Intelligent park energy management method and system

Abstract

The application discloses a kind of wisdom park energy management method and system, specifically related to wisdom park energy distributed storage management field, including by collecting the energy state parameter of energy storage equipment obtains state data, state data is divided into equal length time slice according to time sequence, and in each equal length time slice, power change rate, load offset and output difference are counted, and the statistical result is combined with charging time, discharging time, power value and write time to generate operation record;The application unifies the charging power and discharging power of distributed energy storage equipment according to equal length time slice, constructs scheduling matrix in combination with capacity boundary condition, and synchronizes each device operation intention before execution to quickly identify and eliminate cross-node conflict, so as to solve the problem of energy exchange bus contention, storage queue blockage and log loss caused by multiple autonomous devices in the traditional scheme.

Description

Technical Field

[0001] This invention relates to the field of distributed energy storage management technology in smart parks, and more specifically, to a smart park energy management method and system. Background Technology

[0002] In the energy management system of smart parks, distributed energy storage devices are gradually acquiring local autonomy, enabling them to autonomously decide on charging and discharging timing and power allocation based on parameters such as voltage, current, and load forecasts, and to persist operational data to the distributed storage system. However, when the park's electricity demand curve and renewable energy output curve fluctuate violently at the same time, multiple autonomous devices will trigger local emergency strategies in parallel, competing for the energy exchange bus and storage write channel almost simultaneously. In the absence of a unified planning, sharing, and arbitration mechanism, problems such as power allocation priority reversal, storage queue blockage, and log loss are very likely to occur within a millisecond time window.

[0003] Currently, there is a lack of a mechanism that can synchronize the operational intentions of each device in a timely manner at the decision-making front and quickly resolve cross-node conflicts. As a result, the concurrent competition for energy and storage resources is often difficult to detect and handle in a timely manner, which has become a key bottleneck restricting the stability of smart park energy management. Therefore, it is urgent to propose a smart park energy management method and system to realize the collaborative operation of multiple autonomous energy storage devices and the scheduling of distributed storage resources. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a smart park energy management method and system. By uniformly planning the charging and discharging power of distributed energy storage devices according to equal time slices, constructing a scheduling matrix in conjunction with capacity boundary conditions, and synchronizing the operating intentions of each device before execution to quickly identify and resolve cross-node conflicts, the method solves the problems of energy exchange bus contention, storage queue blocking, and log loss caused by the parallel triggering strategy of multiple autonomous devices in traditional solutions.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a smart park energy management method, comprising the following steps:

[0006] S1. Obtain status data by collecting energy status parameters of energy storage devices, divide the status data into equal time segments according to time series, and statistically analyze the power change rate, load offset and output difference in each equal time segment. Combine the statistical results with charging time, discharging time, power value and writing time to generate operation record.

[0007] S2. By receiving the operation record and detecting the change in energy status parameters, a trigger signal is obtained. The time slices are arranged in order of power change rate from high to low using a differential compression algorithm. Time slices with power change rate higher than a preset threshold are selected to generate broadcast data frames.

[0008] S3. Obtain the set of operation records by receiving local operation records and external operation records, align the set of operation records according to the start time of equal time slices, accumulate the power value within the sliding window, compare it with the upper limit of the power of the energy exchange channel, identify the over-limit equal time slices, and solve the conflict priority table by combining the writing time order.

[0009] S4. Obtain the priority order by receiving the conflict priority table, delay the charging or discharging time of the low-priority running record by the duration of a conflict-equal-length slice, add a delay compensation value to the writing time, and adjust the power value according to the smoothing strategy to generate the adjusted running record.

[0010] S5. Obtain collaborative execution parameters by receiving the adjusted operation records, and perform charging, discharging and data writing operations in parallel. After each operation, calculate the difference between the actual power and the planned power to generate a deviation curve, and output the deviation curve and storage status to the central management system.

[0011] In a preferred embodiment, in S1, state data is obtained by collecting energy state parameters of the energy storage device within a preset sampling interval. The energy state parameters include voltage value, current value, predicted load value and predicted output value. The state data is divided into equal time slices in chronological order, and a unique equal time slice index is assigned to each equal time slice.

[0012] Calculate the power change rate, load offset and output difference in each equal-length segment, bind the calculation results with the corresponding equal-length segment index, and record the binding results in the equal-length segment data table.

