation under the influence of random prediction errors of renewable energy sources. The capacity of the energy storage battery is fixed and cannot be set to be large because the energy storage battery is expensive. And the energy storage battery has an upper SOC safe operation range and a lower SOC safe operation range, and the SOC can ensure safe operation in the range. During the operation of the energy storage battery, the deviation between the actual power and the operation planned power causes the SOC deviation (the product of the power deviation and the time) to be accumulated continuously, so that the SOC deviation between the actual SOC of the energy storage battery and the scheduling plan is larger and larger. When the actual SOC of the energy storage battery reaches the outside of the upper and lower SOC safety ranges at a certain moment in the day, the energy storage battery is overcharged or overdischarged, the energy storage battery controlled by a BMS (battery management system) runs in a power-limited mode or is off-line, and the running stability of the microgrid is reduced at the moment.

In order to deal with the influence of renewable energy sources on the operation of a microgrid, a heuristic coordination control strategy is proposed, such as the following patents: CN 201610158500-a little electric wire netting active real time scheduling method based on envelope control, with renewable energy real time output information as control variable coordinated control unit and energy storage battery, considered the influence of random prediction error. And a rolling optimization method is also adopted, so that the output of the renewable energy sources is repeatedly predicted in a short period in order to cope with the disturbance of random errors. And an economic scheduling model based on prediction errors corrects power fluctuation caused by errors. However, heuristic control has the disadvantage of being less economical than day-ahead optimization, and rolling optimization has the disadvantage of being less capable of coping with the interference of large prediction errors on the system. Most importantly, the methods neglect the accumulative effect of the random prediction error of the output of the renewable energy source, and do not solve the problem that the SOC deviation of the energy storage battery caused by the prediction error causes the limited power operation or off-line of the energy storage battery, and finally the operation stability of the micro-grid is reduced.

Disclosure of Invention

Aiming at the problem that the actual SOC deviation is too large due to the accumulative effect of the energy storage battery power deviation and the limited power operation or off-line of the energy storage battery is caused under the influence of the renewable energy output prediction error, a closed-loop Hybrid model prediction (Hybrid-MPC) control strategy for changing a microgrid operation plan on line according to the real-time SOC state feedback of the energy storage battery is provided, so that the SOC of the energy storage battery is always kept in a normal range under the disturbance of the prediction error on the microgrid, the energy storage battery always operates normally, and the operation stability of the microgrid is kept.

In order to achieve the above purpose, the following technical scheme is adopted, as shown in fig. 1. The Hybrid-MPC control strategy comprises four parts, namely real-time operation simulation, time-domain-reduction rolling optimization, heuristic coordination control and feedback correction. The method comprises the following specific steps:

a combined system hybrid model predictive control method containing a variable-speed seawater pumping and storage unit and chemical energy storage is characterized by comprising the following steps:

step 1, carrying out real-time operation simulation on the microgrid. Setting the slopes of droop curves of the unit and the energy storage battery according to 1, setting the power set point of the droop curve according to a day-ahead operation plan, inputting the output and the load power of the renewable energy source measured in real time into the microgrid, and simulating the actual operation of the microgrid;

step 2, monitoring the actual SOC of the energy storage battery in real time, and judging whether the value is in [ SOC ] _low ,SOC _high ]The range of the interval;

and 3, judging the disturbance magnitude of the prediction error if the actual SOC of the energy storage battery is out of limit. Taking a set value epsilon (such as 20% of the rated output of the renewable energy), judging that the error is small if the difference between the real-time measurement value of the output of the renewable energy and the predicted value before the day is smaller than the set value epsilon, and otherwise, judging that the error is large;

step 4, if the renewable energy output is judged to be small and the prediction error disturbance is judged, adopting a time-domain-reduction rolling optimization strategy (utilizing the advantage that the rolling optimization has strong capability of coping with small disturbance), and changing the operation plan instruction of the microgrid in the remaining time period in the day in real time according to the real-time SOC feedback value of the energy storage battery;

