Hybrid energy storage power supply system for diesel engine substitution and energy management method

Abstract

The invention discloses a hybrid energy storage power supply system for diesel engine replacement and an energy management method, and relates to the technical field of hybrid energy storage power supply and management methods. An assembly with the photoelectric conversion efficiency exceeding 23% and a 10KV-level string type inverter are adopted, and a photovoltaic-energy storage combined MPPT control function is integrated; the lithium iron phosphate energy storage system is matched with a bidirectional PCS and an active equalization technology, and is designed according to 1.5 times of daily peak-valley difference and 2 hours of emergency standby power; the diesel generating set group is started in a grading manner according to load gaps, and is matched with a 10KV / 380V intelligent charging unit; the coordination control center realizes two-dimensional prediction and coordinated scheduling based on an LSTM neural network, couples each unit through a 10KV intelligent bus system, divides load priorities, collects meteorological and equipment parameters, and constructs photovoltaic priority consumption, energy storage stabilization fluctuation, diesel generator guarantee emergency, and flexible and adjustable charging operation modes. The method has the advantages that the energy utilization efficiency is improved, the power supply stability is enhanced, the whole-cycle cost is reduced, limited operation and maintenance scenes are adapted, and diversified power supply requirements are met.

Description

Technical Field

[0001] This invention relates to the field of hybrid energy storage power supply and management methods, and in particular to a hybrid energy storage power supply system and energy management method for replacing diesel engines. Background Technology

[0002] In mining areas and industrial parks in emerging markets such as Africa, power supply has long faced challenges such as weak grid infrastructure, large fluctuations in grid voltage, and frequent power outages, leading to frequent interruptions in production activities. To ensure power supply, these areas often rely on diesel generator sets as the main power source. However, diesel power generation suffers from unstable fuel supply, high maintenance costs, and high fuel consumption per unit of power generation over the long term, which not only drives up electricity costs but also causes serious carbon emissions exceeding standards. At the same time, these regions have abundant solar resources and favorable conditions for developing photovoltaic power generation. However, existing photovoltaic-storage-diesel systems generally suffer from insufficient photovoltaic absorption capacity. During peak photovoltaic output in the daytime, if the energy storage system is already fully charged according to a fixed threshold, the excess electricity cannot be effectively stored or utilized, resulting in a large amount of photovoltaic curtailment. At night or during periods of insufficient sunlight, the energy storage system is depleted prematurely because the previous charging and discharging plan was not dynamically adjusted in conjunction with load fluctuations, forcing a return to diesel power generation, creating a contradiction of both curtailment and power shortage.

[0003] Existing photovoltaic-storage-diesel systems suffer from weak collaborative control capabilities, making them ill-suited for dynamic load scenarios. Most systems rely on fixed thresholds to determine diesel generator start-up / shutdown and energy storage charging / discharging, failing to incorporate photovoltaic output prediction and load variation patterns for global optimization. This leads to significant energy supply-demand mismatch issues. Regarding grid-connected / off-grid switching, traditional systems often depend on a single ATS (Automatic Transfer Switch), lacking coordination with PCS (Power Control System) converter master-slave control. This results in slow switching response, with some systems experiencing switching times far exceeding the tolerance range of sensitive loads, potentially causing power outages for charging stations, production equipment, and other sensitive loads. Furthermore, when multiple diesel generator sets operate in parallel, uneven load distribution frequently occurs, leading to excessive circulating current. This not only affects equipment lifespan but can also cause system voltage and frequency fluctuations, further reducing power supply reliability.

[0004] Meanwhile, the existing systems lack sufficient operational adaptability and scenario compatibility, making it difficult to meet the actual conditions of limited operational resources and diverse needs in Africa. Ordinary photovoltaic-storage-diesel systems can only meet basic power supply needs; to meet charging requirements in scenarios such as mining areas, additional modifications are required, which are complex and inefficient. Furthermore, the systems rely heavily on manual on-site operation and troubleshooting, and the scarcity of local technical personnel leads to long troubleshooting cycles, further exacerbating the risk of power outages. After the diesel generator unit starts up, its output voltage, frequency, phase angle, and other parameters often do not meet power quality requirements, necessitating manual adjustments. This cumbersome operation easily triggers power outages, failing to meet the urgent need in Africa for turnkey microgrid solutions—systems that can be rapidly deployed, operate stably for a long time, and are easy to maintain. Summary of the Invention

[0005] The present invention proposes a hybrid energy storage power supply system and energy management method for replacing diesel engines, in order to solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a hybrid energy storage power supply system for diesel engine replacement, comprising: Photovoltaic array module: The photovoltaic modules are equipped with 10KV string inverters, with each inverter connecting 20-30 modules. It integrates photovoltaic-energy storage joint MPPT control function, and adjusts the grid connection voltage fluctuation through energy storage when there are sudden changes in sunlight. Lithium iron phosphate energy storage module: equipped with bidirectional energy storage converter PCS, integrating V / F and PQ dual-mode switching function, the total capacity is designed according to 1.5 times the daily peak-valley difference + 2 hours of emergency backup power, and adopts active balancing technology to balance the battery cells; Diesel generator set group: The diesel generator set group serves as a backup emergency power source. It is started in stages according to the load gap and the unit output is adjusted in advance through load prediction algorithm to maintain the unit load rate in the range of 60%-80%. Charging Unit: The unit is a 10KV / 380V charging array, adapted to the charging needs of mining areas, and supports flexible power supply mode, dynamically adjusting the charging power according to the energy supply and demand status. Coordination and Control Center: The coordination and control center is the RB-I5martEMS coordination center, which realizes dual-dimensional prediction of photovoltaic output and load demand based on LSTM neural network, plans the coordinated scheduling strategy of photovoltaic-energy storage-diesel power generation 24 hours in advance, and realizes data interaction and command issuance of each unit through TSN time-sensitive network; 10KV bus module: The 10KV bus integrates and manages the photovoltaic array system, lithium iron phosphate energy storage system, diesel generator set group, and charging unit, reducing power transmission losses over medium and long distances. Compared with 0.4KV low-voltage transmission, the loss is reduced by 80% for the same power. Load Allocation Unit: The load allocation unit divides the load into primary production load and secondary auxiliary load, and dynamically allocates power resources according to load priority and energy supply status to achieve continuous power supply to the load; Monitoring Module: The monitoring module includes a meteorological monitoring module and an equipment status monitoring module. The meteorological monitoring module collects real-time data on irradiance, temperature, and wind speed, while the equipment status monitoring module collects voltage, current, temperature, and SOC operating parameters of each unit, providing data support for the coordination and control center. The modules are deeply coupled through a 10KV bus module to construct an operation mode that prioritizes photovoltaic power consumption, smooths fluctuations through energy storage, provides emergency backup through diesel generation, and allows for flexible adjustment of charging.