[0013] The data table of equal-length time slices is mapped to the charging time, discharging time, power value and writing time of the energy storage device to form a running record segment with the equal-length time slice index as the primary key and the charging and discharging parameters and calculation results as combined fields.

[0014] All runtime record fragments are combined according to the time order of the equally long slice index to generate a complete runtime record and output it.

[0015] In a preferred embodiment, in S1, state data is obtained by collecting energy state parameters of the energy storage device within a preset sampling interval. The energy state parameters include voltage value, current value, predicted load value and predicted output value. The state data is divided into equal time slices in chronological order, and a unique equal time slice index is assigned to each equal time slice.

[0016] Calculate the power change rate, load offset and output difference in each equal-length segment, bind the calculation results with the corresponding equal-length segment index, and record the binding results in the equal-length segment data table.

[0017] The data table of equal-length time slices is mapped to the charging time, discharging time, power value and writing time of the energy storage device to form a running record segment with the equal-length time slice index as the primary key and the charging and discharging parameters and calculation results as combined fields.

[0018] All runtime record fragments are combined according to the time order of the equally long slice index to generate a complete runtime record and output it.

[0019] In a preferred embodiment, in S3, by receiving local running records and external running records, the two types of running records are merged in chronological order to generate a running record set, and an equal-length time slice index is assigned to each running record in the running record set. At the same time, the power value and write time are recorded to form a running record table.

[0020] Based on the operation record table, perform a start time alignment operation on the operation record set, aggregate equal-length slices with the same start time, establish a mapping relationship between equal-length slices and power values, and generate a power mapping table.

[0021] A sliding window is set on the power mapping table. The window is slid up step by step according to the window span. The cumulative calculation is performed on the power value within each window range. The cumulative result is compared with the power limit of the energy exchange channel. The index of the same length slice that exceeds the power limit is identified and marked as over-limit state. The index of the same length slice that does not exceed the power limit is marked as normal state, thus forming a power comparison table.

[0022] Read the power comparison table, filter out the indexes of equal-length slices marked as exceeding the limit, sort them according to the magnitude by which the corresponding power value exceeds the power limit, and sort them according to the write time order within equal-length slices with the same magnitude of exceedance. Generate a conflict priority table based on the sorting results.

[0023] In a preferred embodiment, in S4, by receiving the conflict priority table, parsing the equal time slice index and the corresponding priority order in it, constructing the parsing result into a priority sequence, and maintaining the correspondence between the running record and the charging time, discharging time and writing time in the priority sequence;

[0024] In the priority sequence, identify the low-priority operation record one by one, postpone the charging or discharging time of the low-priority operation record by an interval equal to the duration of a conflict-prone long slice, and record the postponed time in the rearranged timetable.

[0025] Compare the write time of each running record with the original write time in the rearranged timetable, calculate the time difference between the two to solve for the delay compensation value, and add the delay compensation value to the write time of the corresponding running record to form a compensated running sequence table.

[0026] For adjacent time slice power values ​​in the runtime sequence table, a weighted allocation is performed on the power values ​​according to a preset smoothing strategy to generate a continuous power curve. The continuous power curve is then combined with the runtime sequence table to generate an adjusted runtime record.

[0027] In a preferred embodiment, in S5, by receiving the adjusted operation record, the timestamp field, power flow direction flag field and state of charge field are extracted, the timestamp field is then made continuous to form a time series, the power flow direction flag field is decoded into a power direction label, the state of charge field is converted into an available margin value, and the parsed result is written into the parameter buffer area.

[0028] By reading the time series, power direction label and available margin value in the parameter buffer, and combining them with the capacity boundary conditions, a scheduling matrix is ​​constructed. Equal time slice information is written into the row vector of the scheduling matrix, and power command values ​​are written into the column vector. During the matrix calculation process, correction operations are performed on entries that exceed the capacity upper limit or fall below the capacity lower limit.

[0029] The actual power value of each equal-length time slice is obtained by calling the power harvester on the energy exchange channel. The difference between the actual power value and the power command value in the scheduling matrix is ​​calculated at each time step to form a deviation sequence. The deviation sequence is then combined with the capacity boundary conditions to perform constraint correction, and the corrected deviation curve is output and written to the deviation buffer.