and 5, if the renewable energy output is judged to be large and the error disturbance is predicted, adopting a heuristic coordination control strategy (by utilizing the advantage that the heuristic control has strong capacity of coping with the large disturbance). In order to ensure the economy of heuristic control, a power instruction at the day before is used as a basic value, and the real-time power and state instructions of the unit and the energy storage battery are adjusted according to the real-time SOC, power and state feedback information of the micro-grid on the basis; the heuristic coordination control strategy is implemented as follows. The strategy consists of two layers of daily planning and real-time operation. Monitoring the SOC state of the energy storage battery in real time, and if the SOC is in an upper safety range and a lower safety range, continuing to operate the micro-grid according to the original day-ahead plan; if the SOC exceeds the upper and lower safety range lines, on the basis of a day-ahead plan instruction, a control instruction of the micro-grid is adjusted in real time according to the SOC of the energy storage battery and unit state feedback information, the working states of the unit and the energy storage battery are changed, the SOC of the energy storage battery is restored to be within the upper and lower safety ranges, and the normal operation of the energy storage is maintained. The heuristic coordination control strategy flow comprises the following control logics:

(1) If the actual SOC of the energy storage battery is less than or equal to the SOC _low Indicating that the energy storage battery is insufficient in electric energy. In order to avoid the over-discharge of the energy storage battery and the failure of continuously supplying power to the load, the energy storage battery needs to be charged. Firstly, the power flow direction of the energy storage battery and the on-off state of the unit are detected. If the power flows into the energy storage battery and the unit is shut down, the renewable energy is abundant, and the renewable energy is fully utilized to charge the energy storage battery;

if power flows into the energy storage battery and the unit is started, the unit is charged by the maximum power of the energy storage battery in order to improve the operation efficiency of the unit, reduce the total fuel consumption and increase the output of the unit. And if the net load causes the output of the unit to reach the maximum value at the moment, but the charging power of the energy storage battery is still lower than the maximum value, the output of the unit is set to the rated value. According to the micro-grid power balance relation, the net load power changes, the charging power of the energy storage battery correspondingly changes, and the value of the net load power changes is the difference between the rated power of the unit and the net load power;

and if the power flows out of the energy storage battery and the unit is shut down, the unit is started emergently to charge the energy storage battery. In order to protect the unit, the output of the unit is the rated minimum value in the first starting period, and the running instruction can be followed only by starting the second period;

and if the power flows out of the energy storage battery and the unit is started, increasing the output of the unit to charge the energy storage battery. If the net load causes the unit output to reach the maximum value, but the charging power of the energy storage battery is still lower than the maximum value, the unit output is set to the rated value. The net load power changes correspondingly, and the charging power of the energy storage battery changes correspondingly, and the value is the difference between the rated power of the unit and the net load power. In order to prolong the service life of the energy storage battery, the energy storage battery is charged to the SOC once the energy storage battery starts to be charged _high 。

(2) If the actual SOC of the energy storage battery is more than or equal to the SOC _high Indicating that the energy storage battery has sufficient electric energy. In order to avoid overcharging of the energy storage battery, the charging of the energy storage battery is limited. Firstly, the power flow direction of the energy storage battery and the on-off state of the unit are detected.

If the power flows out of the energy storage battery, the energy storage battery is discharged, and the unit and the energy storage battery continue to follow the original operation plan;

if the power flows into the energy storage battery and the unit is started, the energy storage battery is still charged, so that the total fuel consumption is reduced while the energy storage battery is protected, and the unit output is reduced to reduce the charging power of the energy storage battery. And if the output of the unit reaches the minimum value, but the energy storage battery is still charged, the energy storage battery starts to discharge by continuously increasing the quantity of the abandoned renewable energy, and if the energy storage battery starts to discharge, the quantity of the abandoned renewable energy stops increasing. If all renewable energy sources are abandoned and the energy storage battery is still charged, the unit is shut down. If the SOC exceeds the maximum allowable value due to the fact that the energy storage battery continues to be charged in the process of abandoning the renewable energy source, directly closing the unit and limiting the charging power of the energy storage battery to be zero, and balancing the load through the output of the renewable energy source;

if power flows into the energy storage battery and the unit is shut down, the energy storage battery starts to discharge by continuously increasing the amount of curtailed renewable energy, and if the energy storage battery starts to discharge, the increase of the amount of curtailed renewable energy is stopped.

