Photovoltaic-storage-chip coordination control method and system based on digital twin driving

Abstract

The present application belongs to the technical field of energy coordinated dispatching, and particularly relates to a light-storage-diesel coordinated control method and system based on digital twin driving. The present application can simultaneously consider the energy storage health state, diesel generator operating state and bus operating state, so that the control process has more comprehensive state input, improves the state perception ability under complex operating environment, predicts the load change trend, photovoltaic output change trend, energy storage state change trend and diesel generator standby demand in the future operating cycle based on digital twin state information, and determines future energy supply demand information according to the prediction result, so that the control strategy can respond to future supply and demand changes in advance, avoids the lag problem caused by the traditional control mode which only adjusts according to the current operating state, and improves the active adjustment ability in the microgrid operating process.

Description

Technical Field

[0001] This invention belongs to the field of energy coordinated dispatch technology, specifically relating to a photovoltaic-storage-diesel coordinated control method and system based on digital twin-driven technology. Background Technology

[0002] With the rapid development of new energy power generation technology, energy storage technology, and microgrid control technology, hybrid power supply systems consisting of photovoltaic power generation, energy storage systems, and diesel generators are increasingly being applied in scenarios such as islands, mining areas, remote areas, communication base stations, emergency power stations, and areas with weak power grids. These systems reduce the fuel consumption of traditional diesel generators by utilizing renewable energy sources such as photovoltaics, while simultaneously using energy storage systems to regulate fluctuations in renewable energy and load changes, thereby improving the economy and reliability of the power supply system. However, in off-grid or weak grid environments, due to the lack of power support from the main grid, photovoltaic output fluctuates significantly due to weather conditions, and the load side exhibits random variations. Diesel generators need to provide stable power supply and backup support, while energy storage systems need to handle power smoothing, peak-valley regulation, and transient response. Therefore, real-time coordinated control is required between photovoltaics, energy storage, and diesel generators.

[0003] Existing photovoltaic-storage-diesel control technologies mostly rely on fixed thresholds or rule-based control, primarily allocating energy based on the current operating status. This makes it difficult to adapt to dynamic operating conditions such as load fluctuations, changes in photovoltaic output, and changes in equipment status. They also fail to adequately consider the health status of energy storage and the economic operating characteristics of diesel generators, which can easily lead to problems such as accelerated energy storage lifespan decay, inefficient diesel engine operation, and frequent start-stops. Furthermore, the lack of multi-source status fusion, future energy demand prediction, and closed-loop correction mechanisms based on actual operating deviations results in insufficient adaptability of scheduling strategies and control precision, making it difficult to meet the stable and efficient operation requirements of photovoltaic-storage-diesel microgrids in complex scenarios. Summary of the Invention

[0004] The purpose of this invention is to provide a photovoltaic, energy storage and diesel generator coordinated control method and system based on digital twin drive, which can realize the intelligent coordinated operation of photovoltaic, energy storage and diesel generator through digital twin state perception, power supply prediction and closed-loop scheduling correction, thereby improving the power supply stability, energy utilization efficiency and equipment life of microgrid.

[0005] The specific technical solution adopted by this invention is as follows:

[0006] A digital twin-driven coordinated control method for light, energy storage, and diesel fuel includes:

[0007] Acquire multi-source operating status data during the operation of the photovoltaic-storage-diesel microgrid, and construct digital twin status information based on the multi-source operating status data;

[0008] Based on digital twin status information, the load change trend information, photovoltaic output change trend information, energy storage status change trend information and diesel generator backup demand information of the photovoltaic-storage-diesel microgrid in the future operation cycle are obtained, and the energy supply demand in the future operation cycle is predicted to obtain the corresponding energy supply demand information.

[0009] Based on energy supply and demand information and digital twin status information, determine diesel generator start-stop control information, diesel generator target output power range and energy storage charging and discharging power constraint information;

[0010] Based on the start-stop control information of the diesel generator, the target output power range of the diesel generator, and the charging and discharging power constraint information of the energy storage, a coordinated scheduling sequence among the photovoltaic, energy storage, and diesel generators is generated under the preset constraint conditions of the photovoltaic-storage-diesel microgrid.

[0011] The photovoltaic, energy storage system and diesel generator are controlled to perform corresponding power regulation operations according to the coordinated scheduling sequence, and the actual operation data during the execution process is collected in real time.

[0012] The actual operating data is compared with the digital twin state information to obtain the state deviation, and the coordination and scheduling sequence is corrected based on the state deviation.

[0013] In a preferred embodiment, multi-source operating status data of the photovoltaic-storage-diesel microgrid is acquired during its operation, and digital twin status information is constructed based on this multi-source operating status data, including:

[0014] The system acquires multi-source operational status data collected from different devices and monitoring nodes during the operation of the photovoltaic-storage-diesel microgrid. This multi-source operational status data includes load operation data, photovoltaic operation data, energy storage operation data, diesel generator operation data, and bus operation data.

[0015] The multi-source operational status data is preprocessed, including time synchronization, data cleaning and state variable transformation, to obtain an operational status dataset with a unified time scale.

[0016] Based on the operating status dataset, we extract equipment status features and electrical status features that characterize the operating status of the photovoltaic-storage-diesel microgrid, and construct digital twin status information based on these features.

[0017] In a preferred embodiment, the digital twin status information includes the energy storage state of charge, energy storage health status, energy storage temperature, available energy storage capacity, diesel generator output power, start-stop status, cumulative operating time, load power, bus voltage, and bus frequency.

[0018] In a preferred embodiment, based on digital twin state information, the load change trend information, photovoltaic output change trend information, energy storage state change trend information, and diesel generator backup demand information of the photovoltaic-storage-diesel microgrid are obtained during the future operating cycle. This allows for the prediction of energy supply demand during the future operating cycle, yielding corresponding energy supply and demand information, including:

[0019] Based on the load status information, photovoltaic operation status information, energy storage status information and diesel generator operation status information in the digital twin status information, the status change characteristics of the photovoltaic-storage-diesel microgrid in the future operation cycle are obtained.

[0020] Based on the characteristics of status changes, predict the load change trend, photovoltaic output change trend, energy storage status change trend, and diesel generator standby demand information for future operating cycles.

[0021] Based on load change trend information, photovoltaic power output change trend information, energy storage status change trend information, and diesel generator standby demand information, determine the energy supply demand information for the future operating cycle.

[0022] In a preferred embodiment, based on energy supply and demand information and digital twin status information, the start-stop control information of the diesel generator, the target output power range of the diesel generator, and the charging and discharging power constraint information of the energy storage are determined, including:

[0023] The system acquires information on load demand changes, photovoltaic output changes, energy storage status changes, and reserve capacity demand from the energy supply demand information, and extracts energy storage status information and diesel generator operating status information from the digital twin status information.

[0024] Based on information on load demand changes, photovoltaic output changes, energy storage status changes, and reserve capacity demand, determine whether the photovoltaic output and energy storage regulation capabilities meet the microgrid's energy supply needs during the current operating cycle.

[0025] When the photovoltaic output and energy storage regulation capacity cannot meet the energy supply demand, the start-up and shutdown conditions of the diesel generator are determined based on the diesel generator operating status information, and corresponding diesel generator start-up and shutdown control information is generated.

[0026] Based on energy demand information, diesel generator operating status information and preset diesel generator economic operating range, determine the target output power range of the diesel generator, so that the output power of the diesel generator is kept in the operating range with low unit power generation fuel consumption.

[0027] Based on the energy storage status information, including energy storage state of charge, energy storage health status, energy storage temperature, and available energy storage capacity, determine the current allowable charging and discharging capacity of the energy storage system.

[0028] Based on the current allowable charging and discharging capacity of the energy storage system and the energy supply demand information, energy storage charging and discharging power constraint information is generated.

[0029] In a preferred embodiment, based on diesel generator start-stop control information, the target output power range of the diesel generator, and energy storage charging and discharging power constraint information, a coordinated scheduling sequence among the photovoltaic, energy storage, and diesel generators is generated under the preset constraints of the photovoltaic-storage-diesel microgrid, including:

[0030] Based on the diesel generator start-stop control information, determine the corresponding operating status of the diesel generator in the future scheduling cycle, and determine the diesel generator power output range in each scheduling period based on the target output power range of the diesel generator.

[0031] Based on the energy storage charging and discharging power constraint information, the charging power range and discharging power range of the energy storage system in each scheduling period within the future scheduling cycle are determined, and the power adjustment range of the energy storage system is determined in combination with the energy supply demand information.

[0032] Based on energy demand information, photovoltaic output change trend information, diesel generator power output range and energy storage system power adjustment range, determine the photovoltaic output power, energy storage charging and discharging power and diesel generator output power corresponding to each scheduling period in the future scheduling cycle.

[0033] Based on the photovoltaic output power, energy storage charging and discharging power and diesel generator output power corresponding to each scheduling period, construct an equipment operation scheduling plan for the future scheduling cycle;

[0034] When the equipment operation scheduling scheme meets the preset constraints of the photovoltaic-storage-diesel microgrid, the equipment operation scheduling scheme will be used as a coordinated scheduling sequence. The preset constraints include power balance constraints, energy storage state of charge constraints, energy storage health state constraints, diesel generator operation constraints, and power supply reliability constraints.

[0035] In a preferred embodiment, the photovoltaic system, energy storage system, and diesel generator are controlled to perform corresponding power regulation operations according to a coordinated scheduling sequence, and real-time operating data during the execution process is collected, including:

[0036] Photovoltaic power regulation commands are generated based on photovoltaic output power, energy storage converter control commands are generated based on energy storage charging and discharging power, and diesel generator control commands are generated based on diesel generator output power.

[0037] The photovoltaic power regulation command, energy storage converter control command and diesel generator control command are sent to the corresponding execution equipment respectively, and the photovoltaic system, energy storage system and diesel generator are controlled to regulate power according to the coordinated scheduling sequence.

[0038] During the power regulation operation, actual operating data of the photovoltaic system, energy storage system, diesel generator and bus side are collected. The actual operating data includes actual energy storage state of charge, actual energy storage health status, actual energy storage temperature, actual energy storage available capacity, actual diesel generator output power, actual start-stop status, actual cumulative running time, actual load power, actual bus voltage and actual bus frequency.

[0039] In a preferred embodiment, the actual operating data is compared with the digital twin state information to obtain the state deviation, and the coordinated scheduling sequence is corrected based on the state deviation, including:

[0040] Extract the actual operating status information at the corresponding moment from the actual operating data collected during the execution of the coordinated scheduling sequence;

[0041] The actual operating status information is compared with the digital twin status information to obtain the status deviation between the actual operating status and the digital twin status.

[0042] Obtain the type and degree of deviation corresponding to the state deviation, and determine whether the output power of photovoltaic, energy storage charging and discharging power and diesel generator output power in the coordinated scheduling sequence need to be adjusted;

[0043] When the state deviation exceeds the preset deviation threshold, the photovoltaic output power command, energy storage charging and discharging power command and diesel generator output power command in the coordinated scheduling sequence are corrected according to the state deviation to obtain the updated coordinated scheduling sequence.

[0044] The system controls the photovoltaic, energy storage, and diesel generators to perform power regulation operations based on the updated coordinated scheduling sequence, and continues to update the digital twin status information using the actual operating data after execution.

[0045] The present invention also provides a photovoltaic-storage-diesel coordinated control system based on digital twin drive, used in the above-mentioned photovoltaic-storage-diesel coordinated control method based on digital twin drive, comprising:

[0046] The twin state module is used to acquire multi-source operating state data during the operation of the photovoltaic-storage-diesel microgrid, and to construct digital twin state information based on the multi-source operating state data.

[0047] The supply and demand module, based on digital twin status information, obtains information on load change trends, photovoltaic output change trends, energy storage status change trends, and diesel generator standby demand information of the photovoltaic-storage-diesel microgrid in the future operating cycle, and predicts the energy supply demand in the future operating cycle to obtain the corresponding energy supply and demand information.

[0048] The control classification module is used to determine the start-stop control information of the diesel generator, the target output power range of the diesel generator, and the energy storage charging and discharging power constraint information based on energy supply and demand information and digital twin status information.

[0049] The coordination and scheduling module generates a coordinated scheduling sequence between photovoltaic, energy storage system and diesel generator based on diesel generator start-stop control information, diesel generator target output power range and energy storage charging and discharging power constraint information, under the preset constraint conditions of photovoltaic-storage-diesel microgrid.

[0050] The actual operation module is used to control the photovoltaic, energy storage system and diesel generator to perform corresponding power regulation operations according to the coordinated scheduling sequence, and to collect the actual operation data in real time during the execution process;

[0051] The scheduling correction module is used to compare the actual running data with the digital twin status information, obtain the status deviation, and correct the coordinated scheduling sequence based on the status deviation.

[0052] And, a photovoltaic-storage-diesel coordinated control terminal based on digital twin technology, including:

[0053] One or more processors;

[0054] A storage device on which one or more programs are stored;

[0055] When one or more programs are executed by one or more processors, the one or more processors implement a digital twin-driven light-storage-diesel coordinated control method.

[0056] The technical effects achieved by this invention are as follows:

[0057] This invention can simultaneously consider the health status of energy storage, the operating status of diesel generators, and the operating status of the bus, enabling the control process to have more comprehensive status input and improving the status perception capability under complex operating environments. Based on digital twin status information, it can predict the load change trend, photovoltaic output change trend, energy storage status change trend, and diesel generator backup demand in the future operating cycle, and determine the future energy supply and demand information based on the prediction results. This allows the control strategy to respond to future supply and demand changes in advance, avoiding the lag problem caused by traditional control methods that only adjust based on the current operating status, and improving the proactive adjustment capability of the microgrid during operation.

[0058] This invention determines the start-stop control information of a diesel generator based on energy supply and demand information and the operating status of the diesel generator. It also determines the target output power range of the diesel generator by combining a preset economic operating range, so that the diesel generator can operate in the operating range with the lowest unit fuel consumption. When the load fluctuation is small, the energy storage system can be used to handle short-term power changes, reducing the frequent start-up and low-load operation of the diesel generator, thereby reducing fuel consumption and mechanical wear of the equipment. It not only considers the state of charge of the energy storage, but also comprehensively considers the health status, temperature and available capacity of the energy storage, so that the energy storage system can dynamically adjust its charging and discharging capacity according to its own health status. When the health status of the energy storage declines, its charging and discharging power and operating range can be limited to reduce the lifespan degradation caused by deep charging and discharging, high-rate operation and high-temperature operation, and improve the long-term operational reliability of the energy storage system.

[0059] This invention, by combining the start-stop status of the diesel generator, the output power range of the diesel generator, and the charging and discharging constraints of energy storage, generates a coordinated scheduling sequence under the constraints of power balance, equipment operation, and power supply reliability. This achieves a reasonable power allocation among photovoltaic, energy storage systems, and diesel generators, enabling the coordinated operation of priority consumption of new energy, flexible adjustment of energy storage, and stable power supply from diesel engines. This improves the overall energy utilization efficiency of the microgrid and realizes a complete closed-loop control process from state perception, trend prediction, coordinated control to feedback optimization. It avoids the problem of independent prediction and execution in traditional control methods and improves the overall operational stability and intelligence level of the photovoltaic-storage-diesel microgrid. Attached Figure Description

[0060] Figure 1 This is a flowchart of the method provided by the present invention;

[0061] Figure 2 This is a system module diagram provided by the present invention;

[0062] Figure 3 This is a schematic diagram of the digital twin closed-loop coordinated control system architecture for a photovoltaic-storage-diesel microgrid provided by the present invention.