[0030] By reading the deviation curve from the deviation cache, combining it with the residual value in the running record, performing feedback index calculation to form a feedback dataset, and outputting the feedback dataset to the central management step.

[0031] In a preferred embodiment, a smart park energy management system includes an intent construction module, an intent synchronization module, a conflict detection module, an intent adjustment module, and a collaborative execution module.

[0032] The intent building module acquires state data by collecting energy state parameters of energy storage devices, divides the state data into equal time slices according to the time series, and statistically analyzes the power change rate, load offset and output difference in each equal time slice. The statistical results are combined with the charging time, discharging time, power value and writing time to generate operation records.

[0033] The intent synchronization module obtains the trigger signal by receiving the operation record and detecting the change in energy status parameters. It uses a differential compression algorithm to arrange equal-length time slices from high to low according to the power change rate, and selects equal-length time slices with a power change rate higher than a preset threshold to generate broadcast data frames.

[0034] The conflict detection module obtains a set of operation records by receiving local operation records and external operation records, aligns the set of operation records according to the start time of equal-length slices, accumulates the power value within a sliding window, compares it with the upper limit of the power of the energy exchange channel, identifies the over-limit equal-length slices, and solves the conflict priority table by combining the writing time order.

[0035] The intent adjustment module obtains the priority order by receiving the conflict priority table, delays the charging or discharging time of the low-priority running record by the duration of a conflict-equal-length slice, adds a delay compensation value to the write time, and adjusts the power value according to the smoothing strategy to generate the adjusted running record.

[0036] The collaborative execution module obtains collaborative execution parameters by receiving the adjusted operation records, performs charging, discharging and data writing operations in parallel, calculates the difference between the actual power and the planned power after each operation to generate a deviation curve, and outputs the deviation curve and storage status to the central management system.

[0037] The technical effects and advantages of this invention are as follows:

[0038] This invention constructs a scheduling matrix to uniformly plan the charging and discharging power of distributed energy storage devices in the time dimension and introduces capacity boundary constraints to ensure that each device receives a consistent power command before execution. This avoids the problems of energy exchange bus contention and storage queue blockage caused by the parallel triggering of autonomous strategies, and fundamentally solves the key problem of the difficulty in timely resolving cross-node conflicts in smart park energy management.

[0039] This invention utilizes time-series segmented calculation to feasiblely allocate charging and discharging power across time slices, ensuring that the system can maintain stable operation even under drastic fluctuations in demand curves and renewable energy output.

[0040] This invention introduces a runtime feedback mechanism on the basis of the scheduling matrix, which enables dynamic correction of power allocation during execution, thereby improving the adaptability of the scheduling strategy to real-time anomalies.

[0041] This invention incorporates distributed storage resources into a unified framework for energy scheduling, ensuring that charging and discharging commands are coordinated with the data storage process. This reduces the risk of log loss and write blocking, and achieves collaborative optimization of energy management and storage management. Attached Figure Description

[0042] Figure 1 This is a flowchart of the method steps of the present invention.

[0043] Figure 2 This is a schematic diagram of the system modules of the present invention.

[0044] Figure 3 This is a flowchart illustrating the process of generating and broadcasting operation records for this invention.

[0045] Figure 4 This is a flowchart of the conflict detection and priority generation process of the present invention.

[0046] Figure 5 This is the revised flowchart for the execution and feedback of the running record in this invention. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Refer to the instruction manual appendix Figure 1-5 An embodiment of the present invention provides a smart park energy management method, comprising the following steps:

[0049] S1. Obtain status data by collecting energy status parameters of energy storage devices, divide the status data into equal time segments according to time series, and statistically analyze the power change rate, load offset and output difference in each equal time segment. Combine the statistical results with charging time, discharging time, power value and writing time to generate operation record.

[0050] S2. By receiving the operation record and detecting the change in energy status parameters, a trigger signal is obtained. The time slices are arranged in order of power change rate from high to low using a differential compression algorithm. Time slices with power change rate higher than a preset threshold are selected to generate broadcast data frames.

[0051] S3. Obtain the set of operation records by receiving local operation records and external operation records, align the set of operation records according to the start time of equal-length slices, accumulate the power value within the sliding window, and compare it with the upper limit of the power of the energy exchange channel. The upper limit of the power of the energy exchange channel includes a static upper limit and a dynamic upper limit. Identify the over-limit equal-length slices and solve the conflict priority table by combining the writing time order.

