A method and system for energy management of diesel-photovoltaic-storage multi-energy complementary systems for isolated grid operation

Abstract

This application provides a method and system for multi-energy complementary energy management of diesel-photovoltaic-storage systems operating in isolated grids, belonging to the field of energy management technology. The method includes: when the residual state of charge of the energy storage unit in the isolated grid is lower than a low-power warning line, acquiring a load forecast time-series curve for a preset future period; identifying the off-peak and peak windows of load changes from the load forecast time-series curve; during the off-peak window, injecting the surplus power exceeding the real-time load power in constant power mode into the energy storage unit and monitoring the terminal voltage ramp-up rate; when the terminal voltage ramp-up rate decays to a preset threshold, switching the constant power mode to load tracking mode and establishing a power response reserve corresponding to the load increment; during the peak window, the diesel generator with the established power response reserve and the energy storage unit collaboratively respond to the peak load. The technical solution provided by this application can achieve orderly replenishment of electrical power in diesel-photovoltaic-storage energy management under low-power warning conditions.

Description

Technical Field

[0001] This application relates to the field of energy management technology, and more specifically, to a method and system for multi-energy complementary energy management of diesel-solar-storage grids operating in isolated grids. Background Technology

[0002] In isolated grid operation scenarios, such as remote islands, mining areas, or areas without electricity, traditional diesel power generation faces challenges such as high fuel costs, significant environmental pressure, and poor power supply reliability. Photovoltaic power generation is intermittent and volatile, and direct grid connection can easily cause frequency and voltage instability. Energy storage systems, as a buffer, can smooth out power fluctuations, but are limited by capacity and cost. Diesel-photovoltaic-storage multi-energy complementary systems can achieve spatiotemporal complementarity and power balance of energy by coordinating diesel generator sets, photovoltaic arrays, and energy storage units, thereby improving power supply reliability and economy. However, in isolated grids, the lack of support from a large power grid makes multi-energy coordination control and energy management a key technical challenge to ensure the stable operation of the system.

[0003] In existing diesel-solar-storage multi-energy complementary energy management systems, the system first monitors load demand, photovoltaic output, energy storage state of charge, and diesel generator operating parameters in real time. Stable power supply is achieved through hierarchical coordinated control. When photovoltaic power is sufficient, it is prioritized to supply the load, and excess energy is stored in batteries. When photovoltaic power is insufficient, energy storage discharges to supplement the deficit. If the energy storage capacity is too low, the diesel generator is started as a backup power source, and the energy storage can be controlled to charge. The energy management system aims to minimize fuel consumption and operating costs, employing model predictive control and other methods to rationally allocate the output of each unit, achieving comprehensive optimization of power quality under islanded grid conditions. However, in diesel-solar-storage energy management for islanded grid operation, when the remaining state of charge of the energy storage unit falls below a preset low-capacity warning line, its voltage external characteristics enter a nonlinear steep drop region. At this time, if subsequent periods experience... With high peak loads and rapid ramp-up rates, energy storage units, constrained by discharge cutoff voltage and discharge rate at low SOC, are unable to respond to transient power demands. Traditional control strategies often directly start the diesel generator and adopt an instantaneous load tracking mode when the battery is low. This forces the generator to operate at low power during off-peak periods, resulting in degraded combustion efficiency and increased carbon buildup. During the subsequent peak load periods, the inherent combustion inertia of the diesel engine and the sluggish response of the governor prevent the provision of power increments that match the load ramp-up rate on a timescale of seconds. As a result, the energy storage unit cannot effectively support subsequent peak load impacts due to the lack of orderly power replenishment under low battery alert conditions, leading to the risk of short-term power shortages in the isolated grid. Therefore, how to achieve orderly power replenishment in diesel-photovoltaic-storage energy management under low battery alert conditions has become a challenge for the industry. Summary of the Invention

[0004] This application provides a method and system for energy management of diesel-photovoltaic-storage multi-energy complementary systems for isolated grid operation, which can realize the orderly replenishment of electrical power in diesel-photovoltaic-storage energy management under low power warning conditions.

[0005] In a first aspect, this application provides a method for energy management of diesel-solar-storage multi-energy complementary systems operating in isolated grids, comprising the following steps:

[0006] When the remaining state of charge of the energy storage unit in the isolated grid is lower than the preset low power warning line at the current moment, the load forecast time series curve for the preset period in the future is obtained.

[0007] Using the minimum backup power duration required for the energy storage unit to maintain stable operation of the isolated grid as a constraint, the low-valley window and peak window of load change within the support duration of the energy storage unit from the current moment are identified from the load forecast time series curve.

[0008] During the off-peak window, the diesel generator is started and operated in constant power mode, and the surplus power exceeding the real-time load power in the constant power mode is injected into the energy storage unit, and the terminal voltage rise rate is continuously monitored during the recovery of the terminal voltage of the energy storage unit.

[0009] When the terminal voltage ramp-up rate is less than a preset threshold, the constant power mode is switched to load tracking mode, and a power response reserve corresponding to the load increment is established based on the load ramp-up characteristics of the peak window and the target speed range of the diesel generator.

[0010] During the peak window, the diesel generators with established power response reserves and the energy storage units respond in concert to the peak load.

[0011] In some embodiments, when the remaining state of charge of the energy storage unit in the isolated grid is lower than a preset low-power warning line at the current moment, obtaining the load forecast time-series curve for a preset future period specifically includes:

[0012] Read the remaining state of charge value of the energy storage unit in the isolated grid at the current moment. When the remaining state of charge value is lower than the preset low power warning line, generate a low power trigger signal.

[0013] Based on the low power trigger signal, the historical load database in the microgrid energy management is retrieved, and a similar daily load data sequence that matches the current operating day type, meteorological condition label, and seasonal attribute is extracted;

[0014] Using a time-series extrapolation prediction algorithm, with the similar daily load data sequence as input, the predicted load power value for a future preset period starting from the current moment is calculated.

[0015] The calculated load forecast power values ​​are arranged in chronological order to generate a load forecast time series curve for the future preset time period.

[0016] In some embodiments, using the minimum backup power duration required for the energy storage unit to maintain stable operation of the isolated grid as a constraint, identifying the low-valley window and peak-valley window of load changes from the current moment to the sustainable duration of the energy storage unit from the load forecast time-series curve specifically includes:

[0017] The available discharge capacity of the energy storage unit is calculated based on the current state of charge of the energy storage unit and the minimum allowable state of charge corresponding to the shortest backup power duration required to maintain stable operation of the islanded grid.

[0018] The available discharge capacity is compared with the load prediction time-series curve for hourly energy balance verification, thereby determining the duration during which the energy storage unit can continuously meet the load demand without triggering the minimum backup power duration constraint from the current moment.

[0019] The curve segment from the current moment to the supported duration is extracted from the load forecast time series curve, and the second derivative sign analysis is performed on the curve segment to mark the inflection point where the load power changes from decreasing to increasing as the valley feature point.