And 6, the two strategies are complementarily matched, and the actual SOC of the energy storage battery is always kept in a normal operation range in an intra-day operation stage under the disturbance of the random prediction error of the renewable energy source through real-time state adjustment, so that the normal operation of the energy storage battery is ensured.

The invention comprehensively considers the accumulated influence of top layer control (dispatching control) and bottom layer control (droop control) on the SOC of the energy storage battery in the composite energy storage combined system on different time scales. Through real-time operation, the actual operation condition of the variable-speed seawater pumped storage unit and the energy storage battery in the power grid, which are interfered by the random prediction error of the renewable energy source, is simulated; by adopting time-domain-reduction rolling optimization, the accumulated influence of small random prediction errors of renewable energy output on the SOC of the energy storage battery is solved, and the running economy of the variable-speed seawater pumped storage unit is considered; the accumulated influence of a large random prediction error of the output of the renewable energy on the SOC of the energy storage battery is dealt with by adopting heuristic control; through feedback correction, the running states of the variable-speed seawater pumped storage unit and the energy storage battery are changed in time, the actual SOC of the energy storage battery is guaranteed to be always maintained in a normal running range, and the unfavorable condition that the energy storage battery exits from running due to overcharge/overdischarge caused by the fact that the actual SOC crosses an upper normal running range and a lower normal running range is avoided. The method has important significance for absorbing large-scale renewable energy by using the composite energy storage combined system containing the variable-speed pumping storage unit and the energy storage battery and maintaining the stability of the combined system in long-time operation.

Drawings

The invention is further illustrated by the following examples in conjunction with the accompanying drawings:

FIG. 1 is a block diagram of the Hybrid-MPC control strategy of the present invention.

Fig. 2 is a heuristic coordination control strategy of the present invention.

Fig. 3a shows the predicted power of the photovoltaic cell at the day before.

FIG. 3b shows the predicted power of the wind turbine day ahead.

Fig. 3c shows the predicted power day ahead of the load.

Fig. 3d is the predicted power day ahead of the payload.

Fig. 4a is the unit output.

Fig. 4b shows the stored energy power.

FIG. 4c is a pre-energy storage day SOC curve.

Fig. 4d shows the stored electric energy of the pumping unit at the warehouse day before.

Fig. 5a shows a net load power small prediction error.

Fig. 5b is the unit output.

Fig. 5c is the stored energy power.

Fig. 5d is an energy storage SOC curve.

Fig. 5e shows the stored electric energy of the pumping unit.

Fig. 6a is the unit output.

Fig. 6b shows the stored energy power.

Fig. 6c is an energy storage SOC curve.

Fig. 6d shows the stored electric energy of the pumping unit.

Fig. 7a is the unit output.

Fig. 7b is the stored energy power.

Fig. 7c shows the net load power large prediction error.

Fig. 7d is an energy storage SOC curve.

Fig. 7e shows the stored electric energy of the pumping unit.

Fig. 8 is a schematic diagram of the overall control method of the present invention.

Detailed Description

An MG-ROS (Micro Grid-Real Operation Simulation) program is developed in MATLAB by using an M language to simulate the actual Operation of the Micro-Grid under the disturbance of a net load random prediction error. The simulated operation microgrid comprises a proper amount of load, a photovoltaic array, a fan, a variable-speed seawater pumped storage unit and an energy storage battery, wherein the parameters are shown in tables 1 and 2. Random errors are added in the load and renewable energy power predicted values to simulate real-time load and real-time renewable energy power input of an actual micro-grid. And setting the power set points of the droop curves of the unit and the energy storage according to the instruction of the operation plan. The MG-ROS program employs the Hybrid-MPC control strategy described previously.