[0052] S4. Obtain the priority order by receiving the conflict priority table, delay the charging or discharging time of the low-priority running record by the duration of a conflict-equal-length slice, add a delay compensation value to the writing time, and adjust the power value according to the smoothing strategy to generate the adjusted running record.

[0053] S5. Obtain collaborative execution parameters by receiving the adjusted operation records, and perform charging, discharging and data writing operations in parallel. After each operation, calculate the difference between the actual power and the planned power to generate a deviation curve, and output the deviation curve and storage status to the central management system.

[0054] In S1, state data is obtained by collecting energy state parameters of the energy storage device within a preset sampling interval. The energy state parameters include voltage value, current value, predicted load value and predicted output value. The state data is divided into equal time slices in chronological order, and a unique equal time slice index is assigned to each equal time slice.

[0055] Within each equal-length time slice, the power change rate, load offset, and output difference are calculated. The calculation results are then bound to the corresponding equal-length time slice index, and the binding results are recorded in the equal-length time slice data table. Specifically, the power change rate is calculated by extracting the power values ​​at adjacent sampling times within the equal-length time slice. The power difference between adjacent sampling times is divided by the sampling interval to obtain the power change rate sequence for that equal-length time slice, and the power change rate sequence is averaged to generate the power change rate for that equal-length time slice. The load offset is calculated by extracting the predicted load value and the load reference value within the equal-length time slice. The difference between the predicted load value and the load reference value is accumulated over sampling times and divided by the number of sampling times to obtain the average load offset for that equal-length time slice. The output difference is calculated by extracting the predicted output value and the target output value within the equal-length time slice. The difference between the predicted output value and the target output value is accumulated over sampling times and divided by the number of sampling times to obtain the average output difference for that equal-length time slice.

[0056] The data table of equal-length time slices is mapped to the charging time, discharging time, power value and writing time of the energy storage device to form a running record segment with the equal-length time slice index as the primary key and the charging and discharging parameters and calculation results as combined fields.

[0057] All runtime record fragments are combined according to the time order of the equally long slice index to generate a complete runtime record and output it.

[0058] In S2, the data of equal-length time slices in the running record are extracted, and the change of energy state parameters for each equal-length time slice is calculated. The change of energy state parameters includes the power change rate and the power change trend. Each change of energy state parameters is associated with the corresponding equal-length time slice index to form a change mapping table.

[0059] Read the change mapping table, sort the equal-length slice indexes by power change rate from high to low, and mark the power change trend during the sorting process to preserve the correspondence between the power fluctuation pattern of equal-length slices and the sorting priority.

[0060] The differential compression algorithm is used to compress the priority sorted equal-length slices. The complete energy state parameters of the first equal-length slice are stored in the reference data area, and the changes in energy state parameters of the remaining equal-length slices are stored in differential form. The reference data area and the differential form are organized into a set of compressed equal-length slices.

[0061] From the set of compressed equal-length segments, select equal-length segments with a power change rate higher than a preset threshold and a continuous and consistent power change trend. Reassemble the selection results with their corresponding equal-length segment indices to construct broadcast data frames. For equal-length segments with a power change rate lower than or equal to the preset threshold, merge them in the order after compression to generate summary equal-length segments and store them in a low-priority buffer.

[0062] In S3, by receiving local and external running records, the two types of running records are merged in chronological order to generate a running record set. Each running record in the running record set is assigned an equal-length time slice index. At the same time, the power value and write time are recorded to form a running record table.

[0063] Based on the operation record table, perform a start time alignment operation on the operation record set, aggregate equal-length slices with the same start time, establish a mapping relationship between equal-length slices and power values, and generate a power mapping table.

[0064] A sliding window is set on the power mapping table. The window is slid up step by step according to the window span. The power value within each window range is accumulated. The accumulated result is compared with the power limit of the energy exchange channel. The power limit of the energy exchange channel includes: the channel reference power limit determined according to the rated power value of the park's energy bus, and the dynamic power limit formed by adjusting according to the safety margin during operation. The two are used as the basis for power comparison. The equal-time slice index that exceeds the power limit is identified and marked as over-limit state, and the equal-time slice index that does not exceed the power limit is marked as normal state, forming a power comparison table containing all equal-time slice indexes and their comparison states.