[0020] By combining the valley feature points with the sustainable duration, the low-valley window and peak window of load change within the sustainable duration of the energy storage unit from the current moment to the load prediction time series curve are determined.

[0021] In some embodiments, determining the valley window and peak window of load variation within the sustainable duration of the energy storage unit from the current moment to the sustainable duration from the current moment by combining the valley feature point with the sustainable duration specifically includes:

[0022] On the load forecast time series curve, with the valley bottom feature point as the center, the curve extends to both sides to the endpoints of the adjacent load change rate crossing zero points to form a valley window;

[0023] The remaining curve segment from the valley window to the end of the supportable duration is obtained, and the segment in the remaining curve segment where the load prediction power value continues to rise and the rate of rise exceeds the preset ramp rate threshold is identified as the peak window.

[0024] In some embodiments, during the off-peak window, starting the diesel generator to operate in constant power mode and injecting the surplus power exceeding the real-time load power in the constant power mode into the energy storage unit specifically includes:

[0025] At the beginning of the low point window, a start command is sent to the diesel generator, and the rated power of the diesel generator is set to the constant active power reference value corresponding to the optimal fuel efficiency point.

[0026] In the constant power mode, the real-time load power of the isolated AC bus is collected in real time, and the difference between the constant active power reference value and the real-time load power is obtained to obtain the surplus power that represents the instantaneous power margin.

[0027] The surplus power is used as the charging power command for the bidirectional converter of the energy storage unit to control the energy storage unit to absorb the surplus power, thereby injecting the surplus power exceeding the real-time load power into the energy storage unit in the constant power mode.

[0028] In some embodiments, continuously monitoring the terminal voltage ramp-up rate during the terminal voltage recovery process of the energy storage unit specifically includes:

[0029] During the process of injecting surplus power into the energy storage unit, the instantaneous value of the DC side voltage of the energy storage unit is continuously collected at a fixed sampling period.

[0030] The voltage change is obtained by performing a differential calculation between the instantaneous DC-side terminal voltage value collected at the current sampling time and the historical instantaneous DC-side terminal voltage value collected at the previous sampling time.

[0031] Divide the change in terminal voltage by the sampling period to calculate the current terminal voltage ramp-up rate.

[0032] In some embodiments, switching the constant power mode to load tracking mode when the terminal voltage ramp rate is less than a preset threshold specifically includes:

[0033] The terminal voltage ramp-up rate is continuously compared with a preset terminal voltage ramp-up rate attenuation threshold. When the terminal voltage ramp-up rate is less than the preset terminal voltage ramp-up rate attenuation threshold, a mode switching trigger signal is generated.

[0034] According to the mode switching trigger signal, the governor control target of the diesel generator is switched from maintaining a constant active power reference value to tracking the real-time load power, so that the diesel generator enters the load tracking mode.

[0035] Secondly, this application provides a diesel-photovoltaic-storage multi-energy complementary energy management system for isolated grid operation, used to execute a diesel-photovoltaic-storage multi-energy complementary energy management method for isolated grid operation. The system includes:

[0036] The acquisition module is used to acquire the load forecast time series curve for a preset period of time when the remaining state of charge of the energy storage unit in the islanded grid is lower than the preset low power warning line at the current moment.

[0037] The processing module is used to identify the low-valley window and peak-valley window of the load change from the current moment to the supportable duration of the energy storage unit within the load forecast time series curve, with the minimum backup power duration required for the energy storage unit to maintain stable operation of the isolated grid as a constraint.

[0038] The processing module is also used to start the diesel generator in constant power mode during the off-peak window, inject the surplus power exceeding the real-time load power in the constant power mode into the energy storage unit, and continuously monitor the terminal voltage rise rate during the terminal voltage rise of the energy storage unit.

[0039] The processing module is also used to switch the constant power mode to the load tracking mode when the terminal voltage ramp rate is less than a preset threshold, and to establish a power response reserve corresponding to the load increment based on the load ramp characteristics of the peak window and the target speed range of the diesel generator.

[0040] An execution module is used to coordinate the response of diesel generators with established power response reserves and energy storage units to peak loads during the peak window.

[0041] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, the processor being configured to acquire the code and execute the above-described diesel-photovoltaic-storage multi-energy complementary energy management method for isolated grid operation.

[0042] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned diesel-photovoltaic-storage multi-energy complementary energy management method for isolated grid operation.

[0043] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0044] The energy management method and system for diesel-solar-storage multi-energy complementary systems for islanded grid operation provided in this application firstly obtains the load forecast time-series curve for a preset future period when the remaining state of charge of the energy storage unit in the islanded grid is lower than a preset low-power warning line at the current moment; secondly, using the minimum backup power duration required for the energy storage unit to maintain stable operation of the islanded grid as a constraint, the low-valley window and peak-valley window of load change within the supportable duration of the energy storage unit from the current moment to the current moment are identified from the load forecast time-series curve; furthermore, during the low-valley window, the diesel generator is started in constant power mode. The system operates by injecting the surplus power exceeding the real-time load power in the constant power mode into the energy storage unit, and continuously monitoring the voltage ramp-up rate during the voltage recovery process of the energy storage unit. Then, when the voltage ramp-up rate is less than a preset threshold, the constant power mode is switched to load tracking mode, and a power response reserve corresponding to the load increment is established based on the load ramp-up characteristics of the peak window and the target speed range of the diesel generator. Finally, during the peak window, the diesel generator with the established power response reserve and the energy storage unit respond to the peak load in a coordinated manner.

[0045] Therefore, this application can achieve orderly replenishment of electrical power in diesel-photovoltaic energy storage management under low power alert conditions. Firstly, when the energy storage unit is under low power conditions, by acquiring the load forecast time-series curve and identifying off-peak and peak windows based on the shortest backup power duration constraint, the start-up and shutdown timing of the diesel generator and the allocation of charging power are based on prior information about future load changes, avoiding charging timing lag or insufficient backup power margin caused by relying solely on real-time status judgments. Secondly, during off-peak windows, the diesel generator is kept constantly running at its optimal fuel efficiency point, and surplus power is injected into the energy storage unit, enabling the diesel generator to maintain its optimal fuel efficiency. The generator operates under steady-state conditions with the lowest fuel consumption, while utilizing the maximum power margin during off-peak load windows to achieve efficient and rapid recharging of the energy storage unit. This balances the fuel economy of the diesel generator with the charging recovery speed of the energy storage unit while ensuring the stability of the islanded grid frequency. Furthermore, by monitoring the voltage rise rate of the energy storage unit in real time and switching from constant power mode to load tracking mode when it decays to a preset threshold, the dynamic response characteristics of the terminal voltage are used as the criterion for switching charging stages. This ensures that the mode switching timing corresponds to the actual electrochemical charging saturation trend, preventing constant power charging from entering the inefficient zone. This addresses the issue of energy waste and prevents overcharging of the energy storage unit. Then, based on load ramp-up characteristics, a power response reserve for the diesel generator is pre-established before the peak window arrives. The speed control system's operating point is offset downwards to reserve rotating reserve capacity. This provides the diesel generator with a feedforward frequency control margin that matches the predicted load increment, solving the inherent defect of the speed controller's lag in tracking peak load ramp-up in a timely manner. It also avoids the risk of the energy storage unit being unable to effectively support subsequent peak load impacts due to a lack of orderly power replenishment, leading to a short-term power shortage in the isolated grid. Finally, the peak window... During this period, the diesel generator releases power response reserves to bear the main component of load ramp-up, and the energy storage unit quickly compensates for speed regulation response deviations to achieve coordinated output. The slow-speed, high-capacity regulation characteristics of the diesel generator and the fast, low-capacity compensation characteristics of the energy storage unit are complemented in a time sequence, so that the combined output curve of the islanded grid under peak load impacts effectively matches the load ramp-up characteristics, thereby suppressing frequency fluctuations within the allowable range and achieving stable and economical operation of the islanded grid throughout the entire process under low energy storage conditions. In summary, the technical solution provided in this application can achieve orderly replenishment of electrical power in diesel-photovoltaic-storage energy management under low energy warning conditions. Attached Figure Description