[0063] Figure 4 This is a schematic diagram of the spatiotemporal prediction model based on physical priors provided by the present invention;

[0064] Figure 5 This is a schematic diagram of the adaptive SOC window and power derating process for energy storage SOH provided by the present invention;

[0065] Figure 6 This invention provides a schematic diagram of the diesel generator's fuel consumption curve, unit fuel consumption curve, and economic operating zone.

[0066] Figure 7 This is a schematic diagram of the multi-objective coordinated scheduling process of energy storage and diesel generator provided by the present invention;

[0067] Figure 8 This is a schematic diagram of the three-layer coordinated control and rolling feedback timing provided by the present invention;

[0068] Figure 9 This is a schematic diagram of the digital twin virtual-real closed-loop correction mechanism provided by the present invention. Detailed Implementation

[0069] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0070] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0071] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.

[0072] Furthermore, the present invention will be described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention in detail, the schematic diagrams are merely examples for ease of explanation and should not limit the scope of protection of the present invention.

[0073] Please see the appendix Figure 1 , Figures 3 to 9 As shown, a coordinated control method for light, energy storage, and diesel fuel based on digital twin-driven technology is provided, including:

[0074] S1. Acquire multi-source operating status data during the operation of the photovoltaic-storage-diesel microgrid, and construct digital twin status information based on the multi-source operating status data;

[0075] S2. Based on the digital twin status information, obtain the load change trend information, photovoltaic output change trend information, energy storage status change trend information and diesel generator backup demand information of the photovoltaic-storage-diesel microgrid in the future operating cycle, predict the energy supply demand in the future operating cycle, and obtain the corresponding energy supply demand information.

[0076] S3. Based on energy supply and demand information and digital twin status information, determine the diesel generator start-stop control information, the target output power range of the diesel generator, and the energy storage charging and discharging power constraint information.

[0077] S4. Based on the start-stop control information of the diesel generator, the target output power range of the diesel generator, and the energy storage charging and discharging power constraint information, a coordinated scheduling sequence among the photovoltaic, energy storage, and diesel generators is generated under the preset constraint conditions of the photovoltaic-storage-diesel microgrid.

[0078] S5. Control the photovoltaic, energy storage system and diesel generator to perform corresponding power regulation operations according to the coordinated scheduling sequence, and collect the actual operation data in real time during the execution process;

[0079] S6. Compare the actual operating data with the digital twin status information to obtain the status deviation, and correct the coordination and scheduling sequence based on the status deviation.

[0080] As described in steps S1 to S6 above, multi-source operating status data during the operation of the photovoltaic-storage-diesel microgrid are collected, including load operating data, photovoltaic operating data, energy storage operating data, diesel generator operating data, and bus operating data. Among them, load operating data is used to characterize the current and historical changes in electricity demand of the microgrid, photovoltaic operating data is used to reflect the changes in new energy power generation capacity, energy storage operating data is used to characterize the current adjustability and health status of the energy storage system, diesel generator operating data is used to describe the supply capacity and operating status of traditional energy, and bus operating data is used to reflect the power supply stability of the entire microgrid. By performing time synchronization, data cleaning, and state quantity conversion processing on the multi-source operating status data, data deviations between different acquisition devices are eliminated, and key state features that can characterize the operating status of the microgrid are extracted. Then, a digital twin state information is constructed, which includes information such as energy storage charge status, energy storage health status, energy storage temperature, energy storage available capacity, diesel generator output power, start-stop status, cumulative operating time, load power, bus voltage, and bus frequency, so that the state in the digital space can correspond to the actual operating state of the microgrid.

[0081] Furthermore, when constructing digital twin state information, a unified digital twin state vector format is adopted:

[0082]

[0083] in:

[0084]

[0085]

[0086]

[0087]

[0088] In the formula (t) represents the overall digital twin state vector of the microgrid. (t) represents the state sub-vector of the diesel generator. (t) represents the state sub-vector of the energy storage battery. (t) represents the load state sub-vector. (t) represents the state sub-vector of the AC bus. (t) represents the output power of the diesel generator. (t) represents the start / stop state of the diesel engine. (t) represents the cumulative operating hours. (t) represents the energy storage state of charge. (t) represents the energy storage health status. (t) represents the temperature of the energy storage battery. (t) represents the currently available energy for energy storage. (t) and (t) represents the load's active power and reactive power, respectively. (t) (t) and (t) represents the AC bus voltage, AC bus frequency, and system power loss, respectively. The above state vector can simultaneously describe the diesel engine operation, energy storage health, load demand, and bus safety status.

[0089] Based on the load status, photovoltaic operation status, energy storage status, and diesel generator operation status contained in the digital twin status information, the system extracts the status features that affect future changes in energy supply demand and predicts the load change trend, photovoltaic output change trend, energy storage status change trend, and diesel generator backup demand information within the future operating cycle. Among these, the load change trend information is used to determine whether future power supply demand will increase or decrease, the photovoltaic output change trend information is used to determine the available power supply capacity of new energy sources, the energy storage status change trend information is used to determine the future power support capacity that the energy storage system can provide, and the diesel generator backup demand information is used to determine whether traditional energy equipment needs to be put into operation in advance. This obtains energy supply demand information within the future operating cycle, enabling coordinated control to shift from a passive response mode based on the current status to an active adjustment mode based on future change trends.

[0090] The prediction model can adopt a spatiotemporal prediction model based on physical priors, constructing load, photovoltaic, energy storage, diesel engine, and bus into a graph structure:

[0091]

[0092] in, To visualize the network system state, For a set of graph nodes, For the graph edge set, It is a dynamic correlation matrix;

[0093]

[0094] in, Represents the physical prior adjacency matrix. This represents a data-driven dynamic correction matrix. This is the normalization function;

[0095] If a graph attention network is used, the node features are updated as follows:

[0096]

[0097] in, For nodes In the Features of layer +1 For nodes In the Features of the layer The attention coefficient between nodes. To share the feature transformation matrix, For nodes The set of neighbors;

[0098] The forecast output should include at least short-term load forecasts. Photovoltaic power output forecast Forecasting of energy storage SOC change trends and diesel engine standby demand;

[0099] Based on the predicted energy supply and demand information and the current digital twin status information, the operating boundaries of the diesel generator and energy storage system are determined. When the prediction results indicate that the photovoltaic output and energy storage regulation capacity cannot meet future energy demand, the diesel generator's start-up status is determined based on its current operating status, cumulative operating time, and energy gap, and corresponding start-up and shutdown control information is generated. Simultaneously, based on future energy demand, the diesel generator's economic operating range, and its current operating status, the target output power range of the diesel generator is determined, ensuring that the diesel generator operates within the lowest possible unit fuel consumption range, avoiding fuel waste caused by prolonged low-load operation. The diesel generator's fuel consumption can be modeled using a lookup table or piecewise linear interpolation based on measured data, assuming a known set of output points and fuel consumption points. ,when At that time, fuel consumption can be expressed as:

[0100]

[0101] in, ( For diesel engines in terms of power Fuel consumption;

[0102] The thermal efficiency of a diesel engine is calculated as follows:

[0103]

[0104] in, For diesel engine thermal efficiency, For diesel fuel density, The lower heating value of diesel oil. This represents the number of MJ corresponding to the electrical energy output per hour;

[0105] The unit fuel consumption for power generation and the economic operating range are as follows:

[0106]

[0107]

[0108] Where SFC stands for fuel consumption per unit of electricity generated. To achieve the minimum unit fuel consumption, To allow for deviation coefficients, coordinated control prioritizes operating the diesel engine at... Inside, Within the economic operating power range, energy storage undertakes the tasks of short-term fluctuations and smoothing adjustments;

[0109] Based on the energy storage state of charge, energy storage health state, energy storage temperature, and available energy storage capacity, the current allowable charge and discharge capacity of the energy storage system is determined, and energy storage charge and discharge power constraint information is generated. The energy storage health state SOH(t) is defined as:

[0110]

[0111] The currently available energy for energy storage is:

[0112]

[0113] in, This refers to the rated energy capacity of the energy storage.

[0114] To protect aging energy storage, the lower and upper limits of SOC narrow as SOH decreases:

[0115]

[0116]

[0117] in, This represents the lower limit of SOC allowed at the current moment. This represents the lower limit of SOC allowed at the current moment. and These represent the lower and upper limits of the base SOC for the new battery, with α being the SOC lower limit narrowing coefficient. β is the narrowing factor of the upper limit of SOC. This is the narrowing coefficient;

[0118] For example, take , , ,when hour:

[0119]

[0120] That is, when the health status of energy storage declines, the lower limit of SOC is increased and the upper limit of SOC is decreased, thereby reducing deep charging and deep discharging;

[0121] Maximum charge and discharge power of energy storage is dated according to SOH:

[0122]

[0123]

[0124] in:

[0125]

[0126] In the formula, (t) represents the energy storage charging power at scheduling step k. (t) represents the energy storage discharge power at scheduling step k. This is the rated maximum charging power of the energy storage. This is the rated maximum discharge power of the energy storage. This is the power derating factor related to SOH. The lowest power capability factor, The reduction index;

[0127] Energy storage should not be charged and discharged simultaneously, therefore a mutual exclusion constraint is set:

[0128]

[0129] The energy storage SOC update equation is:

[0130]

[0131] in, and These refer to energy storage charging efficiency and energy storage discharging efficiency, respectively. The scheduling time step;

[0132] To ensure that the energy storage system meets energy demand while avoiding overcharging and discharging, high-rate operation, and high-power operation under aging conditions, the system determines whether the diesel generator will participate in power supply during the future scheduling cycle based on the start-stop status of the diesel generator, limits the output capacity of the diesel generator in each scheduling period based on the target output power range, determines the charging and discharging range of the energy storage system in each scheduling period based on the energy storage charging and discharging power constraint information, and arranges the photovoltaic output power, energy storage charging and discharging power, and diesel generator output power in each scheduling period in combination with future energy demand information and photovoltaic output change trend information, forming a coordinated scheduling sequence corresponding to multiple future time nodes;

[0133] In an off-grid photovoltaic-storage-diesel scenario incorporating photovoltaics, the power balance relationship is as follows:

[0134]

[0135] in, (k) represents the available photovoltaic power. For the abandoned light power, and ;

[0136] When there is no photovoltaic power or photovoltaic power is ignored, it can be simplified to:

[0137]

[0138] The diesel engine start / stop and output constraints are as follows:

[0139]

[0140] in, This indicates the start / stop status of the diesel engine. For the minimum stable output of the diesel engine, This is the rated maximum output of the diesel engine;

[0141] The diesel engine gradeability constraint is:

[0142]

[0143] in, The maximum power climbing rate of the diesel engine.

[0144] The minimum start-up time and minimum downtime constraints for diesel engines are as follows:

[0145]

[0146] in, To accumulate continuous running time, To accumulate continuous running time, To minimize runtime, This constraint, set as the minimum downtime, is used to avoid short-cycle start-stop cycles of the diesel engine.

[0147] During the generation process, power balance constraints, energy storage state of charge constraints, energy storage health state constraints, diesel generator operation constraints, and power supply reliability constraints are introduced simultaneously, so that the generated coordinated scheduling sequence can not only meet the energy supply and demand balance, but also meet the equipment safe operation requirements.

[0148] The objectives of coordinated scheduling include fuel cost, start-up and shutdown losses, energy storage lifetime loss, curtailment losses, power supply reliability, and control smoothness. The overall objective function is:

[0149]

[0150] Where J is the overall objective function, To normalize the operating cost target, The weighting coefficient for the operating cost target. To normalize the energy storage lifetime loss target, The weighting coefficient for the energy storage lifetime loss target. To achieve the normalized power supply reliability target, The weighting coefficient for power supply reliability targets. To normalize the control smoothness objective, To control the smoothness target weight coefficient, To normalize the light-abandoning penalty target, To determine the weighting coefficients for the light-abandonment penalty targets, each target item is normalized to avoid direct addition of different dimensions;

[0151] The operating cost target can be written as:

[0152]

[0153] Where N is the total number of steps within the scheduling period, This refers to the unit price of diesel fuel. Cost per startup Unit operating and maintenance cost Cost per unit of energy decay;

[0154] Startup event y DG (k) is:

[0155]

[0156] Indicates a diesel engine start-up event;

[0157] Energy storage lifetime loss can be modeled using empirical models:

[0158]

[0159] in, f is the equivalent of energy storage lifetime degradation.deg The empirical function representing the degradation of energy storage lifetime, C rate (k) represents the energy storage charge / discharge rate at the k-th step;

[0160] If an electrochemical mechanism model is not developed, an interpretable engineering approximation can also be used:

[0161]

[0162] in, For throughput-attenuation coefficient, This is the SOC over-limit attenuation penalty coefficient. For indicator functions, For the set of SOC safe intervals, This is the penalty coefficient for temperature exceeding the limit. The safe temperature threshold for energy storage;

[0163] Power supply reliability targets include reserve capacity, voltage-frequency deviation, and SOC safety margin, which can be written as:

[0164]

[0165] in:

[0166]

[0167] in, The penalty coefficient for insufficient reserves. For the required reserve capacity, Available standby capacity, This is the voltage deviation penalty coefficient. This is the frequency deviation penalty coefficient. This refers to the per-unit voltage deviation. This is the frequency per-unit deviation. This is the bus reference voltage. This is the voltage reference value. The rated frequency;

[0168] The smoothing control term is:

[0169]

[0170] The discarded light item is:

[0171]

[0172] Among them, P curt This refers to the power of discarded light.

[0173] In scheduling layer optimization, the decision variable vector is defined as:

[0174]

[0175] The optimization solution can be obtained by NSGA-II, multi-objective particle swarm optimization, mixed integer programming or weighted rolling optimization. If NSGA-II is used, its role is to generate multi-objective Pareto candidate solutions. The final solution can be selected by fuzzy decision, scene weight or safety priority rule.

[0176] Reinforcement learning can be used to output policy priors, rather than directly replacing all optimization control;

[0177] The reinforcement learning state vector can be defined as:

[0178]

[0179] Reinforcement learning action vectors can be defined as:

[0180]

[0181] in, The reference increment for diesel engine power is 'mode', where 'mode' is the operating mode code.

[0182] The immediate reward value for reinforcement learning is:

[0183]

[0184] Where μ1-μ4 are the reinforcement learning reward weight coefficients. For t-step fuel cost, For the cost of energy storage degradation over t steps, This indicates penalties for violating safety constraints at step t, including exceeding SOC limits, voltage limits, frequency limits, and diesel engine output limits. J represents the baseline fuel consumption. start，base J represents the baseline start-stop cost. deg,base Indicates the reference attenuation. This represents the penalty coefficient for benchmark violations.