[0065] Read the power comparison table, filter the indexes of equal-length slices marked as exceeding the limit, sort them according to the magnitude by which the corresponding power value exceeds the power limit, and sort them according to the write time order within equal-length slices with the same magnitude of exceedance. Generate a conflict priority table based on the sorting results and output it to the execution intention adjustment step.

[0066] In S4, by receiving the conflict priority table, parsing the equal time slice index and the corresponding priority order, constructing the parsing result into a priority sequence, and maintaining the correspondence between the running record and the charging time, discharging time and writing time in the priority sequence;

[0067] In the priority sequence, identify the low-priority operation record one by one, postpone the charging or discharging time of the low-priority operation record by an interval equal to the duration of a conflict-prone long slice, and record the postponed time in the rearranged timetable.

[0068] Compare the write time of each running record with the original write time in the rearranged timetable, calculate the time difference between the two to solve for the delay compensation value, and add the delay compensation value to the write time of the corresponding running record to form a compensated running sequence table.

[0069] For adjacent time slice power values ​​in the runtime sequence table, a weighted allocation is performed on the power values ​​according to a preset smoothing strategy to generate a continuous power curve. The continuous power curve is then combined with the runtime sequence table to generate an adjusted runtime record, which is then output to the collaborative execution step.

[0070] In S5, by receiving the adjusted operation record, the timestamp field, power flow direction flag field, and state of charge field are extracted. Then, the timestamp field is made continuous to form a time series, the power flow direction flag field is decoded into a power direction label, and the state of charge field is converted into an available margin value. The available margin value refers to the remaining execution space allowed under the combined effect of capacity boundary conditions and the power limit of the energy exchange channel. The available margin value is calculated by subtracting the current storage capacity value from the storage capacity limit value, comparing the resulting capacity margin with the power margin determined according to the power limit of the energy exchange channel, and taking the smaller value as the available margin value. The parsed result is written to the parameter buffer area.

[0071] By reading the time series, power direction label and available margin value in the parameter buffer, a scheduling matrix is ​​constructed in combination with the capacity boundary conditions. The capacity boundary conditions include the upper limit and lower limit of the storage capacity, which are used to correct the instructions during charging or discharging to prevent exceeding the upper limit or falling below the lower limit. Equal time slice information is written into the row vector of the scheduling matrix, and power instruction values ​​are written into the column vector. During the matrix calculation process, correction operations are performed on entries that exceed the capacity upper limit or fall below the capacity lower limit.

[0072] The actual power value of each equal-length time slice is obtained by calling the power harvester on the energy exchange channel. The difference between the actual power value and the power command value in the scheduling matrix is ​​calculated at each time step to form a deviation sequence. The deviation sequence is then combined with the capacity boundary conditions to perform constraint correction, and the corrected deviation curve is output and written to the deviation buffer.

[0073] By reading the deviation curve from the deviation cache and combining it with the remaining value in the running record, a feedback index is calculated to form a feedback dataset. The feedback dataset is then output to the central management step for subsequent scheduling correction calls.

[0074] It should be noted that in the formula structure involved in this scheme, dimensionless terms can be used as proportional or structural adjustment factors. When combined with quantities with units, they only play a role in numerical scaling and do not introduce new physical dimensions. Therefore, they will not change or confuse the overall unit system. This combination of "dimensionless terms and terms with units" can be understood as a composite structural expression commonly used in mathematical physics modeling. It conforms to the principle of dimensional consistency and has a clear physical interpretation basis.

[0075] Secondly, in the formula structure of this scheme, if multiple variables with different physical units are involved, including but not limited to time, mass or energy variables, their joint appearance is to express the collaborative modeling relationship of multiple physical mechanisms. Each variable can form a unified structure through function mapping, ratio combination or normalization adjustment, with clear units and clear meaning. The overall expression conforms to the principle of dimensional consistency and the conventional formula of engineering modeling.

[0076] In this solution, constants, weights, adjustment factors, threshold parameters, proportional coefficients, etc., are all adjustable control parameters for different application environments. Their values ​​depend on the target equipment configuration, data input characteristics, and performance optimization goals. During the implementation phase, they are set to converge within a reasonable range through model verification, performance constraints, or engineering calibration. Although these parameters do not have a unique preset value, they have clear adjustment logic and calculation paths. They belong to the deterministic setting process in engineering implementation. The purpose of this setting is to ensure that the solution is both universally adaptable and reproducible and operable, without affecting its technical clarity and feasibility.