[0046] Figure 1 This is an exemplary flowchart of a diesel-solar-storage multi-energy complementary energy management method for isolated grid operation, as shown in some embodiments of this application.

[0047] Figure 2 This is an exemplary flowchart illustrating the determination of terminal voltage ramp-up rate according to some embodiments of this application;

[0048] Figure 3This is a schematic diagram of the structure of a diesel-solar-storage multi-energy complementary energy management system for isolated grid operation, as shown in some embodiments of this application.

[0049] Figure 4 This is a schematic diagram of the structure of a computer device that implements a multi-energy complementary energy management method for diesel-solar-storage systems oriented towards isolated grid operation, according to some embodiments of this application. Detailed Implementation

[0050] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0051] refer to Figure 1 This figure is an exemplary flowchart of a diesel-solar-storage multi-energy complementary energy management method for isolated grid operation, according to some embodiments of this application. The figure mainly includes the following steps:

[0052] In step S101, when the remaining state of charge of the energy storage unit in the isolated grid is lower than the preset low power warning line at the current moment, the load forecast time series curve for the preset period in the future is obtained.

[0053] It should be noted that in this application, an isolated grid refers to a power supply system that is disconnected from the main grid and operates independently. The power generation and consumption within the system need to be dynamically balanced in real time. This includes independent power grids formed by self-built power sources due to remote geographical locations that cannot be covered by the main grid. When an isolated grid is in operation, due to the lack of large-capacity support from the main grid, the system has low inertia and weak anti-disturbance capability. Any power fluctuation may cause severe frequency or voltage oscillations. Therefore, the requirements for the speed regulation, voltage regulation performance and automatic control of generators are extremely high.

[0054] In some embodiments, when the remaining state of charge of the energy storage unit in the isolated grid is lower than a preset low-charge warning line at the current moment, the load forecast time-series curve for a preset future period is obtained by the following steps:

[0055] Read the remaining state of charge value of the energy storage unit in the isolated grid at the current moment. When the remaining state of charge value is lower than the preset low power warning line, generate a low power trigger signal.

[0056] Based on the low power trigger signal, the historical load database in the microgrid energy management is retrieved, and a similar daily load data sequence that matches the current operating day type, meteorological condition label, and seasonal attribute is extracted;

[0057] Using a time-series extrapolation prediction algorithm, with the similar daily load data sequence as input, the predicted load power value for a future preset period starting from the current moment is calculated.

[0058] The calculated load forecast power values ​​are arranged in chronological order to generate a load forecast time series curve for the future preset time period.

[0059] In specific implementation, firstly, when the remaining state of charge of the energy storage unit, read by the energy management system through the battery management system, is lower than the preset low-power warning threshold, the comparator logic module inside the energy management system outputs a low-power trigger signal indicating insufficient energy storage capacity. This low-power trigger signal is a binary enable flag used to initiate the subsequent load forecasting process. Secondly, the enable flag in the low-power trigger signal is transmitted to the database call interface of the microgrid central controller, triggering the historical load database to execute a structured query based on a structured query language. The triplet filtering fields in the WHERE clause of the query conditions are set to the running day type equal to the current day's type code, meteorological condition labels, etc. The weather type code and seasonal attribute returned by the current weather forecast interface are equal to the seasonal code mapped according to the current date. Using this query condition, a full-field matching search is performed on the historical load active power data stored in the historical load database, partitioned by day. The active power numerical sequence of the entire day, sampled at the fixed time resolution interval, recorded in the selected hit row is extracted as a similar daily load data sequence. This similar daily load data sequence refers to the time-series data of active power on historical dates that are completely consistent with the predicted target date in three dimensions: date attribute, weather type, and seasonal characteristics. Then, the entire similar daily load data sequence is used as the input tensor of a trained long short-term memory neural network model, and in the inference stage... The segment model unfolds its hidden state step by step. At each step, the load power prediction value output at the current time step is calculated based on the hidden state vector of the previous time step and the feature value of the input at the current time step. The load power prediction value output at this step is used as part of the input at the next time step. The expected load power value at each prediction interval point in the future preset time period starting from the current time is recursively calculated with a fixed time resolution, resulting in a set of load prediction power values ​​indexed by timestamp. The load prediction power value refers to the estimated value of the active power of the power load at each discrete time point in the future. The training process of the long short-term memory neural network model is as follows: First, extract the load prediction values ​​that match the current operating day type, meteorological conditions and seasonal attributes from the historical load database. Similar daily load data sequences were used as training samples, and all load power values ​​were subjected to max-min normalization to map them to the [0,1] interval. Then, a sliding window method was used to construct input-output pairs, with the normalized load value in each window as the input feature and the load value at the next time point as the prediction label. The constructed input-output pairs were randomly divided into training and validation sets in an 8:2 ratio. The training set data was input into a network containing two hidden layers (i.e., 64 and 32 memory units respectively) and a fully connected output layer with a batch size of 32. The mean squared error was used as the loss function, and an adaptive moment estimation optimizer was used for iterative training. After each training cycle, the loss value was calculated using the validation set.When the validation set loss value no longer decreases for 10 consecutive training epochs, training is terminated and the current model parameters are saved as the final prediction model, thus completing the training of the Long Short-Term Memory (LSTM) neural network model. Finally, all predicted load power values ​​are arranged in ascending order of timestamps and plotted as a continuous curve with time as the horizontal axis and load power as the vertical axis, forming a load prediction time series curve for a future preset period. The future preset period can be set according to actual needs and is not limited here.

