CN-122026446-A - Energy storage coordination scheduling method, equipment and medium for multi-energy power system

CN122026446ACN 122026446 ACN122026446 ACN 122026446ACN-122026446-A

Abstract

The invention relates to the technical field of operation and control of power systems, and discloses a method, equipment and medium for energy storage coordination scheduling of a multi-energy power system, wherein the method comprises the steps of establishing a dynamic characteristic model reflecting ageing and loss evolution of an energy storage unit, and expressing the dynamic characteristic model as a dynamic function of historical operation accumulation amount, current charge state and external environment parameters; the method comprises the steps of carrying out fusion analysis on a historical meteorological sequence, a load sequence and a carbon price sequence, outputting interval prediction results of new energy output, system load and carbon price in a future scheduling period, dynamically calculating a trigger index representing system uncertainty according to the interval prediction results, executing optimal scheduling according to the trigger index, updating the state of an energy storage unit according to actual operation data, and correcting the dynamic characteristic model. The invention can intelligently cope with uncertainty, thereby remarkably improving new energy consumption capability and reducing overall operation cost.

Inventors

PAN SHENGHONG
SONG JUNYING
WANG PENGFEI
XU TINGTING
LI YU

Assignees

贵州江源电力建设有限公司

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (9)

1. The energy storage coordination scheduling method of the multi-energy power system is characterized by comprising the following steps of, Establishing a dynamic characteristic model reflecting aging and loss evolution of the energy storage unit, and expressing the dynamic characteristic model as a dynamic function of historical operation accumulation, current charge state and external environment parameters; carrying out fusion analysis on the historical meteorological sequence, the load sequence and the carbon price sequence, and outputting the interval prediction results of new energy output, system load and carbon price in the future scheduling period; The fusion analysis comprises the steps of constructing a sequence prediction model comprising an encoder and a decoder, wherein the decoder utilizes historical weather hiding states, load hiding states and carbon price hiding states output by the encoder to execute weather condition attention retrieval, load self-attention retrieval and carbon price trend attention retrieval at each prediction execution time to respectively generate weather driving context vectors, load trend context vectors and carbon price trend context vectors, and the three context vectors are spliced and then are input together with prediction output as the decoder to update the internal state of the decoder; Dynamically calculating a trigger index representing the uncertainty of the system according to the interval prediction result, and executing optimized scheduling according to the trigger index; acquiring interval prediction results of the current and recent future time periods at each decision moment; Calculating the prediction uncertainty width of new energy output in a first time period in the future according to a preset high quantile predicted value and a preset low quantile predicted value, and dividing the prediction uncertainty width by a median predicted value in a corresponding time period to normalize the prediction uncertainty width to obtain a first uncertainty component; Calculating the instantaneous change rate of the system load at the current moment as a second uncertainty component; the first uncertainty component and the second uncertainty component are weighted and summed to obtain a comprehensive trigger index; Comparing the trigger index with a preset trigger threshold, and starting one-time rolling optimized scheduling after judging when the trigger index is larger than the trigger threshold; The objective function of the optimal scheduling comprises energy storage dynamic loss cost calculated in real time according to the dynamic characteristic model, and the dynamic characteristic model is used as a core constraint of the optimal scheduling; And updating the state of the energy storage unit according to the actual operation data, and correcting the dynamic characteristic model.
2. The method for coordinated scheduling of energy storage in a multi-energy power system according to claim 1, wherein the step of establishing a dynamic characteristic model reflecting aging and loss evolution of the energy storage unit comprises the steps of collecting historical operation data of the energy storage unit in a set operation period, wherein the historical operation data comprises a charge-discharge current sequence, a terminal voltage sequence, an environment temperature sequence and an initial state of charge; And establishing an equivalent internal resistance dynamic model representing the evolution of the internal health state of the energy storage unit by taking the historical operation data as input, wherein the equivalent internal resistance is modeled as a continuous function of the accumulated throughput electric quantity, the current state of charge and the current operation temperature of the energy storage unit.
3. The method for coordinated scheduling of energy storage in a multi-energy power system according to claim 2, wherein the outputting of the interval prediction results of the new energy output, the system load and the carbon price in the scheduling period comprises the steps of constructing a sequence prediction model comprising an encoder and a decoder; The encoder comprises a first sub-encoding network, a second sub-encoding network and a third sub-encoding network which are parallel, and the first sub-encoding network, the second sub-encoding network and the third sub-encoding network are respectively used for processing a historical weather characteristic time sequence, a historical system load time sequence and a carbon price sequence to perform characteristic extraction; Calculating the attention weight corresponding to each feature vector in the sequence in the decoder, and generating a corresponding context vector, wherein the corresponding context vector is spliced and fused and then passes through a fully-connected output layer to output a quantile predicted value of a future continuous scheduling period; the quantile predicted value comprises a high quantile predicted value, a middle quantile predicted value and a low quantile predicted value, and the high quantile predicted value, the middle quantile predicted value and the low quantile predicted value form an interval predicted result together.
4. The method for coordinated scheduling of energy storage in a multi-energy power system of claim 3, wherein said performing an optimized schedule according to said trigger indicator further comprises dynamically setting a look-ahead time window length for a current rolling optimization according to a value of said trigger indicator; presetting a maximum optimized window length and a minimum optimized window length; Calculating the difference value between the trigger index and the trigger threshold value, and determining the window reduction amount; Taking the difference value between the maximum optimized window length and the window reduction amount as a preliminary time window length; and when the preliminary time window length is smaller than the minimum optimized window length, the time window length of the rolling optimization is taken as the minimum optimized window length, otherwise, the preliminary time window length is taken.
5. The method for coordinated scheduling of energy storage in a multi-energy power system of claim 4, wherein taking the dynamic characteristic model as a core constraint of optimal scheduling comprises introducing a state update equation in the dynamic characteristic model as an equality constraint into optimal scheduling; The state updating equation defines a mathematical relationship between the state of the energy storage unit at the end of the current scheduling period, the state of the energy storage unit at the end of the last scheduling period and the charge and discharge power of the current period; And meanwhile, the maximum allowable charging power and the maximum allowable discharging power of the energy storage unit under the current state, which are calculated based on the dynamic characteristic model, are used as the power inequality constraint of the current energy storage unit in the current scheduling period.
6. The method of energy storage coordination scheduling of a multi-energy power system of claim 5, wherein correcting the dynamic characteristic model comprises collecting actual operation data of an energy storage unit after each scheduling period is finished, wherein the actual operation data comprise actual charge and discharge power and state of charge estimated values; utilizing the actual operation data, and adopting a recursive least square algorithm to identify and update the function parameters in the dynamic characteristic model on line; And applying the updated function parameters to the dynamic characteristic model of the next scheduling period to realize closed-loop self-adaptive correction of the model.
7. The energy storage coordination scheduling method of the multi-energy power system according to claim 6, further comprising obtaining a scheduling instruction generated by the optimized scheduling; And in the process of executing control, synchronously executing the updating of the state of the energy storage unit according to the actual operation data and correcting the dynamic characteristic model.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the multi-energy power system energy storage coordination scheduling method of any one of claims 1 to 7.
9. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of the multi-energy power system energy storage coordination scheduling method of any of claims 1 to 7.