[0077] definition In time slice Available margin value;

[0078]

[0079]

[0080]

[0081]

[0082]

[0083] in This is a lower limit constraint on capacity. Capacity upper limit constraint; For variables that evolve continuously over time; In time Power direction label; In time The actual power value; This is a mapping function between power and direction. Used to calculate energy increases or decreases; Let the capacity boundary constraint function be... This is used to ensure that the scoring results do not exceed the upper or lower limits; For the scheduling matrix in the th Line number The column's power command value; Let be the row vector of the scheduling matrix. Time information corresponding to equal-length slices; Let be the column vector of the scheduling matrix. The index corresponding to the power command; For time slices Available margin value; For scheduling functions, Combine time slices, power commands, and capacity conditions to generate scheduling entries; For time slices The power deviation value; For time slices The actual power value obtained from the power harvester; For the scheduling matrix in time slice The corresponding power command value; For time slices The corresponding power command index function;

[0084] in For time slices The corrected power deviation value; For time slices The original power deviation value; For time slices Available margin value; ( ) represents the capacity boundary condition; The constraint correction function is used to correct deviations that exceed the boundary conditions. For time slices The generated feedback dataset entries; For time slices Correction deviation value; For time slices Available margin value; For time slices The remaining value of the running record; For feedback indexing functions, Used to combine the correction deviation with the operating margin to generate feedback data;

[0085] Furthermore, firstly through and The integral result, combined with the capacity boundary conditions ( ) generated in time slice Available margin value Then utilize the available margin value With time index Power Command Index Constructing the scheduling matrix This establishes a correspondence between time slices and power commands; during the execution phase, power is actually collected... With the scheduling matrix in time slice Corresponding power command value The power deviation was calculated based on the difference. Combined with capacity margin With boundary conditions ( The deviation value is corrected to form the corrected deviation. The deviation will eventually be corrected. Available margin With running records Merge and generate feedback dataset This achieves a complete closed loop of capacity boundary constraints, scheduling matching, deviation correction, and feedback output.

[0086] A smart park energy management system includes an intent construction module, an intent synchronization module, a conflict detection module, an intent adjustment module, and a collaborative execution module;

[0087] The intent building module acquires state data by collecting energy state parameters of energy storage devices, divides the state data into equal time slices according to the time series, and statistically analyzes the power change rate, load offset and output difference in each equal time slice. The statistical results are combined with the charging time, discharging time, power value and writing time to generate operation records.

[0088] The intent synchronization module obtains the trigger signal by receiving the operation record and detecting the change in energy status parameters. It uses a differential compression algorithm to arrange equal-length time slices from high to low according to the power change rate, and selects equal-length time slices with a power change rate higher than a preset threshold to generate broadcast data frames.

[0089] The conflict detection module obtains a set of operation records by receiving local operation records and external operation records, aligns the set of operation records according to the start time of equal-length slices, accumulates the power value within a sliding window, compares it with the upper limit of the power of the energy exchange channel, identifies the over-limit equal-length slices, and solves the conflict priority table by combining the writing time order.

[0090] The intent adjustment module obtains the priority order by receiving the conflict priority table, delays the charging or discharging time of the low-priority running record by the duration of a conflict-equal-length slice, adds a delay compensation value to the write time, and adjusts the power value according to the smoothing strategy to generate the adjusted running record.

[0091] The collaborative execution module obtains collaborative execution parameters by receiving the adjusted operation records, performs charging, discharging and data writing operations in parallel, calculates the difference between the actual power and the planned power after each operation to generate a deviation curve, and outputs the deviation curve and storage status to the central management system.

[0092] In practical implementation, firstly, in the energy management of the smart park, the charging and discharging power data of distributed storage devices are continuously acquired, and upper and lower limits of storage capacity are set as constraints. Secondly, these power data are organized in chronological order and, combined with capacity boundary conditions, feasible power allocation results are calculated for each time period. Then, the allocation results are combined into a scheduling matrix, so that there is a clear power command at each time position. Finally, in actual operation, power commands are output step by step according to the scheduling matrix, and power allocation is corrected in combination with operational feedback, so as to always meet the capacity limit.