[0060] It should be noted that the load forecast time series curve in this application refers to an arrayed power distribution sequence that characterizes the trend of power load change over time within a certain time window in the future. Determining the load forecast time series curve can provide the energy management system with prior information on load changes from the current moment until the diesel generator completes the establishment of power response reserves. This allows the system to identify in advance the low-load windows where the load power is relatively low and the peak windows where the load power is rapidly rising from the load forecast time series curve before the energy storage unit's state of charge approaches the minimum allowable boundary. This enables the system to lock the constant power charging time of the diesel generator during the low-load period to maximize the surplus power recovery efficiency. Furthermore, it allows the system to complete the power response reserve bias of the diesel generator speed control system in advance based on the load rise characteristics of the peak window before the peak load arrives, thus avoiding transient power shortages in the diesel generator due to governor response lag during peak load impacts.

[0061] In step S102, the minimum backup power duration required for the energy storage unit to maintain stable operation of the isolated grid is used as a constraint to identify the low-valley window and peak window of load change within the support duration of the energy storage unit from the current moment to the current moment from the load prediction time series curve.

[0062] In some embodiments, the identification of low-valley and peak windows of load variation from the current moment to the sustainable duration of the energy storage unit within the timeframe required for the energy storage unit to maintain stable operation of the isolated grid is achieved through the following steps, constrained by the minimum backup power duration required for the energy storage unit to maintain stable operation:

[0063] The available discharge capacity of the energy storage unit is calculated based on the current state of charge of the energy storage unit and the minimum allowable state of charge corresponding to the shortest backup power duration required to maintain stable operation of the islanded grid.

[0064] The available discharge capacity is compared with the load prediction time-series curve for hourly energy balance verification, thereby determining the duration during which the energy storage unit can continuously meet the load demand without triggering the minimum backup power duration constraint from the current moment.

[0065] The curve segment from the current moment to the supported duration is extracted from the load forecast time series curve, and the second derivative sign analysis is performed on the curve segment to mark the inflection point where the load power changes from decreasing to increasing as the valley feature point.

[0066] By combining the valley feature points with the sustainable duration, the low-valley window and peak window of load change within the sustainable duration of the energy storage unit from the current moment to the load prediction time series curve are determined.

[0067] In specific implementation, firstly, the remaining state of charge (SBC) value of the energy storage unit at the current moment is obtained, and the minimum allowable SBC setting value corresponding to the shortest backup power duration required to maintain stable operation of the isolated grid is retrieved. The difference between the two is multiplied by the rated capacity of the energy storage unit to obtain the available discharge capacity. The available discharge capacity refers to the electrical energy that the energy storage unit can release to the outside without falling below the minimum allowable SBC setting value. The minimum allowable SBC setting value can be set according to energy management requirements and is not limited here. Secondly, the available discharge capacity is used as the upper limit of energy constraints, and hourly energy balance is checked against the load forecast power value at each fixed time resolution interval in the load forecast time series curve starting from the current moment. The balance check method is to accumulate the load forecast power value of each interval multiplied by the time resolution at fixed time steps along the positive time axis. When the accumulated load energy at a certain time step first reaches the available discharge capacity, the corresponding time step is... The time coordinate is used as the extreme time boundary, and the time length from the start time to the extreme time boundary is taken as the sustainable duration. The sustainable duration refers to the critical time length during which the energy storage unit continuously supplies the predicted load with its current available discharge capacity while the state of charge just drops to the minimum allowable state of charge. Then, after determining the sustainable duration, in the array storage structure of the load prediction time series curve, a curve segment is extracted with the time index corresponding to the current moment as the starting point and the time index corresponding to the sustainable duration as the ending point. The load prediction power value of this curve segment is calculated by second-order difference discretization along the time axis. The index positions where the second-order difference value changes from negative to positive and the second-order difference of the previous moment is negative and the second-order difference of the next moment is positive are traversed and searched, and these are marked as valley feature points. Finally, by combining the valley feature points with the sustainable duration, the low valley window and peak window of load change within the sustainable duration of the energy storage unit from the current moment are determined from the load prediction time series curve.

[0068] It should be noted that, in this application, the valley feature point refers to the local minimum point in the load forecast time series curve where the load power changes from a downward trend to an upward trend. It is used to characterize the turning point when the load power of the load forecast time series curve changes from a continuous decline to a continuous rise within the supportable time range. The load power value corresponding to this moment is a local minimum value within this time interval, marking the boundary line between the islanded grid load and the ramp-up power consumption phase. The energy management system uses the time coordinate of the valley feature point to precisely anchor the constant power charging window of the diesel generator to the period with the lowest load level, so that the constant power output of the diesel generator can generate the maximum amount of surplus power after deducting the real-time load power for charging the energy storage unit, thereby maximizing the state of charge recovery efficiency of the energy storage unit under the constraint of the shortest backup power duration.

[0069] In some embodiments, the determination of the valley window and peak window for load changes within the sustainable duration of the energy storage unit from the current moment to the load forecast time series curve by combining the valley feature points with the sustainable duration is achieved through the following steps:

[0070] On the load forecast time series curve, with the valley bottom feature point as the center, the curve extends to both sides to the endpoints of the adjacent load change rate crossing zero points to form a valley window;

[0071] The remaining curve segment from the valley window to the end of the supportable duration is obtained, and the segment in the remaining curve segment where the load prediction power value continues to rise and the rate of rise exceeds the preset ramp rate threshold is identified as the peak window.

[0072] In specific implementation, firstly, using the index position of the valley feature point on the time axis as the center, the first-order difference value of the load forecast power between two adjacent sampling points is calculated by traversing forward and backward on the load forecast time series curve. When traversing forward to the zero-crossing point where the first-order difference value changes from positive to negative, the process stops and records this position as the starting boundary of the valley window. When traversing backward to the zero-crossing point where the first-order difference value changes from negative to positive, the process stops and records this position as the ending boundary of the valley window. The load forecast time series curve segment between the starting boundary and the ending boundary is determined as the valley window. The zero-crossing point of the load change rate refers to the position in the load forecast time series curve where the sign of the first-order difference value of the load forecast power changes. Then, the time interval from the ending boundary of the valley window to the end of the sustainable duration in the load forecast time series curve is extracted as the remaining curve segment. This remaining curve segment is then processed along... The time axis calculates the rate of increase of the predicted load power value between each adjacent sampling point. The longest continuous segment where the rate of increase is continuously greater than the preset ramp rate threshold and the predicted load power value increases monotonically is determined as the peak window. The preset ramp rate threshold refers to the maximum load growth rate that the diesel generator speed control system can track without triggering frequency over-limit. Specifically, the shortest ramp time required for the output power of the diesel generator speed controller to rise from the current output level to full load is obtained through the factory rated parameters of the diesel generator speed controller. At the same time, the load change amplitude limit corresponding to the maximum allowable frequency deviation is obtained from the frequency regulation standard of the isolated microgrid. After converting the two into a power change rate index based on unit time, the smaller value between the inherent ramp rate of the speed control system and the maximum allowable load growth rate limited by the frequency constraint is taken as the preset ramp rate threshold. This will not be elaborated further here.