TABLE 1 parameters of load, photovoltaic, fan

TABLE 2 variable-speed seawater pumped storage unit and operating parameters of energy storage

The load and renewable energy power day-ahead predictions are shown in fig. 3 a-3 d. The unit and the energy storage day-ahead operation plan obtained through optimization calculation according to day-ahead prediction data, the microgrid optimization model and the operation parameters are shown in fig. 4a to 4 d. It can be seen that the planned SOC of the energy storage battery is always within the upper and lower safety range lines, meaning that the energy storage battery can always keep operating normally during the day. In the following simulation, the unit and the energy storage set the droop curves according to the operating plan in the figure. Meanwhile, in the island-type microgrid, the seawater pumped storage unit takes the sea as a lower reservoir and the upper reservoir as a conventional reservoir, so that the change curve of random group output of the electric energy (converted from water quantity according to unit efficiency) stored in the upper reservoir is only calculated. And only the result of the unit running in the power generation mode is given for simplicity.

The actual SOC of the stored energy in the microgrid under small random error disturbances is shown by the dashed lines in fig. 5a to 5 e. In the figure, the concept of virtual energy storage SOC is introduced to highlight the effect of random errors, but actually the SOC will not be negative. Therefore, the deviation between the actual energy storage SOC and the scheduling plan SOC is larger and larger due to the disturbance of the random error, the actual SOC crosses the safety range line at a certain moment, the over-discharge of the energy storage battery is caused, and the energy storage battery must quit the operation at the moment.

The results of applying the reduced time domain roll optimization under small random error perturbations are shown in fig. 6a to 6 d. Therefore, under the disturbance of random errors, the actual SOC of the energy storage battery at the points A1 and A2 just crosses the safety range line, the operation plan of the microgrid in the rest period is immediately changed, the actual SOC of the energy storage battery is always maintained within the upper safety range line and the lower safety range line, and the control effect of the time-reducing domain rolling optimization on the SOC of the energy storage battery is verified.

The results of using heuristic control under large random error perturbations are shown in fig. 7a to 7 e. It can be seen that the energy storage battery actual SOC crosses the lower bound at time B1 and crosses the upper bound at times B2, B3 and B4, respectively. Once the boundary is crossed, the MG-ROS immediately modifies the original day-ahead operation plan instruction, the actual SOC of the energy storage is always maintained within the upper and lower safety range lines, and the control effect of heuristic control on the SOC of the energy storage battery is verified.

Therefore, the invention achieves the following technical effects:

(1) By adopting the Hybrid-MPC control strategy, the SOC of the energy storage battery in the microgrid is always maintained in a normal range under the interference of random prediction errors, and the stable operation of the energy storage battery is ensured.

(2) By maintaining the stable operation of the energy storage battery, the operation stability of the microgrid is enhanced.

The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.

Claims (1)

1. A combined system hybrid model predictive control method containing a variable-speed seawater pumping and storage unit and chemical energy storage is characterized by comprising the following steps:

step 1, carrying out real-time operation simulation on a microgrid; setting the slopes of droop curves of the unit and the energy storage battery according to 1, setting the power set point of the droop curve according to a day-ahead operation plan, inputting the output and the load power of the renewable energy source measured in real time into the microgrid, and simulating the actual operation of the microgrid;

step 2, monitoring the actual SOC of the energy storage battery in real time, and judging whether the value is in [ SOC ] _low ,SOC _high ]The range of the interval;

step 3, judging the disturbance magnitude of the prediction error if the actual SOC of the energy storage battery is out of limit; taking a set value epsilon, if the difference between the real-time output measurement value of the renewable energy and the predicted value before the day is smaller than the set value epsilon, judging the error to be small, otherwise, judging the error to be large;

step 4, if the renewable energy output is judged to be small and the prediction error disturbance is judged, changing the operation plan instruction of the microgrid in the remaining time period in the day in real time by adopting a time-domain-reduction rolling optimization strategy according to the real-time SOC feedback value of the energy storage battery;