[0093] This solution addresses the resource contention issues that easily arise when distributed energy storage devices operate independently in smart park energy management. In particular, when there are drastic fluctuations in electricity demand and renewable energy generation, multiple devices may simultaneously activate local strategies, leading to chaotic power allocation, storage write congestion, and even data loss. This solution synchronizes the operating intentions of each device before decision-making and uses a scheduling matrix to quickly coordinate and resolve conflicts, enabling each energy storage device to operate collaboratively and rationally utilize distributed storage resources, thereby avoiding management instability caused by concurrent contention.

[0094] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart park energy management method, characterized in that, Includes the following steps: S1. Obtain status data by collecting energy status parameters of energy storage devices, divide the status data into equal time segments according to time series, and statistically analyze the power change rate, load offset and output difference in each equal time segment. Combine the statistical results with charging time, discharging time, power value and writing time to generate operation record. S2. By receiving the operation record and detecting the change in energy status parameters, a trigger signal is obtained. The time slices are arranged in order of power change rate from high to low using a differential compression algorithm. Time slices with power change rate higher than a preset threshold are selected to generate broadcast data frames. S3. Obtain the set of operation records by receiving local operation records and external operation records, align the set of operation records according to the start time of equal time slices, accumulate the power value within the sliding window, compare it with the upper limit of the power of the energy exchange channel, identify the over-limit equal time slices, and solve the conflict priority table by combining the writing time order. The S3 includes: receiving local operation records and external operation records, merging the two types of operation records in chronological order to generate an operation record set, assigning an equal-length time slice index to each operation record in the operation record set, and recording the power value and write time to form an operation record table; Based on the operation record table, perform a start time alignment operation on the operation record set, aggregate equal-length slices with the same start time, establish a mapping relationship between equal-length slices and power values, and generate a power mapping table. A sliding window is set on the power mapping table. The window is slid up step by step according to the window span. The cumulative calculation is performed on the power value within each window range. The cumulative result is compared with the power limit of the energy exchange channel. The index of the same length slice that exceeds the power limit is identified and marked as over-limit state. The index of the same length slice that does not exceed the power limit is marked as normal state, thus forming a power comparison table. Read the power comparison table, filter out the indexes of equal-length slices marked as exceeding the limit, sort them according to the magnitude by which the corresponding power value exceeds the power limit, and sort them according to the write time order within equal-length slices with the same magnitude of exceeding the limit. Generate a conflict priority table based on the sorting results. S4. Obtain the priority order by receiving the conflict priority table, delay the charging or discharging time of the low-priority running record by the duration of a conflict-equal-length slice, add a delay compensation value to the writing time, and adjust the power value according to the smoothing strategy to generate the adjusted running record. S5. Obtain collaborative execution parameters by receiving the adjusted operation records, and perform charging, discharging and data writing operations in parallel. After each operation, calculate the difference between the actual power and the planned power to generate a deviation curve, and output the deviation curve and storage status to the central management system.

2. The smart park energy management method according to claim 1, characterized in that: In S1, state data is obtained by collecting energy state parameters of the energy storage device within a preset sampling interval. The energy state parameters include voltage value, current value, predicted load value and predicted output value. The state data is divided into equal time slices in chronological order, and a unique equal time slice index is assigned to each equal time slice. Calculate the power change rate, load offset and output difference in each equal-length segment, bind the calculation results with the corresponding equal-length segment index, and record the binding results in the equal-length segment data table. The data table of equal-length time slices is mapped to the charging time, discharging time, power value and writing time of the energy storage device to form a running record segment with the equal-length time slice index as the primary key and the charging and discharging parameters and calculation results as combined fields. All runtime record fragments are combined according to the time order of the equally long slice index to generate a complete runtime record and output it.