[0073] It should be noted that, in this application, the off-peak window refers to the low-load operating period within the supported duration where the load power is near a local minimum and the trend of change is gradual. Its time boundary is defined by extending from the valley bottom characteristic point to both sides to the zero-crossing point of the adjacent load change rate. In this application, the peak window refers to the period of rapid load growth where the load power continuously rises and the rate of rise exceeds the preset ramp rate threshold after the end of the off-peak window and to the end of the supported duration. By determining the off-peak window, a clear constant power charging start-up time and duration can be provided for the diesel generator, enabling the diesel generator to output in constant power mode during the period of lowest load level, maximizing the injection of surplus power into the energy storage unit to improve its state of charge. Determining the peak window can provide a clear power response reserve establishment time and response intensity basis for the diesel generator, enabling the speed control system to complete speed offset and reserve output adjustment margin before the arrival of peak load. Then, within the peak window, the diesel generator and energy storage unit work together to cope with the rapid load rise, avoiding the drop in island frequency caused by the speed governor response lag.

[0074] In step S103, during the off-peak window, the diesel generator is started and operated in constant power mode, and the surplus power exceeding the real-time load power in the constant power mode is injected into the energy storage unit, and the terminal voltage rise rate is continuously monitored during the terminal voltage rise of the energy storage unit.

[0075] In some embodiments, during the off-peak window, the diesel generator is started and operated in constant power mode, and the surplus power exceeding the real-time load power in the constant power mode is injected into the energy storage unit by the following steps:

[0076] At the beginning of the low point window, a start command is sent to the diesel generator, and the rated power of the diesel generator is set to the constant active power reference value corresponding to the optimal fuel efficiency point.

[0077] In the constant power mode, the real-time load power of the isolated AC bus is collected in real time, and the difference between the constant active power reference value and the real-time load power is obtained to obtain the surplus power that represents the instantaneous power margin.

[0078] The surplus power is used as the charging power command for the bidirectional converter of the energy storage unit to control the energy storage unit to absorb the surplus power, thereby injecting the surplus power exceeding the real-time load power into the energy storage unit in the constant power mode.

[0079] In specific implementation, firstly, within the same control cycle at the start of the time when the current system clock timestamp has entered the low-end window, a communication message containing a start command and rated power is sent to the electronic control unit of the diesel generator set via the controller area network bus. The rated power is assigned to the constant active power reference value corresponding to the optimal fuel efficiency point calibrated in advance through the engine universal characteristic bench test. The optimal fuel efficiency point refers to the steady-state output power condition with the lowest unit fuel consumption rate of the diesel generator. Then, after the diesel generator set starts and completes speed and voltage build-up, its governor and automatic voltage regulator adjust the unit's output power to the constant active power reference value in a closed loop. At the same time, the power transmitter deployed on the incoming side of the isolated AC bus collects the instantaneous values ​​of three-phase voltage and three-phase current in real time at time intervals no less than the power frequency cycle, and calculates the real-time three-phase total active load power as the real-time load power and uploads it to the energy management system. The difference between the power reference value and the real-time load power is used to obtain the surplus power, which represents the instantaneous power margin. The surplus power refers to the power margin remaining after the constant active power output of the diesel generator meets the current load demand of the entire islanded grid, and can be used for energy storage charging. Finally, the surplus power is used as the active power command value of the bidirectional converter of the energy storage and sent to the digital controller of the bidirectional converter of the energy storage through the RS-485 serial communication interface. The bidirectional converter of the energy storage operates in constant power charging control mode. Its inner loop current regulator uses the surplus power as a reference input and generates a pulse width modulation drive signal through the proportional integral regulator to control the switching duty cycle of the IGBT power devices, so that the active power absorbed by the energy storage unit from the AC bus of the islanded grid is equal to the surplus power. This completes the injection of the surplus power exceeding the real-time load power into the energy storage unit in the constant power mode. The bidirectional converter of the energy storage unit refers to a four-quadrant power electronic conversion device that can realize bidirectional power flow between the AC side and the DC side.

[0080] In some embodiments, reference Figure 2 As shown, this figure is an exemplary flowchart illustrating the determination of the terminal voltage ramp-up rate according to some embodiments of this application. In this embodiment, continuous monitoring of the terminal voltage ramp-up rate during the terminal voltage recovery process of the energy storage unit can be achieved by the following steps:

[0081] In step S1031, during the process of injecting surplus power into the energy storage unit, the instantaneous value of the DC side voltage of the energy storage unit is continuously collected at a fixed sampling period.

[0082] In step S1032, the instantaneous value of the DC-side terminal voltage collected at the current sampling time is differentially calculated with the historical instantaneous value of the DC-side terminal voltage collected at the previous sampling time to obtain the change in terminal voltage.

[0083] In step S1033, the change in terminal voltage is divided by the sampling period to calculate the current terminal voltage ramp-up rate.

[0084] In specific implementation, firstly, during the period when the surplus power is injected into the DC bus of the energy storage unit in the form of DC current, the instantaneous value of the DC terminal voltage of the energy storage unit is continuously collected by the voltage Hall sensor on the DC side of the energy storage unit at a fixed sampling period. The sampling period refers to the time interval between two adjacent sampling operations. Then, the instantaneous value of the DC terminal voltage collected at the current sampling moment is differentially calculated with the historical instantaneous value of the DC terminal voltage collected at the previous sampling moment to obtain the terminal voltage change. The terminal voltage change refers to the actual increment value of the DC bus voltage of the energy storage unit within one sampling period. Finally, the terminal voltage change is divided by the sampling period to calculate the terminal voltage ramp-up rate at the current moment.

[0085] It should be noted that the terminal voltage ramp-up rate in this application refers to the rate at which the DC-side terminal voltage of the energy storage unit rises per unit time. As a real-time indicator of the charging acceptance capability during the recovery of the energy storage unit's state of charge, the terminal voltage ramp-up rate can be used to determine whether the energy storage unit is about to enter the constant voltage charging stage or the polarization effect saturation range by continuously monitoring the changing trend of this ramp-up rate. When the terminal voltage ramp-up rate gradually decays from the initial high value to the preset threshold, it indicates that the voltage response in the constant current charging mode of the energy storage unit has become relatively flat due to the internal resistance voltage drop and electrochemical polarization loss. The charging efficiency of continuing to inject at a high rate of constant power has decreased significantly. At this time, the degree of decay of the terminal voltage ramp-up rate is used as a criterion for switching the diesel generator's operating mode, triggering the energy management system to switch the diesel generator from constant power mode to load tracking mode, thereby avoiding fuel waste caused by the diesel generator maintaining high output operation after the energy storage unit's charging efficiency has decreased.

[0086] In step S104, when the terminal voltage ramp-up rate is less than a preset threshold, the constant power mode is switched to load tracking mode, and a power response reserve corresponding to the load increment is established based on the load ramp-up characteristics of the peak window and the target speed range of the diesel generator.

