CN-121395421-B - Intelligent park energy management method and system

Abstract

The invention discloses a method and a system for intelligent park energy management, in particular to the field of intelligent park energy distributed storage management, which comprises the steps of acquiring state data by collecting energy state parameters of energy storage equipment, dividing the state data into equal-length time slices according to time sequences, counting power change rate, load offset and output difference value in each equal-length time slice, and combining the counting result with charging time, discharging time, power value and writing time to generate an operation record; according to the invention, the charging power and the discharging power of the distributed energy storage equipment are planned uniformly according to equal-length time slices, a scheduling matrix is constructed by combining capacity boundary conditions, and the running intention of each equipment is executed in advance to quickly identify and resolve cross-node conflict, so that the problems of energy exchange bus contention, storage queue blocking and log loss caused by a multi-autonomous-equipment parallel triggering strategy in the traditional scheme are solved.

Inventors

• CHENG JIAN

Assignees

• 安徽深亮科技有限公司

Dates

Publication Date

20260508

Application Date

20251022

Claims (6)

1. 1. An energy management method for an intelligent park is characterized by comprising the following steps: S1, acquiring state data by acquiring energy state parameters of energy storage equipment, dividing the state data into equal-length time slices according to time sequences, counting power change rate, load offset and output difference value in each equal-length time slice, and combining a counting result with charging time, discharging time, power value and writing time to generate an operation record; S2, acquiring a trigger signal by receiving an operation record and detecting the energy state parameter variation, and utilizing a differential compression algorithm to arrange long-time slices with the power variation rate from high to low, and screening the long-time slices with the power variation rate higher than a preset threshold value to generate a broadcast data frame; S3, acquiring an operation record set by receiving a local operation record and an external operation record, aligning the operation record set according to the starting time of the equal-length time slices, accumulating power values in a sliding window, comparing the accumulated power values with the upper power limit of an energy exchange channel, identifying an overrun equal-length time slice, and solving a conflict priority table by combining a writing time sequence; The S3 comprises the steps of combining the local operation records and the external operation records according to time sequence to generate an operation record set, distributing long-time slice indexes for each operation record in the operation record set, and simultaneously recording power values and writing time to form an operation record table; Performing initial time alignment operation on the operation record set according to the operation record table, aggregating long-time slices with the same initial time, establishing a mapping relation between the long-time slices and the power values, and generating a power mapping table; Setting sliding windows on a power mapping table, gradually sliding according to window spans, executing accumulation operation on power values in each window range, comparing an accumulation result with the upper power limit of an energy exchange channel, identifying long-time slice indexes exceeding the upper power limit, marking the long-time slice indexes as an overrun state, marking the long-time slice indexes which do not exceed the upper power limit as normal states, and forming a power comparison table; Reading a power comparison table, screening indexes of long time slices marked as overrun states from the indexes, sorting according to the amplitude of the corresponding power value exceeding the upper power limit, sorting according to the writing time sequence in the long time slices with the same exceeding amplitude, and generating a conflict priority table based on the sorting result; S4, acquiring a priority order by receiving a conflict priority table, delaying the charging time or discharging time of the low-priority operation record by the duration of a conflict long-time slice, adding a delay compensation value to the writing time, and adjusting a power value according to a smoothing strategy to generate an adjusted operation record; S5, acquiring collaborative execution parameters by receiving the adjusted operation records, executing charge, discharge and data writing operation in parallel, calculating the difference value between the actual power and the planned power after each operation is completed, generating a deviation curve, and outputting the deviation curve and the storage state to the central management system.
2. 2. The method for intelligent park energy management according to claim 1, wherein in S1, state data are acquired by collecting energy state parameters of the energy storage device in a preset sampling interval, wherein the energy state parameters comprise a voltage value, a current value, a predicted load value and a predicted output value, the state data are divided into equal-length time slices according to time sequence, and unique equal-length time slice indexes are allocated to each equal-length time slice; calculating the power change rate, the load offset and the output difference value in each equal-length time slice respectively, binding each calculation result with the corresponding equal-length time slice index, and recording the binding result to an equal-length time slice data table; The method comprises the steps of corresponding an equal-length time slice data table to charging time, discharging time, power value and writing time of energy storage equipment to form an operation record segment taking an equal-length time slice index as a main key and taking a charging and discharging parameter and a calculation result as a combined field; and combining all the operation record fragments according to the time sequence of the equal-length time slice index to generate a complete operation record and outputting the operation record.