[0007] Furthermore, it also includes: Dynamic Load Simulator: The simulator is designed to implement programmable electronic loads, supporting step / slope load changes to simulate scenarios of concentrated high loads during mining production periods and stable fluctuations in daily power consumption. The coordination control center, based on load simulation data and real-time data collected by the monitoring module, uses the formula dP / dt=k1(P) pv -Pl)+k2(SOC max -SOC)-k3(SOC-SOC min Dynamically adjust the rate of change of charging and discharging power of the energy storage system; where P is the real-time charging and discharging power of the energy storage system, t is time, k1 is the photovoltaic output response coefficient, and P pv For real-time output of the photovoltaic array, Pl is the real-time load power, k2 is the energy storage upper limit adjustment coefficient, and SOC is the state of the load. max SOC is the percentage of the maximum remaining capacity of the energy storage system, and k3 is the lower limit adjustment coefficient of the energy storage system. min This represents the minimum remaining capacity percentage of the energy storage system, and real-time matching between energy storage charging and discharging and photovoltaic output and load fluctuations is achieved through dynamic adjustment.

[0008] Safety interlock module: Real-time monitoring of system voltage, frequency, and temperature parameters, triggering a three-level protection mechanism of early warning, load reduction, and shutdown; when the system voltage deviates by 10±5%kV, the frequency deviates by 50±0.2Hz, or the equipment temperature exceeds 85℃, an early warning signal is first issued and reported to the coordination control center; if the parameters continue to deteriorate for more than 0.3 seconds, the load is automatically reduced by 30%; if the load reduction does not restore normal operation, the power supply to the faulty unit is cut off within 0.5 seconds; the safety interlock module and the coordination control center achieve data interaction through CAN / Modbus dual communication protocols.

[0009] Furthermore, the 10KV string inverter of the photovoltaic array system integrates a photovoltaic-energy storage joint MPPT control algorithm, which improves the photovoltaic output utilization rate to 95% by tracking the maximum power point of the photovoltaic modules and the charging and discharging state of the energy storage system; the bidirectional energy storage converter PCS adopts an improved droop control logic, through formula... Dynamically adjust the output frequency; where f is the PCS output frequency, f ref For a rated output frequency of 50Hz, SOC ref For a target remaining capacity percentage of 80%, k sf Let P be the SOC-frequency correlation coefficient, τ be the integral variable, and P be the frequency correlation coefficient. ref P is the rated output power of the PCS, and k is the real-time output power of the PCS. pfThe power-frequency correlation coefficient is used to automatically increase the output frequency by 0.5Hz to trigger peak clipping when the SOC is less than 20% and the frequency is increased by 0.3Hz through integration and accumulation. When the SOC is greater than 80%, the frequency is reduced by 0.3Hz to absorb excess power and maintain the stability of the system frequency.

[0010] Furthermore, the load forecasting algorithm of the coordination control center adopts a combined forecasting model based on historical load data and real-time monitoring parameters. By analyzing factors such as mine production plans, weather trends, and equipment operating status, it can predict the load demand for the next 24 hours. Based on the load forecasting results, the coordination control center implements a tiered starting strategy for the diesel generator group. When the predicted load gap is less than 30% of the rated capacity, one generator is started; when the gap is between 30% and 60%, two generators are started; and when the gap is greater than 60%, all generators are started. At the same time, the charging and discharging power of the energy storage system is adjusted to supplement the remaining load.

[0011] Furthermore, the 10kV busbar module is equipped with an on-grid switching cabinet (STS), which adopts an integrated design of fast switching, autonomous detection, logic control, and external communication, supporting both active and passive switching modes. During active switching, after the coordination control center issues an off-grid command, the PCS switches to off-grid operation within 0.2 seconds, and the STS simultaneously disconnects the fast switch to achieve the switching. During passive switching, the sampling board collects grid-side voltage information in real time. When a voltage drop is detected, the control board confirms the grid power outage within half a cycle, immediately controls the fast switch to disconnect, and issues a switching command to the PCS, achieving zero-flashover for sensitive loads such as charging piles. The on-grid switching cabinet also has a synchronous grid connection function. During grid connection, it detects the grid voltage phase and amplitude, and adjusts the PCS output parameters through the coordination control center until it matches the grid and completes the grid connection.

[0012] Furthermore, it also includes: Cloud monitoring module: The cloud monitoring module interacts with the coordination and control center via Ethernet to display system operating status, energy flow, and equipment parameter information in real time, supporting remote control and fault diagnosis. The cloud monitoring module has a built-in mining area scenario strategy library, including preset modes for daily power supply, high production load, and emergency power supply, which can be automatically adapted or manually switched according to user needs. When equipment malfunctions, the cloud monitoring module uses fault diagnosis algorithms to locate the fault type and location, generates maintenance guidance plans, and pushes them to management personnel. At the same time, the cloud monitoring module records system operating data, providing historical data support for strategy optimization in the coordination and control center.

[0013] An energy management method for a hybrid energy storage power supply system used as a diesel engine replacement, comprising: Equipment parameter pre-calibration: Establish a photovoltaic MPPT efficiency model, correct the temperature / irradiance coefficient to make the error less than 2%, and calibrate the SOC-OCV curve of the energy storage system using piecewise linear fitting and R... 2>0.99, adjust the parameters of the diesel generator governor and excitation system to make the frequency fluctuation less than 0.5Hz; Joint commissioning: When the photovoltaic output P pv When the load demand is ≥80%, the system enters photovoltaic-dominated mode, the energy storage system charges, and the diesel generator set shuts down. The target is photovoltaic utilization rate >95% and energy storage charging efficiency >92%. pv When the load is less than 50% and the energy storage SOC is less than 30%, the system enters the diesel generator auxiliary mode. The diesel generator starts and carries the load up to 70%, and the energy storage system supplements the remaining power. The target is a diesel generator load rate of 40%-60% and a fuel consumption rate of less than 220g / kWh. When simulating a mains power failure, the system enters the off-grid switching mode. The system switches to off-grid mode within 0.2 seconds. The target is a switching success rate of more than 99% and a frequency stabilization time of less than 1 second. Dynamic strategy optimization: Based on historical operating data, train an LSTM neural network, input meteorological monitoring data, historical load data, and equipment operating parameters, predict photovoltaic output and load demand for the next 24 hours, generate an energy storage charging and discharging schedule, and optimize the number of diesel generator start-stop times to ≤3 times per day; Grid-connected and off-grid switching control: When connected to the grid, the coordination control center prioritizes the use of photovoltaic and grid power to ensure the power supply of the load, and stores the excess power in the energy storage system. When the photovoltaic output is at its peak and the energy storage is fully charged, the power transmitted from the 10KV transmission line to the grid shall not exceed 50kW. When off-grid, the energy storage system serves as a backup power source, combined with diesel generator sets to ensure power supply. Dynamic load allocation: Based on load priority, primary production loads are given priority in receiving power supply. When energy supply is insufficient, the power supply to secondary auxiliary loads is reduced through the load allocation unit. Closed-loop management of equipment status: The monitoring module collects the operating parameters of each unit in real time and uploads them to the coordination control center and cloud monitoring module. When the parameters exceed the preset threshold, the protection mechanism of the safety interlock module is triggered. At the same time, the coordination control center adjusts the operating strategy to form a closed-loop management of monitoring-analysis-control-feedback.