[0185] Reinforcement learning can be pre-trained in a digital twin environment and its robustness can be enhanced by domain randomization. After being transferred to a real system, the output of reinforcement learning does not directly bypass security constraints, but is used as an input to the coordination and scheduling module as optimization weights, candidate policies or initial solutions.

[0186] In tactical-level control, model predictive control or equivalent rolling optimization methods are used to handle short-term prediction and execution biases.

[0187] The MPC state vector is:

[0188]

[0189] The MPC control vector is:

[0190]

[0191] The MPC perturbation vector is:

[0192]

[0193] in, Ambient temperature;

[0194] The state-space model is as follows:

[0195]

[0196] in, , , Let b be the MPC state space matrix, and let b be the MPC constant bias vector.

[0197] The MPC output vector is:

[0198]

[0199] in, This can include SOC, power balance deviation, battery temperature, and diesel engine output deviation. , , Output the equation matrix for MPC;

[0200] The optimization objective of MPC is:

[0201]

[0202] Where U represents the control sequence, N is the decision variable that MPC is trying to solve, and N is the decision variable that MPC is trying to solve. p N represents the prediction time domain length. c Controlling the length of the time domain, y(k+i) is the first... The output of the step, y ref (k+i)th The output reference value of the step. Weighted quadratic norm, Δu(k+i) is the first... The control increment of a step, and the change in control quantity between two adjacent steps. It is a weighted matrix controlling the increment, ξ(k+i) represents the soft constraint relaxation variable, and ρ is the relaxation penalty coefficient. The bigger the size, the heavier the punishment;

[0203] MPC executes the first step of the control quantity in each control cycle and re-solves it based on the latest state in the next cycle. This method can maintain the real-time correctability of the scheduling plan when prediction errors exist.

[0204] Photovoltaic power adjustment commands are generated based on the photovoltaic output power in the coordinated scheduling sequence, energy storage converter control commands are generated based on the energy storage charging and discharging power, and diesel generator control commands are generated based on the diesel generator output power. These commands are then sent to the corresponding execution equipment to ensure that each device operates according to the predetermined scheduling plan. Simultaneously, during the execution of the equipment, actual operating data is continuously collected, including actual energy storage state of charge, actual energy storage health status, actual energy storage temperature, actual energy storage available capacity, actual diesel generator output power, actual start-stop status, actual cumulative operating time, actual load power, actual bus voltage, and actual bus frequency, etc., to reflect the actual operating effect after the scheduling strategy is executed.

[0205] At the execution layer, the energy storage converter can employ sliding mode control, current inner loop PI control, model predictive current control, or other fast power point tracking control methods. The following is an implementation of sliding mode control.

[0206] The current loop model of the energy storage converter is as follows:

[0207]

[0208] Where L is the AC side filter inductance of the converter. This is the inverter output voltage. R is the bus voltage, R is the equivalent resistance of the AC side of the converter, and i is the converter output current.

[0209] The current tracking error is:

[0210]

[0211] Among them, I ref A reference quantity indicating the current output;

[0212] Define the sliding surface:

[0213]

[0214] in, The sliding mode area integral coefficient;

[0215] The sliding surface is then:

[0216]

[0217] Let the approach law be:

[0218]

[0219] in, For reaching-law gain; s˙ sliding mode surface The time derivative, s-sliding surface Φ boundary layer thickness, tanh(s / Φ) smoothing treatment, to avoid high-frequency chattering of sign(s);

[0220] The control law can be obtained as follows:

[0221]

[0222] Get the Lyapunov function:

[0223]

[0224] Along the system trajectory:

[0225]

[0226] Where, V˙ is the derivative with respect to time, Chain-law derivative results;

[0227] Therefore, sliding mode variables As the current converges near the boundary layer, the current tracking error tends to decrease. Replacing the sign function can reduce high-frequency chattering in traditional sliding mode control;

[0228] The actual operating status at the corresponding time is updated based on the collected actual operating data, and compared with the corresponding status in the digital twin status information to obtain the deviation of each status parameter;

[0229] The residual between the digital twin state and the actual state is:

[0230]

[0231] in, This is the actual operating state vector;

[0232] Since different state quantities have different dimensions and scales, the normalized residual is further defined as:

[0233]

[0234] The diagonal normalized matrix is:

[0235]

[0236] Among them, P base Power reference value, H base Operating hour baseline value, T base Temperature reference value, E base Energy reference value, P base Power reference value, Q base Reactive power reference value, V base Voltage reference value, f baseFrequency reference value, P base The power reference value, where 1 in the formula represents the switching state, is dimensionless;

[0237] Model calibration or scheduling weight update is triggered when the following conditions are met:

[0238]

[0239] in, To preset the residual threshold, normalized residual processing can avoid the judgment distortion caused by directly adding variables of different dimensions such as power, voltage, and temperature. When the state deviation exceeds the preset deviation threshold, it indicates that there is a mismatch between the current coordinated scheduling sequence and the actual operating environment. At this time, the photovoltaic output power command, energy storage charging and discharging power command and diesel generator output power command are adjusted according to the degree of deviation to generate an updated coordinated scheduling sequence and execute the control process again.

[0240] The deviation between the predicted load and the actual load is monitored, and the load prediction error is defined as:

[0241]

[0242] in, This represents the load forecast error at time t. This represents the actual load power collected at time t. This represents the predicted load power at time t;

[0243] Using exponentially weighted moving average (EWMA) to smooth the error, the prediction error after EWMA smoothing is defined as:

[0244]

[0245] in, This represents the prediction error after EWMA smoothing at time t. This represents the smoothing coefficient, with typical values ​​ranging from 0.1 to 0.3. This represents the original prediction error at time t. This represents the smoothing error at time t-1;

[0246] The revised load forecast values ​​are:

[0247]

[0248] Voltage and frequency are used in per-unit values ​​when controlling constraints:

[0249]

[0250] in, This is the per-unit voltage value. This is the per-unit value of the frequency. This represents the actual measured value of the bus voltage. Indicates the voltage reference value. This indicates the per-unit value of the bus frequency. This represents the actual measured value of the bus frequency. Indicates the frequency reference value;

[0251] This system enables continuous updates of digital twin status information as actual operating conditions change, achieving closed-loop control between prediction, scheduling, execution, and correction. It simultaneously considers the health status of energy storage, the operating status of diesel generators, and the operating status of the bus, providing more comprehensive status inputs and improving state awareness in complex operating environments. Based on digital twin status information, it predicts load change trends, photovoltaic output change trends, energy storage status change trends, and diesel generator reserve demand within future operating cycles. Based on the prediction results, it determines future energy supply and demand information, allowing the control strategy to respond proactively to future supply and demand changes. This avoids the lag problem caused by traditional control methods that only adjust based on the current operating state, improving the proactive adjustment capability of the microgrid during operation. It determines diesel generator start-stop control information based on energy supply and demand information and diesel generator operating status, and combines this with a preset economic operating range to determine the target output power range of the diesel generators, ensuring that the diesel generators operate within the lowest possible fuel consumption per unit of power generation. When load fluctuations are small, the energy storage system can be used to handle short-term power changes, reducing [the impact of power consumption fluctuations]. The frequent starting and low-load operation of diesel generators reduces fuel consumption and mechanical wear. It considers not only the energy storage's state of charge but also its health status, temperature, and available capacity, enabling the energy storage system to dynamically adjust its charging and discharging capabilities based on its own health. When the energy storage's health deteriorates, its charging and discharging power and operating range can be limited, reducing lifespan degradation caused by deep charging and discharging, high-rate operation, and high-temperature operation, thus improving the long-term reliability of the energy storage system. By combining the diesel generator's start-stop status, output power range, and energy storage charging and discharging constraints, a coordinated scheduling sequence is generated while meeting power balance, equipment operation, and power supply reliability constraints. This achieves a reasonable power allocation among photovoltaic, energy storage, and diesel generators, enabling the priority consumption of new energy sources, flexible adjustment of energy storage, and stable diesel engine power supply to form a coordinated operation, improving the overall energy utilization efficiency of the microgrid. It realizes a complete closed-loop control process from state perception, trend prediction, coordinated control to feedback optimization, avoiding the problem of independent prediction and execution in traditional control methods, and improving the overall operational stability and intelligence level of the photovoltaic-storage-diesel microgrid.

[0252] In a preferred embodiment, multi-source operating status data during the operation of the photovoltaic-storage-diesel microgrid is acquired, and digital twin status information is constructed based on the multi-source operating status data, including:

[0253] S101. Acquire multi-source operating status data collected by different devices and monitoring nodes during the operation of the photovoltaic-storage-diesel microgrid. The multi-source operating status data includes load operating data, photovoltaic operating data, energy storage operating data, diesel generator operating data, and bus operating data.

[0254] S102. Preprocess the multi-source operation status data, including time synchronization, data cleaning and state variable transformation, to obtain an operation status dataset under a unified time scale.

[0255] S103. Based on the operating status dataset, extract the equipment status features and electrical status features that characterize the operating status of the photovoltaic-storage-diesel microgrid, and construct digital twin status information based on the equipment status features and electrical status features.

[0256] It should be noted that the digital twin status information includes energy storage state of charge, energy storage health status, energy storage temperature, available energy storage capacity, diesel generator output power, start-stop status, cumulative operating time, load power, bus voltage, and bus frequency.

[0257] As described in steps S101 to S103 above, during the operation of the photovoltaic-storage-diesel microgrid, in order to achieve comprehensive perception of the energy supply, energy conversion, and energy consumption status within the microgrid, it is necessary to acquire multi-source operating status data from different equipment sides and different electrical monitoring nodes. On the load side, by installing smart meters, power acquisition modules, and power quality detection devices in the microgrid's various distribution cabinets, important load power supply circuits, and user sides, the voltage, current, active power, reactive power, and load start-up and shutdown status during load operation are collected in real time. For example, in the application scenario of an island microgrid, the residential load, communication base station load, and production equipment load are connected to smart meters respectively, and the smart meters are used to collect data according to a preset sampling period (e.g., 1 second / time). The system collects operational data from each load node. Residential loads yielded 30kW of active power, communication base station loads yielded 20kW, and production equipment loads yielded 50kW. On the photovoltaic (PV) side, PV operational data is acquired through PV array monitoring devices, PV inverter acquisition modules, and environmental sensors. Specifically, the actual PV power generation capacity is obtained by collecting the DC-side voltage and current of the PV modules, and the current PV power supply is obtained by collecting the AC-side output power of the inverter. Environmental parameters such as irradiance and module temperature are also collected to help determine changes in PV output. For example, if the DC-side output voltage of the PV modules is 600V and the output current is 80A, then the PV output power is calculated to be 48kW. The power data (kW) is used as photovoltaic (PV) operating status data. On the energy storage side, the energy storage battery management system collects energy storage operation data. Specifically, it uses a battery voltage sampling module to collect the voltage at each cell terminal, a Hall current sensor to collect the energy storage charging and discharging current, and a temperature sensor to collect the battery module temperature. Simultaneously, it reads the State of Charge (SOC), State of Health (SOH), and current available capacity calculated by the battery management system. For example, for an energy storage system with a rated capacity of 500 kWh, if the current remaining capacity is 350 kWh, then based on the ratio of remaining capacity to rated capacity, the current SOC is approximately 70%. If, after long-term operation, the maximum available capacity of the battery decreases to 450 kWh, then based on the ratio of the current maximum available capacity to the rated capacity... With a SOH of approximately 90%, on the diesel generator side, operating data is acquired through the diesel generator controller, speed sensor, fuel flow meter, and output power acquisition module. This includes the diesel generator output power, start / stop status, speed, fuel consumption, and cumulative operating time. For example, the diesel generator output power detection device detects a current output power of 120kW, the controller reads that the diesel generator is currently in a starting state, and the operating time recording module obtains a cumulative operating time of 3200 hours. On the bus side, bus operating data is acquired through voltage transformers, current transformers, and a frequency detection module installed at the AC bus. This includes bus voltage, bus current, bus frequency, and bus power balance status.For example, the rated voltage of the AC bus is 400V. The voltage acquisition module detects a real-time bus voltage of 398V, and the frequency detection module acquires a bus frequency of 49.98Hz. Data collected from the load side, photovoltaic side, energy storage side, diesel generator side, and bus side are uploaded to the energy management and control platform according to a unified communication protocol. This allows for the centralized acquisition of operational information from different devices and nodes, forming multi-source operational status data covering the entire process of energy input, energy storage, energy conversion, and energy consumption. However, because different devices in the photovoltaic-storage-diesel microgrid use different types of data acquisition devices with varying sampling periods, data formats, and data transmission methods, the directly acquired multi-source operational status data may have inconsistencies. To address issues such as inconsistencies, abnormal data, and inconsistent data units, preprocessing of the acquired multi-source operational status data is necessary. This includes time synchronization of the multi-source operational status data, using the system time of the microgrid energy management and control platform as a unified time reference. A corresponding timestamp is appended to each set of collected data, and time alignment is performed based on these timestamps. For example, if the load acquisition module collects load power data at 10:00:00, the energy storage management system collects SOC data at 10:00:05, and the diesel generator controller uploads operational status data at 10:00:01, then 10:00:01 is used as the current data cycle, mapping data collected at different times to the same operational moment. Load status, energy storage status, and diesel engine status can represent the microgrid's operating status at the same point in time. Data cleaning is performed on multi-source operating status data to eliminate erroneous data, abnormal data spikes, and communication anomalies during the acquisition process. For example, under normal operating conditions, the AC bus frequency typically remains within the range of 49.5Hz to 50.5Hz. If a bus frequency of 70Hz is acquired, this data is considered outside the normal operating range and marked as abnormal. For data loss due to communication failures, such as the energy storage temperature sensor failing to upload temperature data at a certain point in time, valid temperature data before and after that point in time is used to supplement the data. For example, if the temperature was 35℃ at one moment and 3℃ at another moment... If the temperature is 7℃, then based on the temperature trend, the current temperature is estimated to be approximately 36℃. For duplicate uploaded data, such as two identical sets of diesel generator power data uploaded at the same time, only one set of valid data is retained, and duplicate data is deleted. Status variables generated by different devices are converted to ensure a unified data representation method for data from different sources. For example, the start / stop status output by the diesel generator controller might be represented by "run" / "stop" characters, while the energy storage system status might be represented by digital codes. Therefore, the diesel generator operating status is converted to a unified digital form, such as "run" for 1 and "stop" for 0. Simultaneously, data in different units is converted to a unified data format, such as energy storage capacity being expressed in kWh.Power is expressed in kW, temperature in °C, and voltage in V, enabling different types of data to be stored and retrieved within the same data structure. Through time synchronization, data cleaning, and state variable conversion, time deviations, abnormal errors, and format differences between data from different devices are eliminated, ultimately resulting in an operational status dataset under a unified time scale. Device status features are extracted from this dataset. For energy storage systems, by reading energy storage voltage, current, cumulative charge / discharge, and capacity information, features such as energy storage state of charge (SOC), energy storage health status, energy storage temperature, and available energy storage capacity are extracted. SOC represents the current remaining energy level; for example, for an energy storage system with a rated capacity of 500 kWh, the current remaining energy... If the capacity is 350kWh, then dividing the remaining capacity by the rated capacity yields 0.7, meaning the current State of Charge (SOC) is 70%. The energy storage health status indicates the degree of battery capacity degradation. For example, if the rated capacity is 500kWh and the maximum usable capacity drops to 450kWh after long-term operation, then dividing 450kWh by 500kWh yields 0.9, meaning the State of Health (SOH) is 90%. The energy storage temperature is obtained from multiple temperature monitoring points. For example, if the battery module temperature monitoring values ​​are 34℃, 35℃, and 36℃, then averaging these values ​​gives the current energy storage temperature as approximately 35℃. The usable energy storage capacity is determined based on the current SOH and the rated capacity. For example, if the rated capacity is 500kWh and the SOH is 90%, then the current usable capacity is... The capacity is 500 multiplied by 0.9, resulting in 450kWh. For the diesel generator, equipment status characteristics such as output power, start / stop status, and cumulative operating time are extracted from the diesel generator's operating data. For example, the power acquisition module obtains that the current output power of the diesel generator is 120kW, and the controller reads the current status as start-up, converting the start-up status to a digital status value of 1. Simultaneously, the operating record shows a cumulative operating time of 3200 hours. Electrical status characteristics are extracted from the operating status dataset. These electrical status characteristics mainly reflect the overall power supply demand and power supply stability of the microgrid, including load power, bus voltage, and bus frequency. For example, the power data collected from each load node is summarized... Calculations show that the residential load is 30kW, the communication load is 20kW, and the production load is 50kW, resulting in a total current load power of 100kW for the microgrid. The current bus voltage (e.g., 398V) is directly read via the bus voltage acquisition module, and the current bus frequency (e.g., 49.98Hz) is read via the frequency detection module. After extracting equipment and electrical status characteristics, these are combined according to a preset status structure to construct digital twin status information. This encapsulates the energy storage charge status, energy storage health status, energy storage temperature, available energy storage capacity, diesel generator output power, start / stop status, cumulative operating time, load power, bus voltage, and bus frequency into a unified digital status description representing the current operating status of the photovoltaic-storage-diesel microgrid.At a given operational moment, the digital twin's state information includes: energy storage state of charge 70%, energy storage health 90%, energy storage temperature 35℃, available energy storage capacity 450kWh, diesel generator output power 120kW, diesel generator operating status (starting), cumulative operating time 3200 hours, load power 100kW, bus voltage 398V, and bus frequency 49.98Hz. This constructed digital twin state information can simultaneously reflect the renewable energy supply capacity, energy storage regulation capacity, diesel generator power supply capacity, and electrical stability of the microgrid, improving the operational stability and coordinated control capabilities of the photovoltaic-storage-diesel microgrid under conditions of renewable energy fluctuations, load changes, and equipment aging.