[0087] In some embodiments, when the terminal voltage ramp rate is less than a preset threshold, switching the constant power mode to load tracking mode is achieved by the following steps:

[0088] The terminal voltage ramp-up rate is continuously compared with a preset terminal voltage ramp-up rate attenuation threshold. When the terminal voltage ramp-up rate is less than the preset terminal voltage ramp-up rate attenuation threshold, a mode switching trigger signal is generated.

[0089] According to the mode switching trigger signal, the governor control target of the diesel generator is switched from maintaining a constant active power reference value to tracking the real-time load power, so that the diesel generator enters the load tracking mode.

[0090] In specific implementation, firstly, the terminal voltage ramp-up rate is continuously compared with a preset terminal voltage ramp-up rate attenuation threshold. When the terminal voltage ramp-up rate is less than the preset terminal voltage ramp-up rate attenuation threshold, a single pulse signal with a width of one control cycle is generated as a mode switching trigger signal. The preset terminal voltage ramp-up rate attenuation threshold refers to the empirical value of the terminal voltage ramp-up rate corresponding to the voltage inflection point of constant current charging to constant voltage charging in the cell parameter table provided by the energy storage unit manufacturer. The mode switching trigger signal is an enable signal used to trigger the state transition of the diesel generator control mode state machine. Then, according to the mode switching trigger signal, the governor control target of the diesel generator is switched from maintaining a constant active power reference value to tracking real-time load power, so that the diesel generator enters the load tracking mode.

[0091] It should be noted that the load tracking mode in this application refers to the operating control mode in which the diesel generator uses the actual active power on the load side as the target value of the speed governor to dynamically follow the power. By determining the load tracking mode, after the voltage ramp-up rate of the energy storage unit decays to a preset threshold, the rated power of the diesel generator can be switched from a constant value to a value that changes synchronously with the real-time load power of the islanded grid. In this way, the diesel generator directly undertakes the real-time balancing of the entire load power, and the energy storage unit no longer receives surplus charging power and exits the active charging state. It is only retained as a buffer unit for short-term power fluctuations in the islanded grid to smooth out the millisecond-level power deviation between load transients and the speed regulation response of the diesel generator. This avoids the extra fuel consumption caused by the diesel generator continuously operating at high output after the charging efficiency of the energy storage unit decreases, while ensuring the continuous stability of the islanded grid voltage frequency.

[0092] In some embodiments, establishing a power response reserve corresponding to the load increment based on the load ramp-up characteristics of the peak window and the target speed range of the diesel generator is achieved through the following steps:

[0093] Extract the ramp rate and power amplitude increment of the load power from the start time of the window to the peak time from the load prediction time series curve of the peak window, as the load ramp-up characteristics of the peak window;

[0094] The target speed response slope for speed regulation of the diesel generator is determined based on the ramp rate in the load ramp characteristics, and the output margin that the diesel generator needs to reserve is determined based on the power amplitude increment in the load ramp characteristics.

[0095] In the governor of the diesel generator, the speed adjustment range is set to a target speed range with the current operating speed as the lower limit and the sum of the current operating speed and the output adjustment margin as the upper limit. The current operating speed is biased to the lower limit of the target speed range. At the same time, the slope parameter of the governor speed response curve corresponding to the upper limit of the target speed range is set as the target speed response slope, forming a power response reserve corresponding to the load increment.

[0096] In specific implementation, firstly, the load power values ​​corresponding to the start time of the peak window and the load power values ​​corresponding to the peak time are extracted from the array structure of the stored load prediction time-series curves. The load power value at the peak time is subtracted from the load power value at the start time of the window to obtain the power amplitude increment. Then, the power amplitude increment is divided by the time span between the peak time and the start time of the peak window to obtain the ramp rate. The ramp rate refers to the increase in active power of the load within the peak window per unit time. The power amplitude increment refers to the total power difference between the load power at the start level and the peak level within the peak window. The two together constitute the load ramp-up characteristics of the peak window. The load ramp-up characteristics describe the intensity and amplitude of the dynamic change of the load rapidly climbing from a low level to a high level within the peak window. Subsequently, the transfer function model of the diesel generator speed control system is called. This model is a first-order inertial plus pure time delay transfer function identified through diesel generator step response test data. The ramp rate in the load ramp-up characteristics is used as the input of the expected output rate of change of the transfer function model. The slope required for the change of the governor speed reference value is then solved as the target. The speed response slope is calculated, and the power amplitude increment in the load ramp-up characteristic is divided by the active power output gain coefficient corresponding to the unit speed change of the diesel generator to obtain the output margin that the diesel generator needs to reserve. The output margin refers to the speed adjustment space that the diesel generator needs to reserve in advance to respond to the full load increment within the peak window. Finally, the speed adjustment range in the governor of the diesel generator is set to a target speed range with the current operating speed as the lower limit and the sum of the current operating speed and the output margin as the upper limit. The current operating speed is biased to the lower limit of the target speed range, and the slope parameter of the governor speed response curve corresponding to the upper limit of the target speed range is set as the target speed response slope to form a power response reserve corresponding to the load increment. The target speed range refers to the continuous speed adjustment range of the diesel generator from the lower limit of the speed bias to the upper limit. The governor speed response curve refers to the time history curve of the actual engine speed transitioning from the initial steady-state value to the target steady-state value when the speed reference value undergoes a step change.

[0097] It should be noted that, in this application, power response reserve refers to the ability of a diesel generator to increase its active power within a peak window by reserving it through speed offset, and to release it along a specified slope. This power response reserve is used to characterize the size of the spinning reserve capacity that a diesel generator can autonomously and quickly call upon through speed ramp without relying on the assistance of an energy storage unit. By determining the power response reserve, the diesel generator can be equipped with a mechanical frequency regulation response capability that matches the power amplitude increment in the load ramp-up characteristics before the rapid load ramp-up arrives in the peak window. When the peak load actually occurs, the speed governor only needs to adjust the speed from the offset lower limit upward along the target speed response slope to release the reserve power, thereby compensating for the inherent response lag defect of the diesel generator speed regulation system and enabling the rate of increase of the unit's output power to match the ramp-up rate of the peak window load.

[0098] In step S105, during the peak window, the diesel generator with established power response reserves and the energy storage unit respond in concert to the peak load.

[0099] In some embodiments, during the peak window, the coordinated response of a diesel generator with established power response reserves to the peak load by an energy storage unit is achieved through the following steps:

[0100] At the beginning of the peak window, the governor control signal of the diesel generator is enabled, so that the diesel generator increases its speed according to the target speed response slope to release the power response reserve.

[0101] The power difference between the active power output ramp rate during the diesel generator response process and the ramp rate of the peak window is calculated in real time, and the power difference is used as the transient power compensation command of the energy storage unit.