3. 3. The method of claim 2, wherein in S2, the long-time slice data in the operation record are extracted, the energy state parameter variation of each long-time slice is calculated, the energy state parameter variation comprises a power variation rate and a power variation trend, and each energy state parameter variation is in one-to-one association with the corresponding long-time slice index to form a variation mapping table; reading a variable quantity mapping table, performing prioritized sorting on the equal long-time slice indexes according to the power change rate from high to low, and marking the power change trend in the sorting process; performing compression processing on the prioritized long-time slices by utilizing a differential compression algorithm, storing the complete energy state parameters of the first long-time slice into a reference data area, storing the energy state parameter variable quantities of the rest long-time slices into a differential form, and organizing the reference data area and the differential form into a compressed long-time slice set; And for the equal-length time slices with the power change rate lower than or equal to the preset threshold value, merging the equal-length time slices according to the sequence after compression to generate summary equal-length time slices and storing the summary equal-length time slices in a low priority buffer area.
4. 4. The method of claim 3, wherein in S4, the long-time slice index and the corresponding priority order are analyzed by receiving the conflict priority table, the analysis result is constructed into a priority sequence, and the corresponding relation between the operation record and the charging time, the discharging time and the writing time is kept in the priority sequence; Identifying low-priority operation records one by one in the priority sequence, delaying the charging time or discharging time of the low-priority operation records by one interval of the duration of the long-time slices such as a conflict, and recording the delayed time to a rearrangement timetable; Comparing the writing time of each operation record with the original writing time in the rearrangement schedule, calculating the time difference between the writing time and the original writing time to solve a delay compensation value, and superposing the delay compensation value to the writing time of the corresponding operation record to form an operation schedule with compensation; and performing weighted distribution on the power values of adjacent equal-length time slices in the operation time sequence table according to a preset smoothing strategy to generate a continuous power curve, and combining the continuous power curve with the operation time sequence table to generate an adjusted operation record.
5. 5. The method of claim 4, wherein in S5, the timestamp field, the power flow direction flag field and the state of charge field are extracted by receiving the adjusted operation record, then the timestamp field is continuously formed into a time sequence, the power flow direction flag field is decoded into a power direction tag, the state of charge field is converted into an available margin value, and the analyzed result is written into the parameter buffer area; the method comprises the steps of constructing a scheduling matrix by reading a time sequence, a power direction label and an available allowance value in a parameter buffer zone and combining capacity boundary conditions, writing long-time slice information in row vectors of the scheduling matrix, writing a power instruction value in column vectors, and executing correction operation on entries exceeding a capacity upper limit or being lower than a capacity lower limit in a matrix calculation process; The method comprises the steps of acquiring actual power values of all equal-time slices by calling a power collector on an energy exchange channel, performing difference calculation on the actual power values and power instruction values in a scheduling matrix time by time to form a deviation sequence, performing constraint correction by combining the deviation sequence with capacity boundary conditions, outputting a corrected deviation curve, and writing the corrected deviation curve into a deviation cache; and reading a deviation curve in the deviation cache, combining the deviation curve with a residual value in the operation record, executing feedback index calculation to form a feedback data set, and outputting the feedback data set to the central management step.
6. 6. An intelligent park energy management system comprising an intention construction module, an intention synchronization module, a conflict detection module, an intention adjustment module and a cooperative execution module, wherein the intelligent park energy management method is applied to the intelligent park energy management system according to claim 1; the intention construction module acquires state data by acquiring energy state parameters of the energy storage equipment, divides the state data into equal-length time slices according to time sequences, counts the power change rate, the load offset and the output difference value in each equal-length time slice, and combines the counting result with the charging time, the discharging time, the power value and the writing time to generate an operation record; the intention synchronization module acquires a trigger signal by receiving an operation record and detecting the change quantity of the energy state parameters, and utilizes a differential compression algorithm to arrange long-time slices with the power change rate from high to low, screen the long-time slices with the power change rate higher than a preset threshold value and generate a broadcast data frame; The conflict detection module acquires an operation record set by receiving a local operation record and an external operation record, aligns the operation record set according to the starting time of the equal-length time slices, accumulates power values in a sliding window, compares the accumulated power values with the upper power limit of an energy exchange channel, identifies an overrun equal-length time slice and solves a conflict priority table by combining a writing time sequence; the intention adjusting module acquires a priority order by receiving the conflict priority table, delays the charging time or the discharging time of the low-priority operation record by the duration of a conflict long-time slice, adds a delay compensation value in the writing time, and adjusts the power value according to the smoothing strategy to generate an adjusted operation record; The collaborative execution module acquires collaborative execution parameters by receiving the adjusted operation records, performs charge, discharge and data writing operations in parallel, calculates the difference value between the actual power and the planned power after each operation is completed, generates a deviation curve, and outputs the deviation curve and the storage state to the central management system.