[0258] In a preferred embodiment, based on digital twin state information, the load change trend information, photovoltaic output change trend information, energy storage state change trend information, and diesel generator backup demand information of the photovoltaic-storage-diesel microgrid are obtained during the future operating cycle. The energy supply demand during the future operating cycle is then predicted to obtain the corresponding energy supply and demand information, including:

[0259] S201. Based on the load status information, photovoltaic operation status information, energy storage status information and diesel generator operation status information in the digital twin status information, obtain the corresponding status change characteristics of the photovoltaic-storage-diesel microgrid in the future operation cycle.

[0260] S202. Based on the characteristics of state changes, predict the load change trend, photovoltaic output change trend, energy storage state change trend, and diesel generator standby demand information within the future operating cycle.

[0261] S203. Based on load change trend information, photovoltaic output change trend information, energy storage status change trend information, and diesel generator standby demand information, determine the energy supply demand information for the future operating cycle.

[0262] As described in steps S201 to S203 above, to accurately predict the future energy demand of the photovoltaic-storage-diesel microgrid, it is necessary to determine the future operating cycle corresponding to the prediction target based on the currently constructed digital twin state information, and obtain the possible changes in the state characteristics of each energy device within this future operating cycle. The future operating cycle refers to the subsequent operating time range pre-set according to coordination and control requirements, using the current microgrid operating time as the prediction starting point. This time range describes the changes in load demand, renewable energy output, and equipment operating status of the photovoltaic-storage-diesel microgrid over a future period. The future operating cycle is set based on the control response speed, equipment regulation capability, and scheduling requirements of the photovoltaic-storage-diesel microgrid. In the coordination and scheduling layer, it can be set to a range of 5 to 30 minutes. For example, if the future operating cycle is set to 15 minutes, then the future operating cycle corresponding to the current time 10:00 is from 10:00 to 10:15. After determining the future operating cycle, based on the load state information, photovoltaic operating state information, energy storage state information, and diesel generator operating state information in the digital twin state information, the state changes within the future operating cycle are analyzed. The system analyzes the situation, acquiring current load power, historical load change data, and load operation time characteristics from load status information. By statistically analyzing load power changes over continuous time periods, it obtains status change characteristics such as load change rate, load growth or decline trend, and load fluctuation range. For example, if the microgrid load power collected at 10:00 AM is 100kW, and load data from the past 15 minutes is also acquired (90kW at 9:45 AM, 94kW at 9:50 AM, 97kW at 9:55 AM, and 100kW at 10:00 AM), analyzing continuous load data reveals a continuously increasing load characteristic and predicts that the load may continue to increase in future operating cycles. Secondly, based on photovoltaic (PV) operation status information, it acquires current PV output power, historical PV power change data, and environmental impact parameters. By analyzing the change in PV output power over time, it obtains PV output change characteristics. For example, if the current PV output power is 50kW, and it decreased from 70kW to 50kW in the past 15 minutes, while the detected light intensity decreased from 850W / m², it further analyzes the changes in PV output power over time. 2 Decreased to 650W / m 2Then, the characteristics of the photovoltaic power output decline are extracted and used as an important basis for future changes in photovoltaic power supply. Next, based on the energy storage status information, the energy storage state of charge, energy storage health status, energy storage temperature, current charging / discharging power, and available capacity information are obtained. Combined with historical energy storage operation trajectories, the characteristics of energy storage status changes are obtained. For example, if the current energy storage SOC is 70%, and the SOC has decreased from 75% to 70% in the past 15 minutes, it indicates that the energy storage is in a continuous discharging state. Simultaneously, if the current SOH is 90% and the battery temperature is 35℃, it is determined that the energy storage is in normal operation and can continue to provide power regulation capabilities. Based on the diesel generator operating status information, the current start / stop status, output power, cumulative operating time, and available backup power information of the diesel generator are obtained. For example, if a diesel generator is currently shut down with a rated power of 200kW and a cumulative operating time of 3200 hours, then by combining future load growth trends, photovoltaic power decline trends, and the remaining regulation capacity of energy storage, the characteristics of diesel generator standby demand changes are extracted. These characteristics are then comprehensively compiled from load change features, photovoltaic output change features, energy storage status change features, and diesel generator standby status features to form the status change characteristics within the future operating cycle. For example, it might predict that in the next 15 minutes, the load will increase by 20kW, photovoltaic power will decrease by 10kW, the energy storage SOC will decrease by 3%, and the diesel generator will need 40kW of standby capacity. The prediction process is based on current digital twin status information, historical operating data, and equipment operating patterns. Currently, load change trend information is predicted. Based on the current load status characteristics, historical load change data matching the current operating scenario is obtained from the historical operation database. For example, load operation data of the same date type, the same time period, and similar environmental conditions within the past 30 days are obtained, and historical load change patterns are analyzed. For instance, if the average load power increased from 100kW to 120kW between 10:00 and 10:15 in the past 30 working days, then combined with the current load growth characteristics, it is predicted that the load power may increase from 100kW to 120kW in the future operating cycle, and load increase trend information is generated. This predicts the photovoltaic output change trend information based on the current photovoltaic operating status information. By analyzing power output, environmental parameters, and historical photovoltaic (PV) operation data, future PV output capacity is determined. For example, if the current PV output power is 50kW, and historically, under similar weather conditions, the PV output power typically decreases by 10kW within the next 15 minutes, while the current solar irradiance continues to decline, then the PV output power is predicted to decrease from 50kW to 40kW within the next operating cycle, generating a trend information on the decline in PV output. This information is then used to predict the trend information on changes in the energy storage state. Based on the current energy storage SOC, SOH, temperature, charging and discharging power, and future load demand forecasts, the changes in the energy storage state within the next operating cycle are calculated. For example, if the current energy storage SOC is 70%, the current energy storage discharge power is 50kW, and continuous discharge is expected within the next 15 minutes...The estimated power consumption is calculated based on the current discharge power and duration to obtain the future SOC trend. Simultaneously, the energy storage capacity decay is assessed based on the SOH status. For example, if the SOH is 90%, the actual usable capacity of the current 500kWh energy storage system is approximately 450kWh. The future energy storage output capacity is adjusted based on the available capacity limit. Diesel generator backup demand information is predicted. Based on the predicted future load demand, photovoltaic output, and energy storage adjustability, potential future energy supply gaps are calculated. For example, if the predicted future load demand is 120kW, the expected photovoltaic output is 40kW, and the maximum allowable energy storage output is 50kW, then the combined photovoltaic and energy storage capacity can provide 90kW of energy supply, which is less than the 120kW load demand. There is a 30kW power supply gap compared to the demand, therefore, diesel generators need to provide at least 30kW of output capacity. Considering power supply reliability requirements, reserve capacity is increased; for example, a 20% reserve margin would add an additional 6kW of reserve demand, resulting in a final diesel generator reserve demand of approximately 36kW. Through a forecasting process, information on load change trends, photovoltaic output change trends, energy storage status change trends, and diesel generator reserve demand is obtained for the future operating cycle. Based on the predicted load power change results, the required load power supply level for each future time point is determined. For example, if the current load power is 100kW, and the forecast indicates the load will gradually increase to 120kW within the next 15 minutes, then the 120kW... W represents the maximum energy demand during the future operating cycle. Based on the trend of photovoltaic (PV) output changes, the future energy supply capacity provided by new energy sources is determined. The predicted PV output power is used as the new energy supply. For example, if the current PV output power is 50kW, and it is predicted to drop to 40kW in the next 15 minutes, then the PV power supply during the future operating cycle is approximately 40kW. If PV output fluctuates, the upper and lower limits of PV output are further determined based on the predicted range. For example, if the predicted PV output range is 35kW to 45kW, this range is used as the new energy supply capacity range. The energy supply capacity that the energy storage system can participate in regulating is determined based on the trend of energy storage status changes. The energy storage's current energy supply capacity is judged based on its SOC, SOH, temperature, and available capacity. The allowable charge and discharge capacity is determined by factors such as the energy storage rated capacity (e.g., 500kWh, SOH 90%), current available capacity (approximately 450kWh), current SOC (State of Charge) of 70%, and temperature of 35°C, all within the normal operating range. Therefore, the energy storage can participate in future regulation. The maximum allowable discharge power determines the power the energy storage can provide; for example, a maximum allowable output power of 50kW indicates a future regulation capacity of 50kW. The power supply task required by the diesel generator is determined based on backup demand information. For instance, based on a future load demand of 120kW, photovoltaic supply of 40kW, and energy storage supply of 50kW, a 30kW power supply gap is calculated. Simultaneously, a 6kW backup demand is added considering backup capacity requirements.This determines that the diesel generator needs to have a backup capacity of at least 36kW. By comprehensively analyzing load demand, photovoltaic power supply capacity, energy storage regulation capacity, and diesel generator backup capacity, energy supply and demand information for the future operating cycle is formed. For example, the energy supply and demand information for the next 15 minutes includes a load demand of 120kW, an expected photovoltaic power supply of 40kW, energy storage allowing to provide 50kW of regulation power, and the diesel generator needing to provide a backup capacity of at least 36kW. This energy supply and demand information can reflect the energy supply and demand status of the microgrid during the future operating cycle, reducing problems such as sudden diesel generator starts, excessive energy storage discharge, and insufficient utilization of new energy sources, thereby improving the economic efficiency and power supply reliability of the photovoltaic-storage-diesel microgrid.

[0263] In a preferred embodiment, based on energy supply and demand information and digital twin status information, diesel generator start-stop control information, diesel generator target output power range, and energy storage charging and discharging power constraint information are determined, including:

[0264] S301. Obtain load demand change information, photovoltaic output change information, energy storage status change information and standby capacity demand information from the energy supply demand information, and extract energy storage status information and diesel generator operating status information from the digital twin status information.

[0265] S302. Based on load demand change information, photovoltaic output change information, energy storage status change information and reserve capacity demand information, determine whether the photovoltaic output and energy storage regulation capacity meet the microgrid's energy supply demand in the current operating cycle.

[0266] S303. When the photovoltaic output and energy storage regulation capacity cannot meet the energy supply demand, determine the start-stop conditions of the diesel generator based on the diesel generator operating status information, and generate the corresponding diesel generator start-stop control information.

[0267] S304. Based on the energy demand information, diesel generator operating status information and preset diesel generator economic operating range, determine the target output power range of the diesel generator, so that the output power of the diesel generator is kept in the operating range with low unit power generation fuel consumption.

[0268] S305. Based on the energy storage state of charge, energy storage health status, energy storage temperature and available energy storage capacity in the energy storage status information, determine the current allowable charging and discharging capacity of the energy storage system.

[0269] S306. Generate energy storage charging and discharging power constraint information based on the current allowable charging and discharging capacity and energy supply demand information of the energy storage system.