[0014] Furthermore, in the dynamic strategy optimization step, the photovoltaic output prediction uses an integral algorithm to quantify the cumulative effect of irradiance, through the formula... Calculate the projected photovoltaic output for the next 24 hours; where P pv_pred The total predicted photovoltaic output for 24 hours is given by I(τ), the predicted irradiance at time τ, η, the photoelectric conversion efficiency of the photovoltaic module, S, the total area of ​​the photovoltaic array, α, the temperature coefficient of the photovoltaic module, and ΔT(τ), which is the deviation between the temperature of the photovoltaic module at time τ and the standard test temperature. The comprehensive influence of irradiance and temperature on photovoltaic output at different times is fully considered through integral calculation.

[0015] Furthermore, in the joint commissioning step, the load power of the diesel generator set is optimally matched through a dynamic adjustment algorithm. When the load power changes, the coordination control center adjusts the load power according to the formula... Real-time adjustment of the diesel generator set's output power; where P deg P represents the real-time output power of the diesel generator set. load P represents the current total load power, β is the diesel generator load factor optimization coefficient, and P bat To ensure the real-time output power of the energy storage system, the diesel generator set is dynamically adjusted to always operate within the optimal load range. Meanwhile, in the off-grid switching mode, the V / F control parameters of the energy storage system PCS and the excitation parameters of the diesel generator set are synchronously adjusted by the coordination control center to control voltage fluctuations within ±5% during the switching process.

[0016] Compared with existing technologies, the beneficial effects of this invention are: (a) Improve energy efficiency This invention effectively addresses the problems of severe solar curtailment and high dependence on diesel power generation in existing systems by employing an operation mode that prioritizes solar power consumption, uses energy storage to mitigate fluctuations, and relies on diesel generators for emergency backup. The system leverages the multi-dimensional predictive capabilities of its coordination and control center to pre-plan the charging and discharging rhythm of the solar-energy storage system. This avoids curtailment caused by fully charged energy storage during peak solar output periods, while simultaneously allocating energy storage discharge rationally during off-peak solar output periods, reducing reliance on diesel generators and lowering carbon emissions. The dynamic charging and discharging regulation of the energy storage system, combined with a tiered start-up strategy for the diesel generator, ensures that the diesel generator always operates within an optimal load range, reducing energy waste during low-load operation and improving overall system energy efficiency, thus maximizing the utilization of clean energy.

[0017] (ii) Enhance power supply stability and continuity This invention optimizes the grid-connected / off-grid switching mechanism and power quality control capabilities, significantly improving power supply stability. Through the synergy of the Time-Sensitive Network (TSN) and PCS master-slave control and sequential operation logic, the grid-connected / off-grid switching time is greatly shortened, ensuring that sensitive loads will not experience power outages during the switching process, thus improving power supply continuity. The bidirectional energy storage converter's V / F and PQ dual-mode switching function and improved control logic can smooth out voltage and frequency fluctuations caused by photovoltaic power output fluctuations and load changes in real time, ensuring that the output power quality meets standards. After the diesel generator starts up, the system monitors and adjusts its voltage, frequency, phase angle, and other parameters in real time to ensure that the diesel generator output meets the load's power demand, avoiding power outages due to substandard power quality, and providing a stable and reliable power guarantee for production and daily life.

[0018] (III) Reduce operation and maintenance difficulty and total life cycle cost This invention significantly reduces maintenance difficulty and labor costs through fully automated control and cloud monitoring. The system can operate unattended, requiring only simple training for operators, reducing reliance on specialized technicians and adapting to the limited maintenance conditions in Africa. Cloud-based diagnostics quickly pinpoints fault types and locations, shortening troubleshooting cycles and minimizing power outages. Increased equipment integration reduces the complexity of transportation and installation, while also lowering subsequent maintenance workload and costs. Furthermore, the system meets charging needs without additional modifications, avoiding the extra investment required for traditional system upgrades. Cost optimization is achieved throughout the entire lifecycle from deployment and operation to maintenance, enhancing the economic feasibility of the project.

[0019] (iv) Enhance scenario adaptability and flexibility This invention addresses the practical needs of scenarios such as mining areas and industrial parks in Africa, exhibiting excellent adaptability. The system incorporates a strategy library specific to mining areas and other scenarios, automatically sensing the power consumption characteristics of different scenarios and adjusting energy allocation and supply modes. It operates efficiently in scenarios involving daily power consumption, high production loads, and emergency power supply without manual intervention. The flexible and adjustable design of the intelligent charging unit can adapt to the charging needs of different types of loads, solving the problem that traditional systems cannot simultaneously handle power supply and charging. The containerized outdoor installation design facilitates rapid deployment in regions such as Africa, adapting to situations with limited on-site installation conditions. Furthermore, the system configuration can be flexibly adjusted according to load scale and energy demand to meet personalized power supply needs in different scenarios. Attached Figure Description