Description

Energy storage coordination scheduling method, equipment and medium for multi-energy power system Technical Field The invention relates to the technical field of operation and control of power systems, in particular to a method, equipment and medium for energy storage coordination scheduling of a multi-energy power system. Background Multi-energy power systems are essentially integrated systems in which multiple energy forms are integrated. The power supply side of the system is provided with various power generation units such as thermal power, hydroelectric power, wind power, photovoltaic power and the like, the power transmission side takes an extra-high voltage and extra-high voltage transmission network as a main body, the energy distribution side is provided with a power distribution network, a heat supply network, a gas network, a hydrogen energy network and the like, the load side is provided with various energy consumption units such as electricity, heat, cold, hydrogen, fuel gas and the like, and the system is matched with various energy storage units such as electricity storage, heat storage, hydrogen storage, gas storage, cold storage, water storage and the like. The whole system is integrated with the supply, transportation, conversion, storage and consumption of multiple energy sources under the combined action of multiple devices in multiple links such as 'source-network-load-storage-conversion'. Because the energy source is various in form and has a large time span, each time and space level of the system needs to reach balance in energy and power. The balance of the short time scale determines the dynamic characteristics and stability of the system, and the economical efficiency and the operation efficiency of the medium-long-term balance determination system. Under the power structure condition that the new energy power generation installation ratio of wind power, photovoltaic and the like reaches 50% -70%, if all aspects such as energy supply economy, energy supply quality, energy supply reliability and the like are guaranteed to achieve good performance, and the requirements of zero carbon and even negative carbon energy supply are met, deep cooperation and efficient cooperation between various energy storage modes such as electricity storage, heat storage, hydrogen storage and the like and various power sources such as wind power, photovoltaic, hydropower, thermal power and the like are required to be achieved. Only on the basis of fully coordinating multi-energy storage and multi-type power supply, the safety and stability level, the carbon emission reduction capability, the energy supply economy and the energy supply quality of a novel power system which takes new energy as a main body and is compatible with various energy forms can be continuously improved. Disclosure of Invention In view of the above problems, the present invention provides a method, an apparatus, and a medium for energy storage coordination scheduling in a multi-energy power system. Therefore, the invention solves the technical problem that the scheduling plan and the actual dynamic capacity of the equipment are not matched due to the static simplification of the energy storage model, the rigidification of the scheduling frame and the single optimization target in the conventional multi-energy power system scheduling. In order to solve the technical problems, the invention provides a technical scheme of the energy storage coordination scheduling method of the multi-energy power system, which comprises the steps of establishing a dynamic characteristic model reflecting aging and loss evolution of an energy storage unit and expressing the dynamic characteristic model as a dynamic function of historical operation accumulation amount, current charge state and external environment parameters; carrying out fusion analysis on the historical meteorological sequence, the load sequence and the carbon price sequence, and outputting the interval prediction results of new energy output, system load and carbon price in the future scheduling period; Dynamically calculating a trigger index representing the uncertainty of the system according to the interval prediction result, and executing optimized scheduling according to the trigger index; The objective function of the optimal scheduling comprises energy storage dynamic loss cost calculated in real time according to the dynamic characteristic model, and the dynamic characteristic model is used as a core constraint of the optimal scheduling; And updating the state of the energy storage unit according to the actual operation data, and correcting the dynamic characteristic model. The method for energy storage coordination scheduling of the multi-energy power system comprises the steps of establishing a dynamic characteristic model reflecting aging and loss evolution of an energy storage unit, collecting historical operation data of the energy storage unit in a set operation period,