[0102] The energy storage unit is controlled to discharge according to the transient power compensation command to fill the instantaneous power deficit caused by the speed regulation response delay of the diesel generator, so that the combined output of the diesel generator and the energy storage unit follows the load ramp-up characteristics of the peak window in real time.

[0103] In practice, firstly, within the same control cycle when the current system clock timestamp has entered the start of the peak window, a speed control mode switching message is sent to the electronic control unit of the diesel generator set via the controller area network bus. This message switches the speed reference value mode of the governor from a fixed bias hold state to a ramp set state, and uses a pre-set target speed response slope as the rate of change parameter of the ramp function. The electronic control unit then increments the speed reference value in equal steps each control cycle according to this ramp function. This causes the electronic speed control actuator of the diesel generator set to adjust the throttle rack position in a closed loop based on the speed reference value, so that the engine speed changes from the target speed along the target speed response slope. The lower limit of the speed range continuously climbs upwards, thereby releasing the power response reserve at a controlled rate. Secondly, during the release of the power response reserve by the diesel generator, the actual output active power of the diesel generator is read from the power transmitter installed on the isolated AC bus. The active power output ramp rate of the diesel generator is calculated by dividing the difference between the output active power recorded in the previous control cycle and the control cycle duration. Simultaneously, the ramp rate in the peak window load ramp characteristic corresponding to the current moment is indexed from the stored load prediction time series curve as the expected ramp rate. The power difference obtained by subtracting the active power output ramp rate of the diesel generator from the expected ramp rate is a regularized value. The output increase of the generator lags behind the load increase. This power difference is used as the transient power compensation command for the energy storage unit. The transient power compensation command refers to the real-time control command used to instruct the bidirectional converter of the energy storage to discharge at a specific active power within a short time scale. Finally, in each control cycle, the transient power compensation command is refreshed to the digital signal processor control register of the bidirectional converter of the energy storage via the Ethernet bus. After receiving the updated compensation command, the control firmware of the bidirectional converter of the energy storage divides the power compensation command value by the current voltage sampling value of the energy storage unit to obtain the DC side discharge current reference value. The inner loop current proportional-integral regulator uses this discharge current reference value and the DC side current to calculate the reference value. The deviation between the actual discharge current value fed back by the Hall sensor is used to calculate the modulation ratio of the space vector pulse width modulation. This is then used to drive a three-phase full-bridge inverter circuit to convert the DC power of the energy storage unit into AC power and inject it into the islanded AC bus. This ensures that the active power output of the energy storage unit is equal to the transient power compensation command value, thus filling the instantaneous power deficit caused by the response delay of the diesel generator due to the speed control actuator and the engine's rotational inertia. The combined output of the diesel generator's active power and the energy storage unit's discharge power follows the real-time load power of the peak window in each control cycle. The combined output refers to the algebraic sum of the active power injected into the islanded AC bus by the diesel generator and the energy storage unit at the same time.

[0104] It should be noted that, in this application, coordinated response to peak load refers to the diesel generator and energy storage unit, which have established power response reserves, jointly undertaking the rapid rise demand of peak load during the peak window according to the master-slave power allocation principle. The diesel generator, as the main power source, continuously increases its output power along the preset target speed response slope to track the main component of load growth. The energy storage unit, as a rapid compensation power source, detects the transient power deviation between the diesel generator's output rise rate and the peak window load rise rate in real time, and injects active power equal to the deviation amplitude at a millisecond-level response speed to fill the transient power tracking error caused by the diesel generator's speed regulation system inertia and actuator delay. The outputs of the two are superimposed in real time on the islanded AC bus, so that the combined output curve is consistent with the load rise characteristics of the peak window in terms of amplitude and change slope, thereby ensuring that the islanded frequency remains within the allowable deviation range during the peak load impact.

[0105] Furthermore, in another aspect of this application, in some embodiments, this application provides a diesel-solar-storage multi-energy complementary energy management system for isolated grid operation, referencing... Figure 3 The figure is a schematic diagram of the structure of a diesel-solar-storage multi-energy complementary energy management system for isolated grid operation, according to some embodiments of this application. This system includes an acquisition module 201, a processing module 202, and an execution module 203, which are described below:

[0106] The acquisition module 201 in this application is mainly used to acquire the load forecast time series curve for a future preset period when the remaining state of charge of the energy storage unit in the islanded grid is lower than the preset low power warning line at the current moment.

[0107] Processing module 202, in this application, is mainly used to identify the low-valley window and peak window of load change from the current time to the supportable time of the energy storage unit from the load forecast time series curve, with the minimum backup power duration required for the energy storage unit to maintain stable operation of the isolated grid as a constraint.

[0108] The processing module 202 is also used to start the diesel generator to operate in constant power mode during the off-peak window, inject the surplus power portion exceeding the real-time load power in the constant power mode into the energy storage unit, and continuously monitor the terminal voltage rise rate during the terminal voltage recovery process of the energy storage unit.

[0109] In addition, the processing module 202 is also used to switch the constant power mode to the load tracking mode when the terminal voltage ramp-up rate is less than a preset threshold, and to establish a power response reserve corresponding to the load increment based on the load ramp-up characteristics of the peak window and the target speed range of the diesel generator.

[0110] The execution module 203 in this application is mainly used to coordinate the response of the diesel generator with the established power response reserve and the energy storage unit to the peak load during the peak window.

[0111] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, the processor being configured to acquire the code and execute the above-described diesel-photovoltaic-storage multi-energy complementary energy management method for isolated grid operation.

[0112] In some embodiments, reference Figure 4 The figure is a schematic diagram of the structure of a computer device implementing a diesel-solar-storage multi-energy complementary energy management method for isolated grid operation, according to some embodiments of this application. The diesel-solar-storage multi-energy complementary energy management method for isolated grid operation in the above embodiments can be implemented through... Figure 4 The computer device shown is used to implement this, and the computer device includes at least one processor 301, a communication bus 302, a memory 303, and at least one communication interface 304.

[0113] The processor 301 can be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more devices used to control the execution of the diesel-photovoltaic-storage multi-energy complementary energy management method for isolated grid operation in this application.

[0114] The communication bus 302 can be used to transmit information between the aforementioned components.

[0115] The memory 303 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 303 may exist independently and be connected to the processor 301 via the communication bus 302. The memory 303 may also be integrated with the processor 301.

[0116] The memory 303 stores program code for executing the scheme of this application, and its execution is controlled by the processor 301. The processor 301 executes the program code stored in the memory 303. The program code may include one or more software modules. In the above embodiments, the determination of the diesel-solar-storage multi-energy complementary energy management method for isolated grid operation can be achieved by the processor 301 and one or more software modules in the program code in the memory 303.

[0117] Communication interface 304 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0118] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0119] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0120] In addition, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described diesel-photovoltaic-storage multi-energy complementary energy management method for isolated grid operation.