[0270] As described in steps S301 to S306 above, information on load demand changes, photovoltaic output changes, energy storage status changes, and reserve capacity demand for the future operating cycle are extracted from the energy supply and demand information. Load demand change information characterizes the changes in the electrical energy consumed by the user side or equipment side over a future period. For example, if it is predicted that the load power will gradually increase from the current 180kW to 230kW within the next 15 minutes, this change process is used as load demand change information. Photovoltaic output change information characterizes the changes in photovoltaic power generation capacity within the future cycle. For example, based on historical irradiance, cloud cover changes, and the current photovoltaic operating status, it is predicted that the photovoltaic output power will increase from 120kW to 230kW within the next 15 minutes. When the kW drops to 80kW, corresponding photovoltaic output change information is generated. Energy storage status change information is used to characterize the future adjustment capability that the energy storage system can provide. For example, based on the current energy storage state of charge (SOC) of 65%, health status of 92%, and temperature of 28℃, if it is predicted that the SOC will decrease to 50% at the end of the future cycle, then the SOC change trend is used as the energy storage status change information. Reserve capacity demand information is used to characterize the adjustable capacity that needs to be reserved to ensure power supply reliability. For example, if the current predicted maximum load is 250kW and the maximum possible photovoltaic output power is 100kW, then at least 150kW of adjustable power supply capacity is needed as reserve capacity. Simultaneously, the current energy storage status information is extracted from the digital twin status information. The system inputs information on the diesel generator's operating status, including energy storage status information such as current state of charge (SBC), health status, temperature, and available capacity. For example, if the current SBC is 65%, the health status is 92%, and the rated capacity is 500 kWh, the available capacity is approximately 460 kWh (500 kWh multiplied by 92%). The diesel generator's operating status information includes its current start / stop status, output power, cumulative operating time, and operating efficiency. For instance, if the diesel generator is currently running with an output power of 100 kW and a cumulative operating time of 3200 hours, this information serves as the diesel generator's status input, allowing the system to simultaneously monitor future energy demand changes and the current adjustable capacity of the equipment. This system aims to avoid problems such as frequent diesel engine start-stops or overuse of energy storage caused by controlling the system solely based on the current load status. It determines the total energy supply demand to be met in future operating cycles based on load demand changes. For example, if the load power is predicted to increase from 100kW to 120kW within the next 15 minutes, then at least 120kW of power supply capacity is required in the current cycle. Secondly, it determines the effective power supply that photovoltaic (PV) systems can provide in future operating cycles based on changes in PV output. For example, if the PV output power is predicted to decrease from 50kW to 40kW, then the PV system can provide 40kW of power supply capacity. Finally, it determines the current regulation capacity of energy storage based on changes in energy storage status and assesses whether the energy storage's state of charge (SOC) is within the allowable discharge range.For example, if the safe operating range of the energy storage's State of Charge (SOC) is set to 20%–90%, and the current SOC is 70%, then the discharge conditions are met. Next, it's determined whether the energy storage's State of Health (SOH) meets the power output requirements. For instance, if the SOH is 90%, higher than the preset health threshold of 80%, then the current output power is allowed. Then, it's determined whether the energy storage temperature is within a safe range. For example, if the normal operating temperature range for energy storage is 0℃–45℃, and the current temperature is 35℃, then the temperature is considered normal. Finally, the adjustable capacity is determined based on the maximum allowable discharge power of the energy storage. For example, if the current maximum discharge power of the energy storage is 50kW, then the energy storage can provide 50kW of power. Alternatively, if the future load demand is 120kW, and the photovoltaic output is expected to be 40kW, then the maximum allowable output of the energy storage is 50kW. If W represents the load demand, then the total power supply capacity of photovoltaic (PV) and energy storage is 90kW, while the load demand is 120kW, a difference of 30kW. Therefore, it is determined that the PV output and energy storage regulation capacity cannot meet the power supply demand, and diesel generators are needed to supplement the power supply. Alternatively, if the future load demand is 100kW, the PV output is 60kW, and the energy storage is allowed to output 50kW, then the total power supply capacity of PV and energy storage reaches 110kW, which is greater than the load demand of 100kW and also meets the reserve capacity requirement. In this case, it is determined that PV and energy storage can meet the power supply demand, and the diesel generator does not need to be started. The participation of the diesel generator in power supply can be dynamically determined based on the future power supply gap, realizing the priority use of new energy sources and diesel engine auxiliary power supply. The power supply gap information is obtained based on the judgment result. For example, if the predicted load demand in the next 15 minutes is 400kW, the expected output of photovoltaics is 100kW, and the maximum allowable output of energy storage is 150kW, then the power supply gap is 150kW. The operating status information of the diesel generator is obtained, including its current start / stop status, cumulative operating time, last shutdown time, and current output power. For instance, if the diesel generator is currently shut down, and more than 2 hours have passed since its last shutdown, and it meets the minimum shutdown time requirement specified by the equipment, then it meets the start-up conditions. The control status of the diesel generator is determined according to preset start / stop judgment conditions. These start-up conditions may include a power supply gap exceeding the maximum regulation capacity of energy storage within the predicted future period, a predicted energy storage SOC below the safety lower limit, insufficient system reserve capacity, or insufficient photovoltaic output. The shutdown conditions for a continuous decline in power output can include the recovery of photovoltaic (PV) output and the energy storage SOC reaching a safe recovery range, a decrease in load demand, and the diesel generator already meeting the minimum operating time requirement. For example, if the minimum operating value of the energy storage SOC is set at 25%, and it is predicted that the SOC will drop to 22% in 30 minutes, while the PV output continues to decline, then it is determined that the diesel generator needs to be started. After determining the start conditions, diesel generator start-stop control information is generated, such as a control command to "start diesel generator No. 1 and keep it running for no less than 60 minutes." If the diesel generator is already running and the power demand decreases, a maintain-run or shutdown control command is generated, ensuring that the diesel generator only participates in power supply when necessary, and reducing ineffective start-stop operations in conjunction with equipment operating constraints.To extend the service life of diesel engines, obtain operating status information of diesel generators and fuel consumption data provided by the manufacturer or on-site test data. For example, test a 200kW rated power diesel generator to obtain fuel consumption corresponding to different output powers: 12L / h at 50kW output power, 18L / h at 100kW output power, 24L / h at 150kW output power, and 34L / h at 200kW output power. Further calculate the unit fuel consumption for power generation, that is, the amount of fuel consumed to generate 1kWh of electricity. For example, at a 100kW output, if 100kWh of electricity is generated per hour and 18L of fuel is consumed, then the unit fuel consumption for power generation is 0.18L / kWh. At the start, the generator produces 150 kWh of electricity per hour, consuming 24 L of fuel. Therefore, the fuel consumption per unit of power generation is approximately 0.16 L / kWh. The 150 kW mark falls within the economic operating range. The economic operating range for the diesel engine is set to 60%–85% of its rated power, i.e., 120 kW–170 kW. When future load demands require the diesel engine to supplement 120 kW of power, the target output power of the diesel engine is set to 120 kW. When the energy supply gap reaches 180 kW, the output power of the energy storage section can be adjusted to keep the diesel engine operating below 170 kW. This prevents fuel waste caused by low-load operation, reduces mechanical carbon buildup and maintenance stress, improves fuel efficiency, and achieves the State of Charge (SOC) of the energy storage. For example, the current SOC... With an OC of 70%, the State of Health (SOH) of the energy storage is obtained. For example, if the current SOH is 90%, it means the actual capacity of the energy storage has dropped to 90% of the rated capacity. The energy storage temperature is also obtained; for example, if the current temperature is 30℃, it is within the optimal operating range. The available capacity of the energy storage is obtained; for example, if the rated capacity is 500kWh, the current available capacity is calculated to be 450kWh based on an SOH of 90%. Charge and discharge limits are determined based on the state. When the SOC is high, for example, exceeding 85%, the maximum charging power is reduced to avoid overcharging. When the SOC is low, for example, below 25%, the discharge power is limited to avoid deep discharge. When the SOH decreases, for example, below 80%, the maximum allowable charge and discharge power is derated; for example, the rated maximum discharge power is reduced to 2... The initial power output is 50kW, which is then adjusted to 180kW based on the energy storage system's health status. When the temperature exceeds 45℃, the power is further reduced, for example, to 50% of the rated power. By comprehensively considering SOC, SOH, temperature, and available capacity information, the current permissible charge and discharge capabilities of the energy storage system are determined. For example, the current maximum charging power is set at 100kW, the maximum discharging power at 180kW, and the SOC operating range is determined to be 30%–85%. This enables adaptive adjustment of the energy storage operating state, improving energy storage safety and extending battery life. Permissible operating parameters for energy storage are obtained, such as a maximum charging power of 100kW, a maximum discharging power of 180kW, and a safe SOC operating range of 30%–85%. Combined with future energy demand information, the direction of energy storage participation in regulation is determined.When the projected future photovoltaic output exceeds the load demand by 50kW, energy storage is needed to absorb the excess energy. This generates a charging power constraint, limiting the energy storage charging power to no more than 100kW. When the future load demand exceeds the photovoltaic power supply capacity by 150kW, energy storage needs to release energy. This generates a discharging power constraint, limiting the energy storage discharging power to no more than 180kW. SOC constraints are set based on future SOC trends. For example, if the current SOC is 80%, continuing to charge at 100kW will exceed 90% in 30 minutes. In this case, the charging power is reduced to 50kW to maintain the SOC. Within 85%, energy storage charging and discharging power constraints are generated, including maximum charging power limits, maximum discharging power limits, SOC upper and lower limits, and charging / discharging state constraints. For example, the generated constraints might be: maximum charging power ≤ 100kW, maximum discharging power ≤ 180kW, SOC maintained between 30% and 85%, and simultaneous charging and discharging prohibited. This ensures that the coordinated scheduling sequence can fully utilize the energy storage's regulation capabilities while preventing a decrease in energy storage lifespan due to overcharging, over-discharging, or high-rate operation, providing a safety boundary for the coordinated operation of photovoltaics, energy storage, and diesel generators.

[0271] In a preferred embodiment, based on diesel generator start-stop control information, the target output power range of the diesel generator, and energy storage charging and discharging power constraint information, a coordinated scheduling sequence among the photovoltaic, energy storage, and diesel generators is generated under the preset constraints of the photovoltaic-storage-diesel microgrid, including:

[0272] S401. Based on the diesel generator start-stop control information, determine the corresponding operating status of the diesel generator in the future scheduling cycle, and determine the diesel generator power output range in each scheduling period based on the target output power range of the diesel generator.

[0273] S402. Based on the energy storage charging and discharging power constraint information, determine the charging power range and discharging power range of the energy storage system in each scheduling period within the future scheduling cycle, and determine the power adjustment range of the energy storage system in conjunction with the energy supply demand information.

[0274] S403. Based on energy demand information, photovoltaic output change trend information, diesel generator power output range and energy storage system power adjustment range, determine the photovoltaic output power, energy storage charging and discharging power and diesel generator output power corresponding to each scheduling period in the future scheduling cycle.

[0275] S404. Based on the photovoltaic output power, energy storage charging and discharging power and diesel generator output power corresponding to each scheduling period, construct an equipment operation scheduling plan for the future scheduling cycle;

[0276] S405. When the equipment operation scheduling scheme meets the preset constraints of the photovoltaic-storage-diesel microgrid, the equipment operation scheduling scheme shall be used as the coordinated scheduling sequence. The preset constraints include power balance constraints, energy storage state of charge constraints, energy storage health state constraints, diesel generator operation constraints, and power supply reliability constraints.

[0277] As described in steps S401 to S405 above, it is necessary to determine the future scheduling period. The future scheduling period represents the time range covered by the scheduling plan generated by the coordinated control. It is determined based on the load change rate of the photovoltaic-storage-diesel microgrid, the response characteristics of the diesel generator, and the energy storage regulation capability. For example, for off-grid microgrids such as those in islands and mining areas, since load changes are usually periodic and the diesel generator needs to run continuously for a long time after startup, the future scheduling period can be set to 30 minutes to 2 hours. In one implementation, by collecting historical load data for 30 consecutive days, the time intervals in which the load power change exceeds 10% of the rated load are statistically analyzed. When the statistics show that the load change is mainly concentrated between 15 minutes and 40 minutes... In this case, a 30-minute scheduling cycle can be used as the future scheduling period, which can be divided into multiple scheduling time periods. For example, the 30-minute scheduling cycle can be divided into 6 consecutive scheduling time periods, each lasting 5 minutes. This allows for the determination of the diesel generator's operating status within each 5-minute time period. When determining the diesel generator's operating status, the status configuration is based on the start, maintain operation, or stop commands in the diesel generator's start-stop control information. For example, if the diesel generator is currently in a stopped state, and based on energy supply and demand information, it is determined that photovoltaic output will decrease within the next 30 minutes, and the energy storage SOC will decrease from 40% to below 25%, then it is considered that photovoltaic and energy storage cannot continuously meet the load demand. In this case, the diesel generator's start-stop control information is in the start state, and it is not... Within the scheduling cycle, all corresponding scheduling periods are set to the running state. If the diesel generator is already running, but a future load reduction and photovoltaic recovery are predicted, and the diesel generator has already met the preset minimum running time (e.g., continuous operation for 60 minutes), then subsequent scheduling periods can be set to the stopped state. After determining the diesel generator's running state, the power output range corresponding to each scheduling period is determined based on the diesel generator's target output power range. The target output power range is derived from the diesel generator's economic operating range, which can be obtained through fuel consumption curves provided by the manufacturer or on-site test data. For example, testing a 200kW rated power diesel generator and recording output power at 50kW, 100kW, and 1... The fuel consumption at 50kW and 200kW was calculated. The fuel consumption per unit of output electrical energy was determined to be lowest around 150kW. Therefore, 120kW to 170kW was set as the economical operating range. When a predicted energy shortage of 130kW is encountered during a certain scheduling period, the diesel generator output power range is set to 120kW to 130kW. When the energy shortage reaches 220kW, since the maximum economical operating power of the diesel generator is 170kW, the diesel generator output power is limited to between 120kW and 170kW, and the remaining power is supplemented by the energy storage system. This allows the diesel generator to plan its operating status in advance according to energy demand and always prioritizes operating within the more fuel-efficient range.To avoid fuel waste and equipment wear caused by frequent diesel engine start-stop and low-load operation in traditional control methods; to obtain energy storage charging and discharging power constraint information, including maximum allowable charging power, maximum allowable discharging power, and SOC operating range. This information is determined comprehensively by the energy storage's state of charge, health status, temperature, and available capacity. For example, an energy storage system with a rated capacity of 500kWh and a rated charging and discharging power of 250kW, currently has an SOC of 70%, a SOH of 90%, and a temperature of 30℃. After derating the power capacity based on the SOH decrease, the current maximum charging power is determined to be 120kW, and the maximum discharging power to be 180kW. Simultaneously, the safe operating range of SOC is set to 30% to 85%. When determining the energy storage power range for each scheduling period within a scheduling cycle, the judgment is based on the energy supply and demand status of each scheduling period. If the predicted output power of photovoltaics exceeds the load demand in a certain scheduling period (e.g., photovoltaic output of 200kW, load demand of 120kW), there is a surplus of 80kW of renewable energy. At this time, energy storage enters charging mode, with its charging power range set from 0kW to 120kW. If the load demand increases in another scheduling period (e.g., load demand of 350kW, photovoltaic output of 100kW), there is a 250kW power supply gap. Since the maximum discharge power of energy storage is 180kW, the discharge power range of energy storage is set from 0kW to 180kW, and the remaining 70kW is handled by diesel generators. The actual energy storage capacity is determined based on the energy supply and demand information. The power regulation range is determined by the following: For example, if the load demand during a future scheduling period is 300kW, the photovoltaic output is 80kW, and the diesel generator outputs 150kW according to the economic range, there will still be a 70kW power supply gap. Therefore, the discharge power of the energy storage system is determined to be 70kW. If the calculated required energy storage output is 220kW, but the maximum discharge capacity of the energy storage is only 180kW, then the energy storage output will be limited to 180kW, and the output power of the diesel generator will be adjusted. This allows the energy storage system to fully consider its own health status and operating boundaries when participating in coordinated control, enabling the energy storage to undertake regulation tasks within a reasonable range, improving the energy storage life protection capability, and simultaneously improving the efficiency of new energy utilization. This also allows for the acquisition of load demand information and photovoltaic output forecasts for each scheduling period within the future scheduling cycle. Information, such as dividing the next 30 minutes into 6 scheduling periods, with the load demand in the first period being 300kW and the predicted photovoltaic output being 150kW, the load demand in the second period being 320kW and the predicted photovoltaic output being 120kW, and the load demand in the third period being 350kW and the predicted photovoltaic output being 80kW, is used to determine the photovoltaic output power first, since photovoltaic power has zero fuel cost. For example, in the first period, 150kW photovoltaic output is directly used for power supply. The diesel engine output power is then determined based on the remaining power demand and the diesel generator's power range. For example, if the remaining load demand in the first period is 150kW, and the economical operating range of the diesel generator is 120kW to 170kW, then the diesel generator output is set to 150kW.Energy storage does not need to participate in discharging. In the third time period, the load demand is 350kW, the photovoltaic output is 80kW, leaving a remaining demand of 270kW. The maximum economical output of the diesel generator is 170kW, so the diesel generator outputs 170kW, and the energy storage outputs 100kW to supplement it. If the photovoltaic output exceeds the load demand in a certain time period, for example, 180kW for photovoltaic output and 100kW for load demand, the excess 80kW is preferentially allocated to the energy storage for charging. The photovoltaic output power, energy storage charging / discharging power, and diesel generator output power are determined separately for each dispatch period, enabling the three types of energy equipment to operate in a complementary relationship. This avoids a single device bearing the entire regulation task, improves photovoltaic absorption capacity, reduces diesel fuel consumption, and minimizes excessive energy storage usage. The process involves establishing a time series corresponding to future scheduling cycles. For example, the next 30 minutes are divided into six 5-minute scheduling periods. Then, the equipment control parameters for each scheduling period are recorded, including photovoltaic output power, energy storage charging or discharging status, energy storage power value, diesel generator start / stop status, and diesel generator output power. For instance, the first scheduling period corresponds to a photovoltaic output of 150kW, energy storage output of 0kW, and diesel generator output of 150kW; the second scheduling period corresponds to a photovoltaic output of 120kW, energy storage discharge of 50kW, and diesel generator output of 150kW; the third scheduling period corresponds to a photovoltaic output of 80kW, energy storage discharge of 100kW, and diesel generator output of 170kW. The data from multiple time points are then processed according to time... By combining these parameters sequentially, a continuous equipment operation scheduling scheme is formed for the future scheduling cycle. This scheme includes not only power commands for each device but also changes in the start / stop status of diesel generators, changes in the charging and discharging direction of energy storage, and the utilization of photovoltaic power. This allows for a description of the entire operation process of the photovoltaic-storage-diesel microgrid throughout the future cycle. By constructing the equipment operation scheduling scheme, the predicted and optimized control results can be converted into specific equipment execution commands, and preset constraints can be obtained. These constraints are derived from equipment technical parameters, operational safety standards, and historical stable operation data. Among these constraints, the power balance constraint originates from the microgrid's power supply demand, meaning that the sum of the photovoltaic output power, diesel generator output power, and energy storage discharge power must meet a negative balance requirement during any scheduling period. For example, if the load demand is 300kW, the photovoltaic output is 100kW, the diesel engine output is 150kW, and the energy storage output is 50kW, then the total power supply is 300kW, which meets the power balance requirements. The energy storage state of charge (SOC) constraint comes from the battery's safe operating range. For example, based on the characteristics of lithium iron phosphate batteries, the SOC operating range is set to 30% to 85%. If the SOC is predicted to remain at 50% after the dispatch plan is executed, the constraint is met. If the predicted SOC drops to 20%, the energy storage output needs to be readjusted. The energy storage health status constraint is used to limit the operating intensity of aging batteries. For example, when the SOH drops below 80%, the maximum charge and discharge power is reduced to 70% of the rated power. The diesel generator operating constraint comes from the equipment's rated parameters.This includes minimum output power, maximum output power, minimum operating time, and ramp-up capability. For example, a diesel engine with a rated power of 200kW has a stable operating range of 50kW to 200kW. Therefore, the output power in the dispatching scheme cannot exceed this range. Power supply reliability constraints stem from bus operation requirements, such as maintaining the AC bus voltage at 380V ± 5% and the frequency at 50Hz ± 0.5Hz. If the equipment operation dispatching scheme meets all constraints, it is used as the final coordinated dispatching sequence. If any constraints are not met, such as the energy storage SOC being below the safe range or the diesel engine output exceeding the limit, the power allocation of photovoltaic, energy storage, and diesel generators is readjusted until all constraints are met. This achieves a closed loop of coordinated dispatching scheme generation and safety verification, improving the operational reliability of the photovoltaic-storage-diesel microgrid and preventing power outages, equipment damage, or energy waste due to unreasonable dispatching strategies.