Description

Intelligent park energy management method and system Technical Field The invention relates to the technical field of intelligent park energy distributed storage management, in particular to an intelligent park energy management method and system. Background In an energy management system of an intelligent park, distributed energy storage equipment gradually has local autonomous capability, can autonomously decide charge and discharge time and power distribution according to parameters such as voltage, current and load prediction and persistence of operation data to the distributed storage system, however, when a park electricity demand curve and a renewable energy output curve fluctuate violently at the same time, a plurality of autonomous equipment can trigger a local emergency strategy in parallel, almost synchronously contend with an energy exchange bus and a storage writing channel, and under the condition of lacking a unified plan sharing and arbitration mechanism, the problems of power distribution priority inversion, storage queue blocking, log loss and the like are very easy to occur in a millisecond time window; At present, a mechanism is lacking, which can timely synchronize the operation intention of each device at the decision front and quickly resolve the cross-node conflict, so that the concurrent contention of energy and storage resources is difficult to timely sense and process, and becomes a key bottleneck for restricting the energy management stability of the intelligent park, and therefore, the need for providing an intelligent park energy management method and system for realizing the cooperative operation of multiple autonomous energy storage devices and the scheduling of distributed storage resources is urgent. Disclosure of Invention In order to overcome the defects in the prior art, the embodiment of the invention provides an intelligent park energy management method and system, which are used for uniformly planning the charging power and the discharging power of distributed energy storage equipment according to equal-length time slices, constructing a scheduling matrix by combining capacity boundary conditions, and rapidly identifying and resolving cross-node conflicts by executing the operation intention of each device in front of the scheduling matrix, so that the problems of energy exchange bus contention, storage queue blocking and log loss caused by a parallel triggering strategy of multiple autonomous devices in the traditional scheme are solved. In order to achieve the purpose, the invention provides the following technical scheme that the intelligent park energy management method comprises the following steps: S1, acquiring state data by acquiring energy state parameters of energy storage equipment, dividing the state data into equal-length time slices according to time sequences, counting power change rate, load offset and output difference value in each equal-length time slice, and combining a counting result with charging time, discharging time, power value and writing time to generate an operation record; S2, acquiring a trigger signal by receiving an operation record and detecting the energy state parameter variation, and utilizing a differential compression algorithm to arrange long-time slices with the power variation rate from high to low, and screening the long-time slices with the power variation rate higher than a preset threshold value to generate a broadcast data frame; S3, acquiring an operation record set by receiving a local operation record and an external operation record, aligning the operation record set according to the starting time of the equal-length time slices, accumulating power values in a sliding window, comparing the accumulated power values with the upper power limit of an energy exchange channel, identifying an overrun equal-length time slice, and solving a conflict priority table by combining a writing time sequence; S4, acquiring a priority order by receiving a conflict priority table, delaying the charging time or discharging time of the low-priority operation record by the duration of a conflict long-time slice, adding a delay compensation value to the writing time, and adjusting a power value according to a smoothing strategy to generate an adjusted operation record; S5, acquiring collaborative execution parameters by receiving the adjusted operation records, executing charge, discharge and data writing operation in parallel, calculating the difference value between the actual power and the planned power after each operation is completed, generating a deviation curve, and outputting the deviation curve and the storage state to the central management system. In a preferred embodiment, in S1, acquiring state data by acquiring energy state parameters of the energy storage device within a preset sampling interval, where the energy state parameters include a voltage value, a current value, a predicted load value and a predicted output value, d