[0020] Figure 1 This is a schematic block diagram of the hybrid energy storage power supply system for diesel engine replacement proposed in this invention; Figure 2 This is a schematic diagram of the hybrid energy storage management method for diesel engine replacement proposed in this invention; Figure 3 This is a schematic diagram comparing the off-grid switching times of different systems for hybrid energy storage management for diesel engine replacement proposed in this invention. Figure 4 This is a schematic diagram showing the relationship between diesel generator load rate and unit power generation fuel consumption in the hybrid energy storage management method for diesel engine replacement proposed in this invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0023] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0024] Reference Figures 1 to 4 A hybrid energy storage power supply system for diesel engine replacement, comprising: Photovoltaic array module: The photovoltaic array system uses photovoltaic modules that are resistant to ultraviolet rays and high and low temperatures, with a photoelectric conversion efficiency of over 23%. It is equipped with 10KV string inverters, with each inverter connected to 20-30 modules. It integrates photovoltaic-energy storage joint MPPT control function, which improves the low light response speed by 30%. When there are sudden changes in light intensity, it can adjust the voltage fluctuation at the grid connection point through energy storage. Lithium iron phosphate energy storage module: The lithium iron phosphate energy storage system is equipped with a bidirectional energy storage converter PCS, which integrates V / F and PQ dual-mode switching functions. The total capacity is designed according to "1.5 times the daily peak-valley difference + 2 hours of emergency backup power". Active balancing technology is used to balance the battery cells and extend the battery cycle life. Diesel generator set group: The diesel generator set group serves as a backup emergency power source. It is started in stages according to the load gap and the unit output is adjusted in advance through load prediction algorithm to maintain the unit load rate in the range of 60%-80%. Intelligent charging unit: The intelligent charging unit is a 10KV / 380V charging array, which is adapted to the charging needs of mining areas and supports a flexible and adjustable power supply mode, dynamically adjusting the charging power according to the energy supply and demand status. Coordination and Control Center: The coordination and control center is the RB-I5martEMS intelligent coordination center, which realizes dual-dimensional prediction of photovoltaic output and load demand based on LSTM neural network, plans the coordinated scheduling strategy of photovoltaic-energy storage-diesel power generation 24 hours in advance, and realizes data interaction and command issuance of each unit through TSN time-sensitive network. 10KV Intelligent Busbar Module: The 10KV intelligent busbar system realizes integrated management and control of photovoltaic array system, lithium iron phosphate energy storage system, diesel generator group and intelligent charging unit, reducing power transmission loss over medium and long distances. Compared with 0.4KV low voltage transmission, the loss is reduced by more than 80% under the same power. Load Allocation Unit: The load allocation unit divides the load into primary production load and secondary auxiliary load, and dynamically allocates power resources according to load priority and energy supply status to achieve continuous power supply to critical loads; Monitoring Module: The monitoring module includes a meteorological monitoring module and an equipment status monitoring module. The meteorological monitoring module collects real-time data on irradiance, temperature, and wind speed, while the equipment status monitoring module collects voltage, current, temperature, and SOC operating parameters of each unit, providing data support for the coordination and control center. The modules are deeply coupled through a 10KV multi-bus system to construct an operating mode that prioritizes photovoltaic power consumption, smooths fluctuations through energy storage, provides backup power for diesel generation, and offers flexible and adjustable charging.

[0025] This invention also includes: Dynamic load simulator: The dynamic load simulator is a programmable electronic load that supports step / slope load changes with a change rate ranging from 0-100kW / s. It can simulate concentrated high loads during mining production periods and stable fluctuations in daily power consumption. The coordination control center, based on the load simulation data and real-time data collected by the monitoring module, uses the formula dP / dt=k1(P pv -Pl)+k2(SOC max -SOC)-k3(SOC-SOC min The rate of change of charging and discharging power of the energy storage system is dynamically adjusted; where P is the real-time charging and discharging power of the energy storage system, t is time, and k1 is the photovoltaic output response coefficient, with a value range of 0.02-0.05 kW / (s·kW). pv Where Pl is the real-time output of the photovoltaic array, k2 is the real-time load power, k2 is the energy storage upper limit adjustment coefficient, with a value range of 0.01-0.03 kW / (s·%), and SOC is... max, where SOC is the percentage of the maximum remaining capacity of the energy storage system, k3 is the lower limit adjustment coefficient for energy storage, with a value ranging from 0.015 to 0.04 kW / (s·%), and SOC is the percentage of the real-time remaining capacity of the energy storage system. min The minimum remaining capacity percentage of the energy storage system is used to achieve real-time matching between energy storage charging and discharging and photovoltaic output and load fluctuations through dynamic adjustment, thereby reducing the frequency of diesel generator start-up and shutdown.

[0026] Safety Interlock Module: The safety interlock module monitors system voltage, frequency, and temperature parameters in real time, triggering a three-level protection mechanism of early warning, load reduction, and shutdown. When the system voltage deviates by 10±5%kV, the frequency deviates by 50±0.2Hz, or the temperature of critical equipment exceeds 85℃, an early warning signal is first issued and reported to the coordination control center. If the parameters continue to deteriorate for more than 0.3 seconds, the load is automatically reduced by 30%. If the load reduction does not restore normal operation, the power supply to the faulty unit is cut off within 0.5 seconds. The safety interlock module and the coordination control center achieve data interaction through CAN / Modbus dual communication protocols, ensuring rapid response and execution of protection commands and reducing power outages or safety accidents caused by equipment failures.

[0027] In this invention, the 10KV string inverter of the photovoltaic array system integrates a photovoltaic-energy storage joint MPPT control algorithm. By tracking the maximum power point of the photovoltaic modules and the charging and discharging state of the energy storage system, the photovoltaic power utilization rate is increased to over 95%. The bidirectional energy storage converter PCS adopts an improved droop control logic, using the formula... Dynamically adjust the output frequency; where f is the PCS output frequency, f ref For a rated output frequency of 50Hz, SOC ref For a target remaining capacity percentage of 80%, k sf The SOC-frequency correlation coefficient ranges from 0.002 to 0.005 Hz / (s·%), where τ is the integral variable and P is the frequency correlation coefficient. ref P is the rated output power of the PCS, and k is the real-time output power of the PCS. pf It is the power-frequency correlation coefficient, with a value range of 0.001-0.003Hz / (s·kW). When SOC<20%, the output frequency is automatically increased by 0.5Hz through integral accumulation to trigger peak clipping. When SOC>80%, the frequency is reduced by 0.3Hz to absorb surplus power and maintain system frequency stability.

[0028] In this invention, the load forecasting algorithm of the coordination control center adopts a combined forecasting model based on historical load data and real-time monitoring parameters. By analyzing multiple factors such as mine production plans, weather trends, and equipment operating status, it can predict the load demand for the next 24 hours, with the forecasting error controlled within ±8%. Based on the load forecasting results, the coordination control center implements a tiered starting strategy for the diesel generator group. When the predicted load gap is less than 30% of the rated capacity, one unit is started; when the gap is between 30% and 60%, two units are started; and when the gap is greater than 60%, all units are started. At the same time, the charging and discharging power of the energy storage system is adjusted to supplement the remaining load. This strategy reduces the unit fuel consumption of the diesel generator group by 15% and the annual operating cost by 42%, reducing the inefficiency caused by frequent start-stop or low-load operation of the units.

[0029] In this invention, the 10KV intelligent bus module is equipped with a grid-connected switching cabinet (STS), which adopts an integrated design of fast switching, high-precision detection, logic control, and external communication, supporting both active and passive switching modes. During active switching, after the coordination control center issues an off-grid command, the PCS switches to off-grid operation within 0.2 seconds, and the STS simultaneously disconnects the fast switch, achieving seamless switching. During passive switching, the sampling board collects grid-side voltage information in real time. When a voltage drop is detected, the control board confirms the grid power outage within half a cycle, immediately controls the fast switch to disconnect, and issues a switching command to the PCS. The entire switching process does not exceed 20ms, achieving zero-flashover for sensitive loads such as charging piles. The grid-connected switching cabinet also has a synchronous grid connection function. During grid connection, it detects the grid voltage phase and amplitude, and adjusts the PCS output parameters through the coordination control center until it matches the grid, thus completing grid connection and reducing grid connection impact.