[0278] In a preferred embodiment, the photovoltaic system, energy storage system, and diesel generator are controlled to perform corresponding power regulation operations according to a coordinated scheduling sequence, and real-time operating data during the execution process is collected, including:

[0279] S501. Generate photovoltaic power adjustment command based on photovoltaic output power, generate energy storage converter control command based on energy storage charging and discharging power, and generate diesel generator control command based on diesel generator output power.

[0280] S502: Send photovoltaic power regulation commands, energy storage converter control commands, and diesel generator control commands to the corresponding execution devices to control the photovoltaic system, energy storage system, and diesel generator to perform power regulation according to the coordinated scheduling sequence;

[0281] S503. During the power regulation operation, collect the actual operating data of the photovoltaic system, energy storage system, diesel generator and bus side. The actual operating data includes the actual energy storage state of charge, actual energy storage health status, actual energy storage temperature, actual energy storage available capacity, actual diesel generator output power, actual start-stop status, actual cumulative running time, actual load power, actual bus voltage and actual bus frequency.

[0282] As described in steps S501 to S503 above, the coordinated scheduling sequence contains information such as the target output power of the photovoltaic system, the target charging and discharging power of the energy storage system, and the target output power of the diesel generator for multiple future scheduling periods. The scheduling target for the corresponding time period is obtained according to the current control cycle. For example, if the current scheduling cycle is 15 minutes, and the coordinated scheduling sequence determines the target output power of the photovoltaic system to be 120kW, the discharge power of the energy storage system to be 50kW, and the output power of the diesel generator to be 150kW, then the target power is converted into equipment-side commands. For the photovoltaic system, the actual operating power of the photovoltaic array, the maximum output capacity of the inverter, the current ambient irradiance, and the component temperature are obtained. The target photovoltaic output power in the coordinated scheduling sequence is then compared with... The current actual output power is compared to obtain the power regulation requirement. For example, if the current actual photovoltaic output power is 80kW, but the dispatch sequence requires an output of 120kW, then an additional 40kW of output power is calculated. Based on the photovoltaic inverter's maximum power point tracking range, DC-side voltage state, and AC-side bus voltage state, the increased power requirement is converted into the inverter's active power setpoint, power factor control parameters, or maximum power point tracking adjustment parameters, generating the corresponding photovoltaic power regulation command. For the energy storage system, the energy storage converter control command is determined based on the energy storage charging and discharging power in the coordinated dispatch sequence. The current state of charge (SOC), state of health (SOH), battery temperature, and current charging / discharging direction of the energy storage are obtained. When the dispatch sequence requires the energy storage to discharge 5... At 0kW, it is determined whether the current SOC of the energy storage meets the discharge requirements. For example, if the current SOC is 65%, the dynamic allowable operating range is 20% to 90%, and the maximum allowable discharge power is 100kW, then the 50kW discharge demand meets the constraints, and a discharge control command for the energy storage converter is generated. If the current energy storage SOC is 25%, and considering the minimum allowable SOC after the SOH decreases to 30%, then the discharge power is reduced or the discharge command is canceled to avoid over-discharge of the energy storage. The energy storage converter control command typically includes power direction control information, target active power value, current reference value, and charging / discharging mode switching information. For example, the target discharge power of 50kW corresponds to the output discharge mode, current reference value, and power limit parameters. For diesel generators, according to coordination... The target output power and start / stop status in the scheduling sequence generate control commands for the diesel generator. The current state of the diesel generator is determined; for example, if it is currently shut down, but the coordinated scheduling sequence requires the generator to output 150kW, a start command is generated based on the minimum start time, start delay time, and reserve capacity requirements. After the generator starts, the fuel supply and speed control parameters are adjusted according to the target output power range to gradually reach the target value. For example, if the rated power of the diesel generator is 200kW, and the economic operating range is 50%–80% of the rated power, the target output power is controlled within the range of 100kW–160kW, rather than directly increasing it to 200kW, to reduce unit fuel consumption.The system converts the equipment operation plans output from the coordination and scheduling layer into execution commands that can be recognized by the control systems of photovoltaic inverters, energy storage converters, and diesel generators. This achieves effective integration between scheduling strategies and equipment control, avoiding control deviations caused by scheduling results that cannot be directly executed, and improving the operational stability of the photovoltaic-storage-diesel microgrid. Control command transmission paths are established based on equipment communication addresses. For example, the photovoltaic system connects to the photovoltaic inverter controller via RS485, CAN communication, or industrial Ethernet; the energy storage system communicates with the energy storage converter through the energy storage management system; and the diesel generator connects to the microgrid energy management controller through the unit controller. Control commands are also converted according to equipment type, such as converting photovoltaic power adjustment commands into inverter-recognizable commands. The active power setpoint is used to convert the energy storage inverter control commands into charging and discharging current reference values, and the diesel generator control commands into start-up control signals, speed regulation signals, and power output target values. For example, when the coordinated dispatch sequence requires the photovoltaic system to provide 120kW of power, the energy storage system to discharge 50kW, and the diesel generator to output 150kW in the current period, a 120kW power target value is sent to the photovoltaic inverter. The inverter adjusts its power output according to the current DC input power, and at the same time, a 50kW discharge command is sent to the energy storage inverter, causing the energy storage system to release 50kW of power to the AC bus. For the diesel generator, start-up or operation-holding commands are sent according to the start-stop status, and the fuel supply is adjusted through the speed control system to enable the diesel engine to output power. When the power reaches 150kW, the adjustment is not completed all at once, but rather in stages. For example, when the diesel generator is increased from 50kW to 150kW, to avoid mechanical shock, the output power is increased by 20kW per minute according to the diesel engine's ramp-up constraint, gradually reaching the target power over 5 minutes. The energy storage system then quickly adjusts its output power based on the converter's response speed to compensate for the power gap during the diesel engine's adjustment process. Simultaneously, it continuously monitors changes in bus voltage and frequency. When a low bus voltage or frequency drop is detected, the energy storage output power can be temporarily corrected based on the real-time status to ensure stable power supply. By combining dispatch commands with the equipment execution process, the interaction between photovoltaic, energy storage, and diesel generators is achieved. The coordinated operation of the system prioritizes the use of renewable energy by photovoltaics, with energy storage playing a rapid power balancing role and diesel generators ensuring stable power supply. This reduces the inefficient operation time of diesel engines, minimizes unreasonable charging and discharging of energy storage, and improves the overall energy efficiency of the microgrid. Corresponding data acquisition devices are configured according to the equipment type. For energy storage systems, the energy storage management system collects data such as individual battery voltage, battery pack current, battery temperature, SOC, SOH, and remaining usable capacity. SOC is calculated by the energy storage management system based on historical charge / discharge volumes, battery rated capacity, and the current integral result. For example, for an energy storage system with a rated capacity of 500 kWh and a current remaining capacity of 350 kWh, the SOC is approximately 70%.SOH (State of Charge) is calculated by comparing the current available capacity with the initial rated capacity. For example, if the current maximum chargeable capacity is 450 kWh and the initial capacity is 500 kWh, then the SOH is approximately 90%. For diesel generators, the unit controller collects data on engine speed, output power, operating status, number of starts, cumulative operating time, and fuel consumption. For example, if the current output power of the diesel engine is 150 kW, the cumulative operating time is 3200 hours, and the fuel flow rate is 35 L / h, this data is used as the diesel generator's operating status data input into the digital twin status information. For photovoltaic systems, the inverter collects data on the DC input power, AC output power, module temperature, and inverter operating status of the photovoltaic array. For example, if the current actual output power of the photovoltaic array is... If the power output is 118kW and the module temperature is 45℃, this data is recorded as photovoltaic operation data. For the bus side, AC bus voltage, frequency, active power, and reactive power are collected using voltage sensors, current sensors, and a frequency detection module. For example, if the bus voltage is 400V and the frequency is 50Hz, this data serves as the overall microgrid operating status data. The collected data is aggregated according to a unified timestamp; for example, fast electrical parameters are collected every 100 milliseconds, equipment status parameters every second, and health status parameters are updated every minute. This ensures that data from different sampling periods are matched to a unified time scale. By collecting real-time operational data, the actual operating status after the coordinated scheduling sequence is obtained, and the actual status is fed back to the digital twin. In this model, the digital twin model refers to a virtual mapping model constructed in the digital space based on IoT data acquisition, state perception, data processing, and dynamic update technologies, corresponding to the actual operating state of the photovoltaic-storage-diesel microgrid. By collecting real-time operating data from the photovoltaic system, energy storage system, diesel generator, and bus side, it digitally describes the equipment status, electrical status, and energy status of the actual physical system, and achieves synchronous updates between the virtual model and the actual system. In this scheme, the digital twin model collects operating data such as photovoltaic output power, energy storage voltage, current, state of charge (SOC), state of health (SOH), temperature, available capacity, diesel generator output power, start / stop status, cumulative operating time, load power, bus voltage, and frequency. The data is processed through time synchronization, anomaly removal, and state transition to form a unified state dataset. Based on the processed state data, equipment operation characteristics and electrical operation characteristics are extracted to construct a digital twin state information system encompassing energy storage status, diesel generator status, load status, and bus status. For example, with a rated energy storage capacity of 500 kWh and a state of health (SOH) of 92%, the digital twin model can calculate the current available capacity to be approximately 460 kWh. Combined with the current state of charge (SOC), the actual available energy storage capacity is determined. During diesel generator operation, corresponding operating status descriptions are generated based on output power, start / stop status, and cumulative operating time, ensuring that the digital twin state continuously reflects the actual operation of the photovoltaic-storage-diesel microgrid.This process provides an accurate data foundation for state deviation calculation, model updates, and coordinated scheduling optimization. Simultaneously, it can promptly detect equipment anomalies, such as deviations from expected energy storage SOC, insufficient diesel engine output, and bus frequency fluctuations, allowing for proactive adjustments to control strategies and improving system operational safety and reliability.

[0283] In a preferred embodiment, the actual operating data is compared with the digital twin state information to obtain the state deviation, and the coordination and scheduling sequence is corrected based on the state deviation, including:

[0284] S601. Extract the actual operating status information at the corresponding moment based on the actual operating data collected during the execution of the coordination and scheduling sequence;

[0285] S602. Compare the actual operating status information with the digital twin status information to obtain the status deviation between the actual operating status and the digital twin status.

[0286] S603. Obtain the deviation type and degree corresponding to the state deviation, and determine whether the photovoltaic output power, energy storage charging and discharging power and diesel generator output power in the coordinated scheduling sequence need to be adjusted.

[0287] S604. When the state deviation exceeds the preset deviation threshold, the photovoltaic output power command, energy storage charging and discharging power command and diesel generator output power command in the coordinated scheduling sequence are corrected according to the state deviation to obtain the updated coordinated scheduling sequence.

[0288] S605. Control the photovoltaic, energy storage system and diesel generator to perform power regulation operations according to the updated coordinated scheduling sequence, and continue to update the digital twin status information using the actual operating data after execution.