[0121] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0122] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for energy management of diesel-solar-storage multi-energy complementary systems for isolated grid operation, characterized in that, Includes the following steps: When the remaining state of charge of the energy storage unit in the isolated grid is lower than the preset low power warning line at the current moment, the load forecast time series curve for the preset period in the future is obtained. Using the minimum backup power duration required for the energy storage unit to maintain stable operation of the isolated grid as a constraint, the low-valley window and peak window of load change within the support duration of the energy storage unit from the current moment are identified from the load forecast time series curve. During the off-peak window, the diesel generator is started and operated in constant power mode, and the surplus power exceeding the real-time load power in the constant power mode is injected into the energy storage unit, and the terminal voltage rise rate is continuously monitored during the recovery of the terminal voltage of the energy storage unit. When the terminal voltage ramp-up rate is less than a preset threshold, the constant power mode is switched to load tracking mode, and a power response reserve corresponding to the load increment is established based on the load ramp-up characteristics of the peak window and the target speed range of the diesel generator. During the peak window, the diesel generators with established power response reserves and the energy storage units respond in concert to the peak load.

2. The method as described in claim 1, characterized in that, When the remaining state of charge of the energy storage units within the isolated grid is lower than the preset low-power warning line at the current moment, the specific steps for obtaining the load forecast time-series curve for the preset future period include: Read the current state of charge value of the energy storage unit in the isolated grid. When the remaining state of charge value is lower than the preset low power warning line, generate a low power trigger signal. Based on the low power trigger signal, the historical load database in the microgrid energy management is retrieved, and a similar daily load data sequence that matches the current operating day type, meteorological condition label, and seasonal attribute is extracted; Using a time-series extrapolation prediction algorithm, with the similar daily load data sequence as input, the predicted load power value for a future preset period starting from the current moment is calculated. The calculated load forecast power values ​​are arranged in chronological order to generate a load forecast time series curve for the future preset time period.

3. The method as described in claim 1, characterized in that, Using the minimum backup power duration required for the energy storage unit to maintain stable operation of the isolated grid as a constraint, the low-valley window and peak-valley window for load changes within the sustainable duration of the energy storage unit, identified from the load forecast time-series curve, specifically include: The available discharge capacity of the energy storage unit is calculated based on the current state of charge of the energy storage unit and the minimum allowable state of charge corresponding to the shortest backup power duration required to maintain stable operation of the islanded grid. The available discharge capacity is compared with the load prediction time-series curve for hourly energy balance verification, thereby determining the duration during which the energy storage unit can continuously meet the load demand without triggering the minimum backup power duration constraint from the current moment. The curve segment from the current moment to the supported duration is extracted from the load forecast time series curve, and the second derivative sign analysis is performed on the curve segment to mark the inflection point where the load power changes from decreasing to increasing as the valley feature point. By combining the valley feature points with the sustainable duration, the low-valley window and peak window of load change within the sustainable duration of the energy storage unit from the current moment to the load prediction time series curve are determined.

4. The method as described in claim 3, characterized in that, By combining the valley feature points with the sustainable duration, the low-valley window and peak window of load change within the sustainable duration of the energy storage unit from the current moment to the load prediction time series curve are determined, specifically including: On the load forecast time series curve, with the valley bottom feature point as the center, the curve extends to both sides to the endpoints of the adjacent load change rate crossing zero points to form a valley window; The remaining curve segment from the valley window to the end of the supportable duration is obtained, and the segment in the remaining curve segment where the load prediction power value continues to rise and the rate of rise exceeds the preset ramp rate threshold is identified as the peak window.

5. The method as described in claim 1, characterized in that, During the aforementioned off-peak window, starting the diesel generator and operating it in constant power mode, and injecting the surplus power exceeding the real-time load power in the constant power mode into the energy storage unit specifically includes: At the beginning of the low point window, a start command is sent to the diesel generator, and the rated power of the diesel generator is set to the constant active power reference value corresponding to the optimal fuel efficiency point. In the constant power mode, the real-time load power of the isolated AC bus is collected in real time, and the difference between the constant active power reference value and the real-time load power is obtained to obtain the surplus power that represents the instantaneous power margin. The surplus power is used as the charging power command for the bidirectional converter of the energy storage unit to control the energy storage unit to absorb the surplus power, thereby injecting the surplus power exceeding the real-time load power into the energy storage unit in the constant power mode.

6. The method as described in claim 1, characterized in that, Continuously monitoring the terminal voltage rise rate during the terminal voltage recovery process of the energy storage unit specifically includes: During the process of injecting surplus power into the energy storage unit, the instantaneous value of the DC side voltage of the energy storage unit is continuously collected at a fixed sampling period. The voltage change is obtained by performing a differential calculation between the instantaneous DC-side terminal voltage value collected at the current sampling time and the historical instantaneous DC-side terminal voltage value collected at the previous sampling time. Divide the change in terminal voltage by the sampling period to calculate the current terminal voltage ramp-up rate.

7. The method as described in claim 1, characterized in that, When the terminal voltage ramp-up rate is less than a preset threshold, switching the constant power mode to load tracking mode specifically includes: The terminal voltage ramp-up rate is continuously compared with a preset terminal voltage ramp-up rate attenuation threshold. When the terminal voltage ramp-up rate is less than the preset terminal voltage ramp-up rate attenuation threshold, a mode switching trigger signal is generated. According to the mode switching trigger signal, the governor control target of the diesel generator is switched from maintaining a constant active power reference value to tracking the real-time load power, so that the diesel generator enters the load tracking mode.

8. A diesel-solar-storage multi-energy complementary energy management system for isolated grid operation, used to execute the diesel-solar-storage multi-energy complementary energy management method for isolated grid operation as described in any one of claims 1 to 7, characterized in that, The system includes: The acquisition module is used to acquire the load forecast time series curve for a preset period of time when the remaining state of charge of the energy storage unit in the islanded grid is lower than the preset low power warning line at the current moment. The processing module is used to identify the low-valley window and peak-valley window of the load change from the current moment to the supportable duration of the energy storage unit within the load forecast time series curve, with the minimum backup power duration required for the energy storage unit to maintain stable operation of the isolated grid as a constraint. The processing module is also used to start the diesel generator in constant power mode during the off-peak window, inject the surplus power exceeding the real-time load power in the constant power mode into the energy storage unit, and continuously monitor the terminal voltage rise rate during the terminal voltage rise of the energy storage unit. The processing module is also used to switch the constant power mode to the load tracking mode when the terminal voltage ramp rate is less than a preset threshold, and to establish a power response reserve corresponding to the load increment based on the load ramp characteristics of the peak window and the target speed range of the diesel generator. An execution module is used to coordinate the response of diesel generators with established power response reserves and energy storage units to peak loads during the peak window.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code and execute the diesel-solar-storage multi-energy complementary energy management method for isolated grid operation as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the diesel-solar-storage multi-energy complementary energy management method for isolated grid operation as described in any one of claims 1 to 7.

Images