[0030] This invention also includes: Cloud-based monitoring module: The cloud-based monitoring module interacts with the coordination and control center via Ethernet, displaying real-time system operating status, energy flow, and equipment parameter information, and supporting remote control and fault diagnosis. The cloud-based monitoring module has a built-in mine area scenario strategy library, including multiple preset modes such as daily power supply, high production load, and emergency power supply, which can be automatically adapted or manually switched according to user needs. When equipment malfunctions, the cloud-based monitoring module uses fault diagnosis algorithms to locate the fault type and location, generates maintenance guidance plans, and pushes them to management personnel, reducing fault handling time from several days to within one hour. Simultaneously, the cloud-based monitoring module records system operating data, providing historical data support for strategy optimization by the coordination and control center, thereby improving system operating efficiency.

[0031] The present invention also discloses an energy management method for a hybrid energy storage power supply system for diesel engine replacement, comprising the following steps: Equipment parameter pre-calibration: Establish a photovoltaic MPPT efficiency model, correct the temperature / irradiance coefficient to make the error less than 2%, calibrate the SOC-OCV curve of the energy storage system using piecewise linear fitting with R²>0.99, and adjust the parameters of the diesel generator governor and excitation system to make the frequency fluctuation less than 0.5Hz; Multi-modal joint commissioning: When the photovoltaic output Ppv ≥ 80% of the load demand, the system enters the photovoltaic-dominated mode, charging the energy storage system and shutting down the diesel generator set. The target is a photovoltaic utilization rate > 95% and an energy storage charging efficiency > 92%. When Ppv < 50% of the load and the energy storage SOC < 30%, the system enters the diesel generator-assisted mode, starting the diesel generator and carrying the load up to 70%, while the energy storage system supplements the remaining power. The target is a diesel generator load rate of 40%-60% and a fuel consumption rate < 220g / kWh. When simulating a mains power failure, the system enters the off-grid switching mode, switching to off-grid mode within 0.2 seconds. The target is a switching success rate > 99% and a frequency stabilization time < 1 second. Dynamic strategy optimization: Based on historical operating data, train an LSTM neural network, input meteorological monitoring data, historical load data, and equipment operating parameters, predict photovoltaic output and load demand for the next 24 hours, generate an energy storage charging and discharging schedule, and optimize the number of diesel generator start-stop times to ≤3 times per day; Grid-connected / off-grid switching control: When connected to the grid, the coordination control center prioritizes the use of photovoltaic and grid power to ensure the power supply to the load, and stores excess energy in the energy storage system. When the photovoltaic output is at its peak and the energy storage is fully charged, the power transmitted from the 10KV transmission line to the grid is controlled to not exceed 50kW. When off-grid, the energy storage system serves as a backup power source, combined with diesel generator sets to ensure power supply, so that the power supply reliability reaches 99.7%. Dynamic load allocation: Based on load priority, primary production loads are given priority in power supply. When energy supply is insufficient, the power supply of secondary auxiliary loads is reduced through the load allocation unit to ensure that critical production is not affected. Closed-loop management of equipment status: The monitoring module collects the operating parameters of each unit in real time and uploads them to the coordination control center and cloud monitoring module. When the parameters exceed the preset threshold, the protection mechanism of the safety interlock module is triggered. At the same time, the coordination control center adjusts the operating strategy to form a closed-loop management of "monitoring-analysis-control-feedback".

[0032] In this invention, the dynamic strategy optimization step uses an integral algorithm to accurately quantify the cumulative irradiance effect in photovoltaic power output prediction, through the formula... Calculate the projected photovoltaic output for the next 24 hours; where P pv_predThe total photovoltaic output is predicted for 24 hours. I(τ) is the predicted irradiance at time τ (unit: W / m²), η is the photovoltaic module photoelectric conversion efficiency, S is the total area of ​​the photovoltaic array (unit: m²), α is the photovoltaic module temperature coefficient (unit: % / ℃), and ΔT(τ) is the deviation of the photovoltaic module temperature from the standard test temperature (25℃) at time τ (unit: ℃). By fully considering the comprehensive impact of irradiance and temperature on photovoltaic output at different times through integral calculation, the prediction accuracy is improved by 15%-20%, providing a reliable basis for energy storage charging and discharging plans and diesel generator start-up and shutdown planning.

[0033] In this invention, during the multimodal joint commissioning step, the load power of the diesel generator set is optimally matched through a dynamic adjustment algorithm. When the load power changes, the coordination control center adjusts the load power according to the formula... Real-time adjustment of the diesel generator set's output power; where P deg P represents the real-time output power of the diesel generator set. load Where P is the current total load power, β is the diesel generator load rate optimization coefficient, with a value ranging from 0.6 to 0.8, and P... bat To ensure the real-time output power of the energy storage system (positive for discharging and negative for charging), dynamic adjustment is used to ensure that the diesel generator set always operates within the optimal load range, reducing low-load operating time and lowering fuel consumption and equipment wear. Simultaneously, in off-grid switching mode, the V / F control parameters of the energy storage system PCS and the excitation parameters of the diesel generator set are synchronously adjusted by the coordination control center to keep voltage fluctuations within ±5% during the switching process, reducing shutdowns of sensitive loads due to voltage fluctuations and improving the stability and reliability of the system's off-grid operation.

[0034] Specific implementation methods of hybrid energy storage power supply systems and energy management methods for diesel engine replacement: Example 1: Hybrid Energy Storage Power Supply System for a 1 Million Tons / Year Production Capacity Mining Area in Africa This embodiment is applied to a metal mining area in Africa with a capacity of 1 million tons per year. The daily power load of this mining area includes mining equipment, ore dressing production lines, and mining office and living facilities. Among them, the mining equipment and ore dressing production lines are primary production loads, requiring continuous and stable power supply, while the office and living facilities are secondary auxiliary loads, whose power supply can be dynamically adjusted according to the energy supply status. Given the characteristics of the mining area—a weak power grid, high diesel supply costs, and abundant solar resources—a hybrid energy storage power supply system is configured to replace diesel engines, as detailed below: (I) Overall System Configuration The system includes a photovoltaic array system, a lithium iron phosphate energy storage system, a diesel generator set group, an intelligent charging unit, a coordination and control center, a 10KV intelligent bus system, a load distribution unit, a monitoring module, a dynamic load simulator, a safety interlocking module, and a cloud monitoring module. The photovoltaic array system has an installed capacity of 3MW, using UV-resistant and high / low temperature resistant photovoltaic modules with a photoelectric conversion efficiency exceeding 23%. It is equipped with five 10KV string inverters, each connecting 25 photovoltaic modules, and integrates photovoltaic-energy storage joint MPPT control function, improving low-light response speed by 30%. The lithium iron phosphate energy storage system has a total capacity of 6MWh, designed for "1.5 times the daily peak-valley difference + 2 hours of emergency backup power". The daily peak load of the mining area is 2000kW, the valley load is 800kW, the peak-valley difference is 1200kW, 1.5 times the peak-valley difference is 1800kWh, and the 2-hour emergency backup power is 4000kWh, thus matching the total capacity of 6MWh. It is equipped with six 500kW bidirectional energy storage converters (PCS) with integrated V / F and PQ dual-mode switching function. The diesel generator set group includes two 1MW units as backup emergency power. The intelligent charging unit is 10KV / 38 The system features a 0V charging array with 10 380V DC charging piles to meet the charging needs of mining area engineering vehicles. The coordination and control center utilizes the RB-I5martEMS intelligent coordination hub, based on an LSTM neural network for prediction and scheduling. A 10KV intelligent bus system connects various energy units and loads, enabling high-voltage transmission and integrated management. The load allocation unit uses software algorithms to prioritize loads and dynamically allocate power. The monitoring module includes two sets of meteorological monitoring equipment and eight sets of equipment status monitoring devices, collecting parameters such as irradiance, temperature, wind speed, and voltage, current, temperature, and SOC of each unit. The dynamic load simulator is a programmable electronic load supporting step / slope load changes from 0-100kW / s. The safety interlock module integrates voltage, frequency, and temperature sensors and a protection control unit. The cloud monitoring module connects to the coordination and control center via Ethernet for remote monitoring and fault diagnosis.