[0289] As described in steps S601 to S605 above, the time identifier corresponding to the current execution time of the coordinated scheduling sequence is obtained. For example, if the coordinated scheduling sequence is generated according to a 15-minute cycle, and the current time is 10:15, the time window for verifying the current operating status is determined to be from 10:15 to 10:30. Based on this time window, the actual operating data within the corresponding time range is extracted from the data acquisition database. This includes the actual SOC, actual SOH, battery temperature, and actual available capacity collected by the energy storage system; the actual output power, actual start-stop status, and cumulative operating time collected by the diesel generator; the actual output power collected by the photovoltaic system; and the actual voltage, actual frequency, and actual load power collected by the bus side. For example, at 10:15, the data acquisition... The data acquisition module obtained the following data: Energy Storage SOC at 68%, energy storage temperature at 32℃, actual usable energy storage capacity at 450kWh, actual diesel generator output power at 140kW (operating), photovoltaic output power at 115kW, bus voltage at 398V, and bus frequency at 49.98Hz. Since the data came from different acquisition nodes, it was further filtered based on a unified timestamp. For example, data collected between 10:15:00 and 10:15:01 was selected as the current status data. For missing data, interpolation between adjacent time points was used to fill in the gaps. For instance, diesel generator power data was missing at 10:15:00, while power data collected at 10:14:59 was 138kW and power data collected at 10:15:01 was 140kWh. If the output power is 2kW, then the average of the two, 140kW, is taken as the actual output power at 10:15. After time matching, the actual collected data is reorganized into actual operating status information according to the data structure corresponding to the digital twin status information. This includes actual energy storage status information, actual diesel generator status information, actual renewable energy output status information, and actual bus operating status information. For example, the actual energy storage SOC, temperature, and available capacity are combined to form the actual energy storage operating status; the diesel generator output power, start-stop status, and cumulative operating time are combined to form the actual diesel generator operating status; the actual photovoltaic output power is used to form the actual renewable energy operating status; and the bus voltage and frequency are used to form the actual electrical status. This allows for the conversion of discretely collected equipment operating data. To ensure consistency between the actual operating status information and the digital twin's status information structure, and to avoid errors in status judgment due to inconsistent data time and format, thereby improving the accuracy of digital twin feedback correction, a correspondence is established according to the type of state variable. Each parameter in the actual operating status is matched one-to-one with the predicted state parameters in the digital twin's status information. For example, the digital twin's status information records a predicted energy storage SOC of 70% at the current moment, while the actual operating status information shows an energy storage SOC of 68%; the digital twin's status information shows a predicted diesel generator output power of 150kW, while the actual acquired output power is 140kW; and the digital twin's status information shows a predicted bus frequency of 50Hz, while the actual acquired frequency is 49.98Hz.The degree of difference between each state parameter was calculated. For the energy storage SOC state, the SOC state deviation was obtained by comparing the actual SOC with the digital twin SOC; that is, the difference between the actual value of 68% and the predicted value of 70% was 2 percentage points. For the diesel generator output power state, the power deviation of 10kW was obtained by comparing the actual output power of 140kW with the predicted output power of 150kW. For the bus frequency state, the frequency deviation of 0.02Hz was obtained by comparing the actual frequency of 49.98Hz with the target frequency of 50Hz. Since the units of different state parameters are different (e.g., power is in kW, voltage is in V, temperature is in °C, and SOC is in percentage), further scaling was performed on each state deviation. Standardization is performed to obtain the allowable variation ranges for each state parameter. For example, the allowable variation range for diesel generator power is 0–200kW, for energy storage SOC is 0–100%, and for bus frequency is 49.5Hz–50.5Hz. Then, the actual deviations are compared with the corresponding allowable ranges. For instance, a diesel engine power deviation of 10kW represents 5% of the maximum power range of 200kW, an energy storage SOC deviation of 2% represents 2% of the SOC range of 100%, and a frequency deviation of 0.02Hz represents 2% of the allowable frequency variation range of 1Hz. Based on the standardized deviation values ​​of each state variable, a set of state deviations for the current moment is formed. This avoids the influence of a single large numerical variable on the overall judgment result, ensuring that power, voltage, and frequency are considered correctly. Different types of state variables, such as rate and temperature, can be compared under a unified standard, thereby accurately determining the operational differences between the digital twin model and the real microgrid. The type of deviation is determined based on the equipment type to which the state variable belongs. For example, when the actual energy storage SOC is detected to be lower than the digital twin's predicted value, this deviation is marked as an energy storage state deviation; when the actual output power of the diesel generator is detected to be lower than the planned output power, this deviation is marked as a diesel generator execution deviation; when the bus frequency is detected to be lower than the target frequency, this deviation is marked as an electrical stability deviation. For instance, if the coordinated dispatch sequence requires energy storage to discharge 50kW, and the predicted SOC drops from 70% to 65% after 10 minutes of execution, but the actual detected SOC drops to 62%, then the energy storage energy consumption rate is determined. A speed exceeding the predicted rate results in an energy storage state deviation. The degree of deviation is determined based on its magnitude. For example, state deviation levels can be set, including three levels: normal deviation, concern deviation, and severe deviation. The deviation levels are derived from historical operating data statistics and the safe operating range of the equipment. Actual operating data of the microgrid is collected after more than 30 days of continuous operation. The fluctuation range of each state variable under normal operating conditions is statistically analyzed. A fluctuation exceeding the normal fluctuation range by 1 times the standard range is defined as a concern deviation, and a fluctuation exceeding 2 times the standard range is defined as a severe deviation. For example, based on historical data statistics, the normal power tracking error of diesel generators is usually less than 5kW. Therefore, within 5kW is defined as a normal deviation, 5kW to 15kW as a concern deviation, and greater than 15kW as a severe deviation.After determining the type and degree of deviation, the next step is to assess whether the power of equipment in the coordinated scheduling sequence needs adjustment. The actual state of the equipment corresponding to the state deviation is compared with the current scheduling target. For example, if a diesel generator is planned to output 150kW but actually outputs 148kW (a deviation of only 2kW), it is considered that the diesel engine still meets the scheduling requirements and no adjustment is needed. If the planned output is 150kW but the actual output is 120kW (a deviation of 30kW), it indicates that the diesel engine cannot execute according to the scheduling sequence and requires increasing the diesel engine output or adding energy storage discharge compensation. For energy storage systems, when the SOC deviation is small and the bus voltage and frequency remain stable, no adjustment is needed for the energy storage power. However, when the SOC decline rate is significantly higher than the predicted value, leading to SOC deviation in future cycles... When the risk of C falling below the safety lower limit, it is necessary to reduce the discharge power or start the diesel engine earlier. This can identify the operating deviations caused by different equipment and avoid frequent adjustments to the control strategy due to short-term small fluctuations, thus improving the stability of the coordinated control process. The preset deviation threshold is determined based on the equipment's safe operation requirements, historical operating fluctuation range, and control accuracy requirements. Long-term operating data of the photovoltaic-storage-diesel microgrid is collected, such as data from 90 consecutive days of operation. The deviation range of each parameter under normal operating conditions is statistically analyzed. For example, the energy storage SOC prediction error is usually within ±3%, the diesel engine power tracking error is usually within ±5kW, and the bus frequency fluctuation is usually less than 0.05Hz. The deviation threshold is set according to the safety margin. For example, the SOC deviation threshold is set to 5%, and the diesel engine power... The deviation threshold is set to 10kW, and the bus frequency deviation threshold is set to 0.1Hz. When a state deviation is detected to exceed the corresponding threshold, a coordinated scheduling sequence correction is initiated. For example, the original coordinated scheduling sequence requires 120kW output from photovoltaics, 50kW from energy storage, and 150kW output from diesel generators within the next 15 minutes. However, during actual operation, it is found that photovoltaics can only output 80kW due to cloud cover, resulting in a 40kW new energy power gap. At the same time, the energy storage SOC decreases faster than predicted. Therefore, power is redistributed according to the current state, and the power gap to be compensated is determined based on the state deviation. For example, if photovoltaics decreases by 40kW, then energy storage or diesel generators need to compensate by 40kW. It is also determined whether the current SOC of energy storage allows for increased discharge. With an OC of 35%, a minimum dynamic SOC of 30%, and a maximum allowable increase in discharge power of 30kW, the energy storage system will increase output by 30kW, and the remaining 10kW will be compensated by the diesel generator. A new equipment operation plan will be generated, for example, adjusted to 80kW photovoltaic output, 80kW energy storage discharge, and 160kW diesel generator output. Power balance, SOC constraints, diesel generator operating range, and power supply reliability constraints will be rechecked. Once the new equipment operation plan meets all constraints, it will be used as the updated coordinated scheduling sequence. Through a state deviation feedback correction mechanism, the coordinated scheduling process can be transformed from unidirectional planned control to closed-loop dynamic control, improving the system's adaptability to photovoltaic fluctuations, load changes, and equipment state changes.The updated coordination and scheduling sequence is analyzed. For example, if the new scheduling result requires 80kW photovoltaic output, 80kW energy storage discharge, and 160kW diesel generator output, corresponding equipment control commands are generated. The photovoltaic system adjusts the inverter output according to the new target power, the energy storage system adjusts the inverter current direction and output power according to the new discharge power, and the diesel generator adjusts the fuel supply and speed according to the new output power target. After the equipment executes the new scheduling command, actual operating data continues to be collected, such as new energy storage SOC, diesel engine output power, photovoltaic output power, and bus voltage and frequency data. The new actual operating data is input into the digital twin state update process. During the digital twin state update, the latest collected data replaces the changed state parameters. For example, if the original digital twin state had an energy storage SOC of 7... If the actual SOC is 68% after data collection and updating, then the energy storage SOC will be updated to 68%. The equipment change trend will be updated based on continuous operating data. For example, if the actual output power of the diesel engine is 150kW, 155kW, and 160kW for three consecutive cycles, the diesel engine operating status change trend will be updated. A new digital twin status information system will be reconstructed, including energy storage status, diesel engine status, photovoltaic status, and bus status. Through continuous collection, comparison, correction, and updating, the digital twin model will always remain synchronized with the actual operating status of the photovoltaic-storage-diesel microgrid, achieving virtual-real closed-loop control. This reduces the cumulative error caused by long-term operation of the prediction model, improves the accuracy of the scheduling strategy, and automatically adjusts the control strategy when equipment ages, the environment changes, or the load pattern changes, ensuring the long-term safe, efficient, and stable operation of the photovoltaic-storage-diesel microgrid.

[0290] Please see the appendix Figure 2 As shown, the present invention also provides a photovoltaic-storage-diesel coordinated control system based on digital twin drive, used in the above-mentioned photovoltaic-storage-diesel coordinated control method based on digital twin drive, comprising:

[0291] The twin state module is used to acquire multi-source operating state data during the operation of the photovoltaic-storage-diesel microgrid, and to construct digital twin state information based on the multi-source operating state data.

[0292] The supply and demand module, based on digital twin status information, obtains information on load change trends, photovoltaic output change trends, energy storage status change trends, and diesel generator standby demand information of the photovoltaic-storage-diesel microgrid in the future operating cycle, and predicts the energy supply demand in the future operating cycle to obtain the corresponding energy supply and demand information.

[0293] The control classification module is used to determine the start-stop control information of the diesel generator, the target output power range of the diesel generator, and the energy storage charging and discharging power constraint information based on energy supply and demand information and digital twin status information.

[0294] The coordination and scheduling module generates a coordinated scheduling sequence between photovoltaic, energy storage system and diesel generator based on diesel generator start-stop control information, diesel generator target output power range and energy storage charging and discharging power constraint information, under the preset constraint conditions of photovoltaic-storage-diesel microgrid.

[0295] The actual operation module is used to control the photovoltaic, energy storage system and diesel generator to perform corresponding power regulation operations according to the coordinated scheduling sequence, and to collect the actual operation data in real time during the execution process;

[0296] The scheduling correction module is used to compare the actual running data with the digital twin status information, obtain the status deviation, and correct the coordinated scheduling sequence based on the status deviation.