3. The smart park energy management method according to claim 2, characterized in that: In S2, the data of equal-length time slices in the running record are extracted, and the change of energy state parameters for each equal-length time slice is calculated. The change of energy state parameters includes the power change rate and the power change trend. Each change of energy state parameters is associated with the corresponding equal-length time slice index to form a change mapping table. Read the change mapping table, perform priority sorting on the equal-length slice index according to the power change rate from high to low, and mark the power change trend during the sorting process; The differential compression algorithm is used to compress the priority sorted equal-length slices. The complete energy state parameters of the first equal-length slice are stored in the reference data area, and the changes in energy state parameters of the remaining equal-length slices are stored in differential form. The reference data area and the differential form are organized into a set of compressed equal-length slices. From the set of compressed equal-length segments, select equal-length segments with a power change rate higher than a preset threshold and a continuous and consistent power change trend. Reassemble the selection results with their corresponding equal-length segment indices to construct broadcast data frames. For equal-length segments with a power change rate lower than or equal to the preset threshold, merge them in the order after compression to generate summary equal-length segments and store them in a low-priority buffer.

4. The smart park energy management method according to claim 3, characterized in that: In S4, by receiving the conflict priority table, parsing the equal time slice index and the corresponding priority order, constructing the parsing result into a priority sequence, and maintaining the correspondence between the running record and the charging time, discharging time and writing time in the priority sequence; In the priority sequence, identify the low-priority operation record one by one, postpone the charging or discharging time of the low-priority operation record by an interval equal to the duration of a conflict-prone long slice, and record the postponed time in the rearranged timetable. Compare the write time of each running record with the original write time in the rearranged timetable, calculate the time difference between the two to solve for the delay compensation value, and add the delay compensation value to the write time of the corresponding running record to form a compensated running sequence table. For adjacent time slice power values ​​in the runtime sequence table, a weighted allocation is performed on the power values ​​according to a preset smoothing strategy to generate a continuous power curve. The continuous power curve is then combined with the runtime sequence table to generate an adjusted runtime record.

5. The smart park energy management method according to claim 4, characterized in that: In S5, by receiving the adjusted operation record, the timestamp field, power flow direction flag field and state of charge field are extracted. Then, the timestamp field is made continuous to form a time series, the power flow direction flag field is decoded into a power direction label, the state of charge field is converted into an available margin value, and the parsed results are written to the parameter buffer. By reading the time series, power direction label and available margin value in the parameter buffer, and combining them with the capacity boundary conditions, a scheduling matrix is ​​constructed. Equal time slice information is written into the row vector of the scheduling matrix, and power command values ​​are written into the column vector. During the matrix calculation process, correction operations are performed on entries that exceed the capacity upper limit or fall below the capacity lower limit. The actual power value of each equal-length time slice is obtained by calling the power harvester on the energy exchange channel. The difference between the actual power value and the power command value in the scheduling matrix is ​​calculated at each time step to form a deviation sequence. The deviation sequence is then combined with the capacity boundary conditions to perform constraint correction, and the corrected deviation curve is output and written to the deviation buffer. By reading the deviation curve from the deviation cache, combining it with the residual value in the running record, performing feedback index calculation to form a feedback dataset, and outputting the feedback dataset to the central management step.

6. A smart park energy management system, comprising the smart park energy management method described in claim 1, characterized in that, It includes an intent building module, an intent synchronization module, a conflict detection module, an intent adjustment module, and a collaborative execution module; The intent building module acquires state data by collecting energy state parameters of energy storage devices, divides the state data into equal time slices according to the time series, and statistically analyzes the power change rate, load offset and output difference in each equal time slice. The statistical results are combined with the charging time, discharging time, power value and writing time to generate operation records. The intent synchronization module obtains the trigger signal by receiving the operation record and detecting the change in energy status parameters. It uses a differential compression algorithm to arrange equal-length time slices from high to low according to the power change rate, and selects equal-length time slices with a power change rate higher than a preset threshold to generate broadcast data frames. The conflict detection module obtains a set of operation records by receiving local operation records and external operation records, aligns the set of operation records according to the start time of equal-length slices, accumulates the power value within a sliding window, compares it with the upper limit of the power of the energy exchange channel, identifies the over-limit equal-length slices, and solves the conflict priority table by combining the writing time order. The intent adjustment module obtains the priority order by receiving the conflict priority table, delays the charging or discharging time of the low-priority running record by the duration of a conflict-equal-length slice, adds a delay compensation value to the write time, and adjusts the power value according to the smoothing strategy to generate the adjusted running record. The collaborative execution module obtains collaborative execution parameters by receiving the adjusted operation records, performs charging, discharging and data writing operations in parallel, calculates the difference between the actual power and the planned power after each operation to generate a deviation curve, and outputs the deviation curve and storage status to the central management system.

Images