[0035] (II) Working process and technical details of each module Photovoltaic Array System: The photovoltaic modules are encapsulated with double-layer glass, capable of withstanding ambient temperature variations from -30℃ to 60℃. An anti-UV coating is applied to the surface to reduce efficiency degradation caused by prolonged exposure to sunlight. The 10KV string inverter incorporates a photovoltaic-energy storage joint MPPT control algorithm, which tracks the maximum power point of the photovoltaic modules in real time while receiving charging and discharging status signals from the energy storage system's PCS. For example, if cloud cover causes irradiance to drop sharply from 1000W / m² to 400W / m², the inverter quickly adjusts its output power and sends a fluctuation signal to the coordination control center. The coordination control center then instructs the energy storage system's PCS to discharge synchronously, replenishing the power gap and keeping the grid connection point voltage fluctuation within 10±5%KV, preventing sudden voltage drops from affecting the operation of mining equipment.

[0036] Lithium iron phosphate energy storage system: The energy storage battery uses lithium iron phosphate batteries. Each individual cell has a nominal voltage of 3.2V and a nominal capacity of 280Ah. Every 16 cells form a battery module, with a nominal voltage of 51.2V, a capacity of 280Ah, and energy of 14.336kWh. Every 15 modules form a battery cluster, with a nominal voltage of 768V, a capacity of 280Ah, and energy of 215.04kWh. Every 3 battery clusters are connected to a 500kW PCS, forming one energy storage sub-unit with a capacity of 645.12kWh. Ten energy storage sub-units constitute a total energy storage system with a capacity of 6MWh. The PCS uses an improved droop control logic, through the formula... The output frequency is dynamically adjusted, where f is the PCS output frequency. ref For a rated output frequency of 50Hz, SOC ref The target remaining capacity percentage is 80%, SOC is the real-time remaining capacity percentage of the energy storage system, and k sf The SOC-frequency correlation coefficient has a value of 0.003 Hz / (s·%). P is the integral variable. ref The PCS has a rated output power of 500kW, P is the real-time output power of the PCS, and k is the power of the PCS. pf This is the power-frequency correlation coefficient, with a value of 0.002 Hz / (s·kW). When the energy storage SOC drops to 19% or below 20%, The integral of the term, assuming a duration of t = 167s, yields a value of (80-19) × 0.003 × 167 ≈ 0.5Hz, resulting in an output frequency. If the real-time power P = 450kW, then At this point, the power of other PCS is synchronously adjusted by the coordination control center, ultimately increasing the output frequency by 0.5Hz, triggering load shaving of the mining equipment, and preventing excessive discharge of the energy storage. Simultaneously, the energy storage system employs active balancing technology, using the BMU unit to collect the voltage of individual batteries in real time. When the voltage difference between individual batteries exceeds 50mV, the balancing circuit is activated to balance the charge of each battery cell, increasing the energy storage cycle life from 4000 cycles in ordinary systems to over 6000 cycles.

[0037] Diesel generator set group: The coordination and control center uses a load forecasting algorithm, combined with the mine's production plan and meteorological monitoring data, to predict the load demand for the next 24 hours. The mining peak is from 8:00 AM to 8:00 PM, and the maintenance low point is from 8:00 PM to 8:00 AM. The forecast error is controlled within ±8%. For example, if the predicted load at 10:00 AM is 1800kW and the predicted photovoltaic output is 1200kW, the load gap is 600kW, accounting for 30% of the total diesel generator capacity. In this case, one 1MW diesel generator set is started, and the energy storage system is simultaneously instructed to discharge at 300kW to supplement the remaining 300kW gap, ensuring that the diesel generators actually carry 600kW, with a load rate of 60%, which is within the optimal range of 60%-80%. If the load is predicted to drop to 1000kW after 8 PM and the photovoltaic output to 100kW, resulting in a load shortfall of 900kW (45% of the total diesel generator capacity), then two diesel generator sets will be started, each with a load capacity of 450kW, resulting in a load rate of 45%. At this time, the coordination control center will instruct the energy storage system to discharge at a power of 200kW, increasing the load capacity of each diesel generator to 550kW, while maintaining the optimal load rate of 55%. After the diesel generators start, the system will collect the three-phase voltage, frequency, and phase angle of their output in real time. If the voltage deviation exceeds 10±5%KV (e.g., the voltage rises to 10.6KV), the frequency deviates from 50±0.2Hz (e.g., the frequency drops to 49.7Hz), or the phase angle deviates from 120 degrees (e.g., the phase angle difference is 118 degrees), the system will automatically adjust the diesel generator controller parameters until the power quality requirements are met before closing the diesel generator power switch, thus preventing substandard power from entering the system.

[0038] Intelligent charging unit: The 10KV / 380V charging array is connected to the power grid through a 10KV intelligent bus system. The coordination and control center dynamically adjusts the charging power based on photovoltaic output, energy storage SOC, and load demand. During peak daytime photovoltaic output (e.g., 2500kW, 1800kW load, 60% energy storage SOC), the charging array is instructed to charge at 500kW. In the evening, when photovoltaic output decreases (e.g., 800kW, 1500kW load, 75% energy storage SOC), the charging power is reduced to 200kW. At night, when the energy storage SOC is below 30% and the diesel generator starts, charging is suspended to prioritize power supply to production loads. The charging array has overcurrent, overvoltage, and overtemperature protection functions. When the charging current exceeds 120% of the rated current (e.g., 100A), the current is automatically reduced to the rated current. When the charging voltage exceeds 380±10%V, charging stops and an alarm is issued.