[0297] As described above, the twin state module acquires multi-source operating status data through acquisition devices deployed on the photovoltaic array side, energy storage converter side, energy storage battery management system side, diesel generator control side, and AC bus side. Specifically, the photovoltaic side acquires data such as photovoltaic module output power, voltage, current, irradiance, and ambient temperature to reflect the new energy generation status; the energy storage side acquires data such as battery voltage, current, temperature, state of charge (SOC), state of health (SOH), and available capacity to reflect the remaining regulation capacity and lifespan of the energy storage system; the diesel generator side acquires output power, speed, start-stop status, cumulative operating time, and fuel consumption status to reflect the current operating capacity of the diesel engine; and the bus side acquires bus voltage, frequency, and load power to reflect the micro-level operating status. The system synchronizes the overall power supply status of the power grid, removes abnormal data, and transforms state variables. It converts data from different sampling periods and formats into a unified time-scale dataset and further extracts energy storage health characteristics, equipment operation characteristics, and electrical state characteristics. This forms a digital twin state information system containing information such as energy storage state of charge, energy storage health status, energy storage temperature, available energy storage capacity, diesel generator output power, start / stop status, cumulative operating time, load power, bus voltage, and bus frequency. For example, when the energy storage system collects data showing a battery voltage of 720V, current of 150A, temperature of 28℃, SOC of 65%, and SOH of 92%, and the diesel generator collects data showing an output power of 80kW and is in operation... The system is in a state with a cumulative operating time of 3200 hours. State parameters are synchronously mapped to a digital twin space, enabling the digital twin state information to reflect the real-time operating status of the equipment. The supply and demand module uses the digital twin state information as input to predict energy supply trends within future operating cycles. The prediction cycle is determined based on the current time, historical operating data, and equipment status, for example, using a prediction time window of 15 minutes, 30 minutes, or 1 hour. Based on load power change records, historical photovoltaic output change records, energy storage SOC change trends, and diesel generator standby capacity information in the digital twin state information, potential changing state characteristics within future cycles are extracted. For example, if the load power changes from 120kW to 120kW within the past hour... If the power output of photovoltaic (PV) power is increased to 160kW, while the historical output power for the same period decreases from 100kW to 40kW, the system determines that there may be a power supply gap in the future. Based on the current SOC (State of Charge) of the energy storage (SOC) at 45%, the maximum discharge power at 150kW, and the SOH (State of Harshness) status, the system assesses the continuous support capacity that the energy storage can provide. Combining this with the current shutdown status, start-up time, and available capacity of the diesel generator, the system determines future reserve requirements. By comprehensively analyzing load change trends, PV output change trends, energy storage status change trends, and diesel generator reserve requirements, the system obtains energy supply and demand information for the future period. This avoids the problem of traditional control methods that adjust based solely on the current power status. The control classification module uses energy supply and demand information and digital twin status information...The energy supply capacity of different energy units is classified and assessed, and the operating control parameters of diesel generators and energy storage systems are determined. Based on the changes in load demand and photovoltaic output in future energy demand information, it is determined whether the current power supply of photovoltaics and the adjustable capacity of energy storage can meet the load demand. When photovoltaics and energy storage can meet the energy demand, the diesel generator remains in a shutdown state, and the fluctuations in new energy sources are balanced through the charging and discharging of energy storage. When the forecast results show that photovoltaics and energy storage cannot meet the load demand, the start-up and shutdown control information of the diesel generator is determined based on the operating status, minimum start-up time, and standby requirements. For example, if the predicted load demand in the next 15 minutes is 300kW, the photovoltaic output is expected to be 80kW, and the maximum sustainable discharge of energy storage is... With a power capacity of 120kW, the photovoltaic and energy storage systems can provide a maximum of 200kW of power, leaving a 100kW power gap. At this point, the system generates diesel generator start-up control information. Simultaneously, it determines the target output power range based on the diesel generator's economic operating range. For example, if the diesel generator's fuel consumption is lowest between 40% and 80% of its rated power, the system controls the diesel engine's output power to remain within this range. The system dynamically determines the maximum charging and discharging power of the energy storage system based on its SOC, SOH, and temperature status, preventing overcharging and discharging under low health or high temperature conditions. This achieves coordination between the economical operation of the diesel generator and the protection of the energy storage's lifespan. The coordination and scheduling module uses the diesel generator start-up and shutdown control information, the target output power range, and the energy storage charging and discharging... Power constraint information is used to generate a coordinated dispatch sequence under the condition of meeting the microgrid operation constraints. Multiple dispatch periods are divided according to the future dispatch cycle; for example, using a 5-minute dispatch time step, the next hour is divided into 12 dispatch periods. Within each dispatch period, the photovoltaic output power, energy storage charging and discharging power, and diesel generator output power are determined. During the dispatch process, power balance constraints must be met simultaneously: the sum of photovoltaic output power, diesel generator output power, and energy storage discharging power should meet the load demand, while the energy storage charging power cannot exceed the surplus power of new energy sources. State of Charge (SOC) constraints must be met; for example, the energy storage SOC should be maintained within the range of 20% to 90%. State of Harm (SOH) constraints must also be met; for example, the maximum charging and discharging power is reduced when the SOH decreases. The system satisfies the operating constraints of diesel generators, such as limiting the number of start-stop cycles, minimum operating time, and maximum ramp speed, and the power supply reliability constraints, such as maintaining a certain reserve capacity. By selecting equipment operating combinations that meet these constraints, a coordinated scheduling sequence is formed for future cycles. For example, if the predicted load for a certain scheduling cycle is 250kW, the photovoltaic output is 100kW, and the energy storage is allowed to discharge 80kW, then a diesel generator output of 70kW is sufficient to meet the power demand. Simultaneously, the diesel engine is kept within its economic operating range, preventing a single device from bearing the entire load change and achieving a reasonable division of labor among photovoltaics, energy storage, and diesel engines. The actual operation module converts the corresponding control objectives into executable control commands for the equipment based on the scheduling sequence generated by the coordinated scheduling module.The system sends commands to the photovoltaic inverter, energy storage converter, and diesel generator control unit. For example, when the coordinated scheduling sequence requires the photovoltaic output power to remain at 100kW, the system generates a power adjustment command for the photovoltaic inverter and sends it to the photovoltaic controller via the communication interface, causing the inverter to adjust its power output. When the energy storage output is required to be 50kW, the system generates a control command for the energy storage converter based on the charging and discharging direction, controlling the change in battery-side current. When the diesel generator output is required to be 120kW, the system sends a power adjustment command to the diesel generator speed control unit, adjusting the diesel engine fuel supply and speed to the target state. During execution, the system continuously collects actual operating data, including energy storage SOC, SOH, temperature, available capacity, diesel engine output power, start / stop status, cumulative operating time, load power, bus voltage, and frequency, to determine the actual execution effect of the equipment and realize the conversion of the scheduling strategy into actual equipment control actions. The scheduling correction module performs closed-loop correction of the coordinated scheduling process by comparing the actual operating data with the digital twin state information, matching the actual operating data according to the collection time to form the actual operating state at the time corresponding to the digital twin state information, and adjusting the actual state accordingly. The system compares the actual state with the digital twin state item by item. For example, if the digital twin predicts a bus voltage of 400V while the actual collected voltage is 392V, an 8V voltage deviation occurs. If the digital twin predicts a storage SOC of 60% while the actual SOC is 57%, a 3% SOC deviation occurs. Further analysis of the deviation type determines whether the scheduling strategy needs adjustment. For instance, if the load prediction deviation causes the storage SOC to drop faster than expected, or if the bus frequency deviates from the allowable range, the current coordinated scheduling sequence is considered insufficient to meet actual operational needs, requiring readjustment of photovoltaic output power, storage charging / discharging power, or diesel generator output power. When the state deviation exceeds preset thresholds, such as a voltage deviation exceeding 5% of the rated voltage, a SOC deviation exceeding 5%, or a power deviation exceeding 10% of the rated capacity, a new scheduling instruction is generated based on the deviation direction, and the coordinated scheduling sequence is updated. Simultaneously, the corrected operational results are fed back to the digital twin state module to update the digital twin state information, ensuring the digital twin model continuously approximates the actual operating state. This achieves a dynamic closed loop between prediction, scheduling, and execution, improving the system's adaptability to load changes, energy fluctuations, and equipment state changes.

[0298] And, a photovoltaic-storage-diesel coordinated control terminal based on digital twin technology, including:

[0299] One or more processors;

[0300] A storage device on which one or more programs are stored;

[0301] When one or more programs are executed by one or more processors, the one or more processors implement a digital twin-driven light-storage-diesel coordinated control method.

[0302] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.

Claims

1. A coordinated control method for light, energy storage, and diesel fuel based on digital twin-driven technology, characterized in that: include: Acquire multi-source operating status data during the operation of the photovoltaic-storage-diesel microgrid, and construct digital twin status information based on the multi-source operating status data; Based on digital twin status information, the load change trend information, photovoltaic output change trend information, energy storage status change trend information and diesel generator backup demand information of the photovoltaic-storage-diesel microgrid in the future operation cycle are obtained, and the energy supply demand in the future operation cycle is predicted to obtain the corresponding energy supply demand information. Based on energy supply and demand information and digital twin status information, determine diesel generator start-stop control information, diesel generator target output power range and energy storage charging and discharging power constraint information; Based on the start-stop control information of the diesel generator, the target output power range of the diesel generator, and the charging and discharging power constraint information of the energy storage, a coordinated scheduling sequence among the photovoltaic, energy storage, and diesel generators is generated under the preset constraint conditions of the photovoltaic-storage-diesel microgrid. The photovoltaic, energy storage system and diesel generator are controlled to perform corresponding power regulation operations according to the coordinated scheduling sequence, and the actual operation data during the execution process is collected in real time. The actual operating data is compared with the digital twin state information to obtain the state deviation, and the coordination and scheduling sequence is corrected based on the state deviation.

2. The light-storage-diesel coordinated control method based on digital twin drive according to claim 1, characterized in that, Acquire multi-source operational status data during the operation of the photovoltaic-storage-diesel microgrid, and construct digital twin status information based on the multi-source operational status data, including: The system acquires multi-source operational status data collected from different devices and monitoring nodes during the operation of the photovoltaic-storage-diesel microgrid. This multi-source operational status data includes load operation data, photovoltaic operation data, energy storage operation data, diesel generator operation data, and bus operation data. The multi-source operational status data is preprocessed, including time synchronization, data cleaning and state variable transformation, to obtain an operational status dataset with a unified time scale. Based on the operating status dataset, we extract equipment status features and electrical status features that characterize the operating status of the photovoltaic-storage-diesel microgrid, and construct digital twin status information based on these features.

3. The light-storage-diesel coordinated control method based on digital twin drive according to claim 1, characterized in that, The digital twin status information includes energy storage state of charge, energy storage health status, energy storage temperature, available energy storage capacity, diesel generator output power, start-stop status, cumulative operating time, load power, bus voltage, and bus frequency.

4. The coordinated control method for light, energy storage, and diesel fuel based on digital twin drive according to claim 1, characterized in that, Based on digital twin status information, the system acquires load change trends, photovoltaic output change trends, energy storage status change trends, and diesel generator backup demand information for the photovoltaic-storage-diesel microgrid during future operating cycles. This information is used to predict energy supply demand during future operating cycles, yielding corresponding energy supply and demand information, including: Based on the load status information, photovoltaic operation status information, energy storage status information and diesel generator operation status information in the digital twin status information, the status change characteristics of the photovoltaic-storage-diesel microgrid in the future operating week are obtained. Based on the characteristics of status changes, predict the load change trend, photovoltaic output change trend, energy storage status change trend, and diesel generator standby demand information for future operating cycles. Based on load change trend information, photovoltaic power output change trend information, energy storage status change trend information, and diesel generator standby demand information, determine the energy supply demand information for the future operating cycle.

5. The light-storage-diesel coordinated control method based on digital twin drive according to claim 1, characterized in that, Based on energy supply and demand information and digital twin status information, the start-stop control information for diesel generators, the target output power range for diesel generators, and the charging and discharging power constraints for energy storage are determined, including: The system acquires information on load demand changes, photovoltaic output changes, energy storage status changes, and reserve capacity demand from the energy supply demand information, and extracts energy storage status information and diesel generator operating status information from the digital twin status information. Based on information on load demand changes, photovoltaic output changes, energy storage status changes, and reserve capacity demand, determine whether the photovoltaic output and energy storage regulation capabilities meet the microgrid's energy supply needs during the current operating cycle. When the photovoltaic output and energy storage regulation capacity cannot meet the energy supply demand, the start-up and shutdown conditions of the diesel generator are determined based on the diesel generator operating status information, and corresponding diesel generator start-up and shutdown control information is generated. Based on energy demand information, diesel generator operating status information and preset diesel generator economic operating range, determine the target output power range of the diesel generator, so that the output power of the diesel generator is kept in the operating range with low unit power generation fuel consumption. Based on the energy storage status information, including energy storage state of charge, energy storage health status, energy storage temperature, and available energy storage capacity, determine the current allowable charging and discharging capacity of the energy storage system. Based on the current allowable charging and discharging capacity of the energy storage system and the energy supply demand information, energy storage charging and discharging power constraint information is generated.

6. The light-storage-diesel coordinated control method based on digital twin drive according to claim 1, characterized in that, Based on diesel generator start-stop control information, the target output power range of the diesel generator, and energy storage charging and discharging power constraints, a coordinated scheduling sequence among the photovoltaic, energy storage, and diesel generators is generated, under the preset constraints of the photovoltaic-storage-diesel microgrid, including: Based on the diesel generator start-stop control information, determine the corresponding operating status of the diesel generator in the future scheduling cycle, and determine the diesel generator power output range in each scheduling period based on the target output power range of the diesel generator. Based on the energy storage charging and discharging power constraint information, the charging power range and discharging power range of the energy storage system in each scheduling period within the future scheduling cycle are determined, and the power adjustment range of the energy storage system is determined in combination with the energy supply demand information. Based on energy demand information, photovoltaic output change trend information, diesel generator power output range and energy storage system power adjustment range, determine the photovoltaic output power, energy storage charging and discharging power and diesel generator output power corresponding to each scheduling period in the future scheduling cycle. Based on the photovoltaic output power, energy storage charging and discharging power and diesel generator output power corresponding to each scheduling period, construct an equipment operation scheduling plan for the future scheduling cycle; When the equipment operation scheduling scheme meets the preset constraints of the photovoltaic-storage-diesel microgrid, the equipment operation scheduling scheme will be used as a coordinated scheduling sequence. The preset constraints include power balance constraints, energy storage state of charge constraints, energy storage health state constraints, diesel generator operation constraints, and power supply reliability constraints.

7. The light-storage-diesel coordinated control method based on digital twin drive according to claim 1, characterized in that, The system controls the photovoltaic, energy storage, and diesel generators to perform corresponding power regulation operations according to the coordinated scheduling sequence, and collects real-time operating data during the execution process, including: Photovoltaic power regulation commands are generated based on photovoltaic output power, energy storage converter control commands are generated based on energy storage charging and discharging power, and diesel generator control commands are generated based on diesel generator output power. The photovoltaic power regulation command, energy storage converter control command and diesel generator control command are sent to the corresponding execution equipment respectively, and the photovoltaic system, energy storage system and diesel generator are controlled to regulate power according to the coordinated scheduling sequence. During the power regulation operation, actual operating data of the photovoltaic system, energy storage system, diesel generator and bus side are collected. The actual operating data includes actual energy storage state of charge, actual energy storage health status, actual energy storage temperature, actual energy storage available capacity, actual diesel generator output power, actual start-stop status, actual cumulative running time, actual load power, actual bus voltage and actual bus frequency.

8. The light-storage-diesel coordinated control method based on digital twin drive according to claim 1, characterized in that, The actual operational data is compared with the digital twin state information to obtain the state deviation, and the coordination and scheduling sequence is corrected based on the state deviation, including: Extract the actual operating status information at the corresponding moment from the actual operating data collected during the execution of the coordinated scheduling sequence; The actual operating status information is compared with the digital twin status information to obtain the status deviation between the actual operating status and the digital twin status. Obtain the type and degree of deviation corresponding to the state deviation, and determine whether the output power of photovoltaic, energy storage charging and discharging power and diesel generator output power in the coordinated scheduling sequence need to be adjusted; When the state deviation exceeds the preset deviation threshold, the photovoltaic output power command, energy storage charging and discharging power command and diesel generator output power command in the coordinated scheduling sequence are corrected according to the state deviation to obtain the updated coordinated scheduling sequence. The system controls the photovoltaic, energy storage, and diesel generators to perform power regulation operations based on the updated coordinated scheduling sequence, and continues to update the digital twin status information using the actual operating data after execution.

9. A photovoltaic-storage-diesel coordinated control system based on digital twin drive, applied to the photovoltaic-storage-diesel coordinated control method based on digital twin drive as described in any one of claims 1 to 8, characterized in that, include: The twin state module is used to acquire multi-source operating state data during the operation of the photovoltaic-storage-diesel microgrid, and to construct digital twin state information based on the multi-source operating state data. The supply and demand module, based on digital twin status information, obtains information on load change trends, photovoltaic output change trends, energy storage status change trends, and diesel generator standby demand information of the photovoltaic-storage-diesel microgrid in the future operating cycle, and predicts the energy supply demand in the future operating cycle to obtain the corresponding energy supply and demand information. The control classification module is used to determine the start-stop control information of the diesel generator, the target output power range of the diesel generator, and the energy storage charging and discharging power constraint information based on energy supply and demand information and digital twin status information. The coordination and scheduling module generates a coordinated scheduling sequence between photovoltaic, energy storage system and diesel generator based on diesel generator start-stop control information, diesel generator target output power range and energy storage charging and discharging power constraint information, under the preset constraint conditions of photovoltaic-storage-diesel microgrid. The actual operation module is used to control the photovoltaic, energy storage system and diesel generator to perform corresponding power regulation operations according to the coordinated scheduling sequence, and to collect the actual operation data in real time during the execution process; The scheduling correction module is used to compare the actual running data with the digital twin status information, obtain the status deviation, and correct the coordinated scheduling sequence based on the status deviation.

10. A coordinated control terminal for photovoltaic, energy storage, and diesel fuel based on digital twin technology, characterized in that: include: One or more processors; A storage device on which one or more programs are stored; When one or more programs are executed by one or more processors, the one or more processors implement the digital twin-driven light-storage-diesel coordinated control method according to any one of claims 1 to 8.

Images