step 5, if the output of the renewable energy sources is judged to be large and the prediction error disturbance is judged, adopting a heuristic coordination control strategy; in order to ensure the economy of heuristic control, a power instruction at the day before is used as a basic value, and the real-time power and state instructions of the unit and the energy storage battery are adjusted according to the real-time SOC, power and state feedback information of the micro-grid on the basis; the heuristic coordination control strategy execution process is as follows; the strategy consists of two layers of day-ahead planning and real-time operation; monitoring the SOC state of the energy storage battery in real time, and if the SOC is in an upper safety range and a lower safety range, continuing to operate the micro-grid according to the original day-ahead plan; if the SOC exceeds the upper and lower safety range lines, on the basis of a day-ahead plan instruction, adjusting a control instruction of the microgrid in real time according to the SOC of the energy storage battery and unit state feedback information, changing the working states of the unit and the energy storage battery, enabling the SOC of the energy storage battery to be restored to the upper and lower safety ranges, and maintaining the normal operation of the energy storage battery; the heuristic coordination control strategy flow comprises the following control logics:

(1) If the actual SOC of the energy storage battery is less than or equal to the SOC _low Indicating that the energy storage battery is insufficient in electric energy; in order to avoid the over-discharge of the energy storage battery and the failure of continuously supplying power to the load, the energy storage battery needs to be charged; firstly, detecting the power flow direction of an energy storage battery and the on-off state of a unit; if the power flows into the energy storage battery and the unit is shut down, the renewable energy is abundant, and the renewable energy is fully utilized to charge the energy storage battery;

if power flows into the energy storage battery and the unit is started, in order to improve the operation efficiency of the unit, reduce the total fuel consumption and increase the output of the unit, the energy storage battery is charged with the maximum power; if the output of the unit is up to the maximum value due to the net load at the moment, but the charging power of the energy storage battery is still lower than the maximum value, the output of the unit is set to be a rated value; according to the micro-grid power balance relation, the net load power changes, the charging power of the energy storage battery correspondingly changes, and the value of the net load power changes is the difference between the rated power of the unit and the net load power;

if the power flows out of the energy storage battery and the unit is shut down, the unit is started emergently to charge the energy storage battery; in order to protect the unit, the output of the unit is the rated minimum value in the first starting period, and the running instruction can be followed only by starting the second period;

if the power flows out of the energy storage battery and the unit is started, the output of the unit is increased to charge the energy storage battery; if the output of the unit is up to the maximum value due to the net load, but the charging power of the energy storage battery is still lower than the maximum value, the output of the unit is set to a rated value; the net load power changes and the charging power of the energy storage battery correspondingly changes, and the value is the difference between the rated power of the unit and the net load power; in order to prolong the service life of the energy storage battery, the energy storage battery is charged to the SOC once the energy storage battery starts to be charged _high ；

(2) If the actual SOC of the energy storage battery is more than or equal to the SOC _high Indicating that the electric energy of the energy storage battery is sufficient; in order to avoid the overcharge of the energy storage battery, the charging of the energy storage battery needs to be limited; firstly, detecting the power flow direction of an energy storage battery and the on-off state of a unit;

if the power flows out of the energy storage battery, the energy storage battery is discharged, and the unit and the energy storage battery continue to follow the original operation plan;

if the power flows into the energy storage battery and the unit is started, the energy storage battery is still charged, so that the total fuel consumption is reduced while the energy storage battery is protected, and the unit output is reduced to reduce the charging power of the energy storage battery; if the output of the unit reaches the minimum value, but the energy storage battery is still charged, the energy storage battery starts to discharge by continuously increasing the quantity of the abandoned renewable energy, and if the energy storage battery starts to discharge, the quantity of the abandoned renewable energy stops increasing; if all renewable energy sources are abandoned and the energy storage battery is still charged, the unit is closed; if the SOC exceeds the maximum allowable value due to the fact that the energy storage battery continues to be charged in the process of abandoning the renewable energy source, directly closing the unit and limiting the charging power of the energy storage battery to be zero, and balancing the load through the output of the renewable energy source;

if power flows into the energy storage battery and the unit is shut down, the energy storage battery starts to discharge by continuously increasing the amount of curtailed renewable energy, and if the energy storage battery starts to discharge, the increase of the amount of curtailed renewable energy is stopped.

Images