[0039] The coordination and control center uses an LSTM neural network to build a predictive model. Inputting past month's meteorological data (including irradiance, temperature), load data (including mining equipment and office power consumption), and equipment operating parameters (including photovoltaic efficiency and energy storage SOC), the model is trained with 5000 samples to predict photovoltaic output and load demand for the next 24 hours. For example, if the predicted irradiance at 9 AM the next day is 900 W / m², photovoltaic output is 1800 kW, and load is 1600 kW, the energy storage charging and discharging plan is generated: from 0:00 to 6:00 AM, during the off-peak period of grid power, if grid power is available, the energy storage charges at 400 kW; from 9:00 to 11:00 AM, the energy storage discharges at 200 kW; from 2:00 to 4:00 PM, during the peak photovoltaic output period, the energy storage charges at 500 kW; and from 8:00 to 10:00 PM, the energy storage discharges at 300 kW. Simultaneously, the coordination and control center uses a Time-Sensitive Network (TSN) to issue commands to each unit at 10 ms intervals, ensuring real-time data interaction and control commands, and avoiding coordination misalignment due to communication delays.

[0040] 10KV Intelligent Busbar System: Utilizing 10KV cross-linked polyethylene insulated cables, this system connects the photovoltaic inverter, energy storage PCS, diesel generator step-up module, and load distribution unit. The cable cross-sectional area is selected based on the transmission power; a 120mm² cable is chosen for 3MW photovoltaic output, compared to the 600mm² cable required for 0.4KV low-voltage transmission, resulting in over 80% reduction in line loss for the same power output. The busbar system is equipped with a 10KV intelligent switchgear, incorporating voltage and current sensors and protection devices. When the busbar current exceeds 120% of the rated current (e.g., 1000A), exceeding 1200A, non-critical load circuits are automatically disconnected. When the busbar voltage drops below 9.5KV, the energy storage PCS is instructed to increase its output voltage to maintain busbar voltage stability.

[0041] Load Allocation Unit: The mine's load is divided into primary production load and secondary auxiliary load using software algorithms. Primary production load includes mining equipment and ore dressing production lines, with a total power of 1500kW. Secondary auxiliary load includes office lighting and dormitory electricity, with a total power of 800kW. When energy supply is insufficient, such as 800kW of photovoltaic output, 500kW of energy storage discharge, and diesel generator not running, resulting in a total power supply of 1300kW, priority is given to ensuring 1300kW of the primary production load (1500kW), reducing the power supply to the secondary auxiliary load to 0. When energy supply is sufficient, such as 2500kW of photovoltaic output, 300kW of energy storage discharge, and a total power supply of 2800kW, both the primary load (1500kW) and the secondary load (800kW) are ensured, with the remaining 500kW used for energy storage charging.

[0042] Monitoring Modules: The meteorological monitoring module collects irradiance, ambient temperature, and wind speed data every 10 seconds. Irradiance measurement range is 0-2000 W / m² with an accuracy of ±5%. Ambient temperature measurement range is -40℃ to 80℃ with an accuracy of ±0.5℃. Wind speed measurement range is 0-30 m / s with an accuracy of ±0.3 m / s. Data is uploaded to the coordination and control center via RS485 protocol. The equipment status monitoring module collects photovoltaic inverter output voltage and current data every 5 seconds. Photovoltaic inverter output voltage measurement accuracy is ±0. The accuracy of current measurement reaches ±0.5%, and the accuracy of single-cell voltage and temperature measurement of energy storage batteries reaches ±10mV and ±1℃. The accuracy of diesel generator output frequency and power measurement reaches ±0.01Hz and ±1%, and the accuracy of charging array charging current and voltage measurement reaches ±1A and ±0.5V. Abnormal data, such as single-cell battery voltage exceeding 3.65V or falling below 2.5V, are uploaded in real time and trigger alarms.

[0043] Dynamic load simulator: Simulates the step load during startup of mining equipment in a mine, such as a sudden increase from 500kW to 1200kW, with a change rate of 70kW / s, and the smooth fluctuation during daily operation, such as ±100kW, with a change rate of 10kW / s. The coordination and control center, based on the load simulation data and real-time data from the monitoring module, uses the formula dP / dt=k1(P) pv -Pl)+k2(SOC max -SOC)-k3(SOC-SOC min The rate of change of the charging and discharging power of the energy storage system is dynamically adjusted, where P is the real-time charging and discharging power of the energy storage system, t is time, and k1 is the photovoltaic output response coefficient, with a value of 0.03 kW / (s・kW). pv The real-time output of the photovoltaic array is 1800kW, Pl is the real-time load power of 1200kW, k2 is the energy storage upper limit adjustment coefficient, with a value of 0.02kW / (s·%), and SOC. max SOC represents 90% of the maximum remaining capacity of the energy storage system, SOC represents 65% of the real-time remaining capacity of the energy storage system, k3 is the lower limit adjustment coefficient for energy storage, with a value of 0.025 kW / (s·%), and SOC min The minimum remaining capacity percentage of the energy storage system is 20%. Substituting the parameters, we calculate dP / dt = 0.03 × (1800 - 1200) + 0.02 × (90 - 65) - 0.025 × (65 - 20) = 18 + 0.5 - 1.125 = 17.375 kW / s. This means that the energy storage charging and discharging power is adjusted at a rate of 17.375 kW / s to quickly match the step change of the load from 500 kW to 1200 kW and avoid system power imbalance.

[0044] Safety interlock module: Real-time monitoring of system voltage, frequency, and temperature of critical equipment. The system voltage standard is 10±5%KV, the frequency standard is 50±0.2Hz, and critical equipment includes photovoltaic inverters, PCS, diesel generator IGBTs, and transformers. The temperature threshold is 85℃. When the bus voltage is detected to rise to 10.6KV, exceeding the 10+5%KV upper limit, a warning signal is first sent to the coordination control center. If the voltage does not drop below 10.5KV within 0.3 seconds, the photovoltaic inverter output power is automatically reduced by 30%. If the voltage still does not recover after another 0.2 seconds (cumulatively 0.5 seconds), the photovoltaic inverter power supply circuit is immediately cut off. The safety interlock module and the coordination control center exchange data via CAN / Modbus dual communication protocols. The CAN protocol is used for real-time protection command transmission with a delay of less than 10ms, while the Modbus protocol is used for uploading historical alarm data, ensuring rapid response and data integrity of the protection mechanism.

[0045] Cloud-based monitoring module: Connects to the coordination and control center via Ethernet to display the system's operating status in real time, including the start / stop status of each unit, power flow, energy data including daily photovoltaic power generation, diesel fuel consumption, energy storage charging and discharging, and equipment parameters including voltage, current, temperature, and SOC. It has a built-in mining scenario strategy library, including peak mining mode, off-peak maintenance mode, and emergency power supply mode. In peak mining mode, production load is prioritized, and the charging array reduces power. In off-peak maintenance mode, energy storage charges while the diesel generator shuts down. In emergency power supply mode, the diesel generator and energy storage provide combined power. These modes can be automatically switched according to the mining area's production plan or manually selected by management personnel. When the diesel generator experiences low oil pressure, the cloud-based monitoring module uses a fault diagnosis algorithm to compare historical fault data with real-time parameters, locating the fault type as "oil pump malfunction." It generates a repair guidance plan, including checking the oil pump wiring and replacing the oil filter, and pushes it to the mining area management personnel's mobile phone. Simultaneously, it schedules the backup diesel generator to start, reducing the fault handling time from the traditional 3 days to within 1 hour.

[0046] (III) System Operation Performance Data Characterization Table 1 Comparison of System Operation Performance with Existing Photovoltaic Storage and Diesel Storage Systems in Example 1: Table 1 shows that the system of this invention has significant improvements in energy utilization, power supply reliability, and operation and maintenance efficiency compared to existing photovoltaic-storage-diesel systems. The photovoltaic curtailment rate has decreased from 25% to 5%, thanks to the coordinated scheduling of photovoltaic and energy storage in the control center, avoiding curtailment caused by full energy storage during peak photovoltaic output. The diesel generator's fuel consumption per unit of power generation has decreased from 220g / kWh to 187g / kWh, because the system uses load forecasting and tiered start-up strategies to ensure the diesel generator always operates within the optimal load range of 60%-80%, reducing fuel waste during low-load operation. Furthermore, the off-grid switching time has been shortened from 500ms to 20ms, meeting the zero-flashover requirements of sensitive loads such as charging piles and mining equipment through the coordinated control of the STS switching cabinet and PCS. Reliability improved from 85% to 99.7%, relying on energy storage to smooth fluctuations, safety interlock protection, and diesel generator backup for emergency use, reducing the number of power outages; fault handling time was shortened from 72 hours to 1 hour, with remote diagnostic functions of cloud monitoring significantly improving fault diagnosis efficiency; maintenance personnel were reduced from 6 people managing 3 systems to 1 person, with fully automated control and cloud monitoring reducing reliance on manpower; photovoltaic power utilization increased from 75% to 95%, with MPPT control and dynamic power regulation fully tapping the potential of photovoltaic power; the average daily start-up and shutdown frequency of diesel generators decreased from 5 times to 2 times, extending the service life of diesel generators and reducing maintenance costs.

[0047] Example 2: Hybrid Energy Storage Power Supply System in a Medium-Sized Industrial Park in Africa This example is applied to a medium-sized industrial park in Africa, which includes three machine processing plants, two employee dormitories, and one supporting commercial area. The machine processing plants' production lines are primary production loads requiring continuous power supply, with a total power of 1200kW. Dormitory lighting, air conditioning, and the commercial area's power consumption are secondary auxiliary loads, with a total power of 800kW. The park's mains voltage fluctuates within ±15%, with an average of 2-3 power outages per day. Traditionally, the system relies on two 1.5MW diesel generators, resulting in high operating costs and significant carbon emissions. Therefore, a hybrid energy storage power supply system is configured to replace the diesel generators, as detailed below: (I) Overall System Configuration The system adopts a parallel design of three independent subsystems. Each subsystem includes a photovoltaic array, energy storage system, PCS, STS switching cabinet, and MPPT converter. It integrates a coordination and control center, a 10KV intelligent bus system, load distribution unit, monitoring module, dynamic load simulator, safety interlock module, and cloud monitoring module. The total photovoltaic installed capacity is 2MWp, divided into three photovoltaic sub-arrays: two 670kWp sub-arrays and one 660kWp sub-array. Each sub-array is equipped with six 120kW MPPT converters. The total energy storage capacity is 2.57MWh, divided into three energy storage sub-units: two 771kWh sub-units and one 1028kWh sub-unit. Each sub-unit uses lithium iron phosphate batteries and is equipped with one 500kW PCS. The diesel generator set group includes three 500kW units as emergency backup power. The intelligent charging unit is a 10KV / 380V charging array with eight 380V AC charging piles, adaptable to various environments. The system includes charging for logistics vehicles and employee electric vehicles; the coordination and control center is the RB-I5martEMS intelligent coordination center, which enables coordinated scheduling of the three subsystems; the 10KV intelligent bus system aggregates the outputs of the three subsystems and connects them to the park's 10KV high-voltage load and 380V low-voltage load respectively; the load distribution unit achieves load priority management through a combination of hardware loops and software algorithms; the monitoring module includes three meteorological monitoring stations and twelve equipment status monitoring devices; the dynamic load simulator supports load changes from 0-100kW / s; the safety interlock module covers all critical equipment; and the cloud monitoring module enables remote operation and maintenance and strategy adjustment.

[0048] (II) Working Process and Technology of Each Module Detailed photovoltaic array and MPPT converter: The photovoltaic modules are monocrystalline silicon modules with a photoelectric conversion efficiency of 23.5%, which can operate stably in environments ranging from -25℃ to 55℃. Each 670kWp photovoltaic sub-array contains 2234 300W modules, divided into 25 strings. Every 6 strings are connected to one 120kW MPPT converter, for a total of 6 sets. The MPPT converter adopts multi-channel MPPT control technology, with each channel independently tracking the maximum power point of one string. When the output of some modules decreases due to shading, it does not affect other strings, improving the overall efficiency of the photovoltaic array by 10%-15%. The output of the MPPT converter is connected to the DC bus, which is connected to the energy storage system and PCS. The control center coordinates and instructs the MPPT converter to adjust the output power according to the energy storage SOC and load demand. For example, when the energy storage SOC reaches 90% and the load demand is lower than the photovoltaic output, the MPPT converter reduces the output power to avoid photovoltaic power surplus.

[0049] Lithium iron phosphate energy storage system: Each 771kWh energy storage sub-unit contains 3 battery clusters, each battery cluster consists of 16 battery modules. The battery modules are 1P16S, with a nominal voltage of 51.2V, a capacity of 314Ah, and an energy of 16.076kWh. The 16 modules are connected in series to form a 1P256S battery cluster, with a nominal voltage of 819.2V, a capacity of 314Ah, and an energy of 257kWh. The 3 battery clusters are connected in parallel to a 500kW PCS to form a 771kWh energy storage sub-unit; the 1028kWh energy storage sub-unit contains 4 of the above battery clusters, which are connected in parallel to a 500kW PCS. The energy storage system's BMS adopts a three-level architecture, with one CBMU master control module and 16 MBMU slave control modules working together. The MBMUs collect individual battery voltage and temperature data every 2 seconds, with voltage measurement accuracy reaching ±10mV and temperature measurement accuracy reaching ±1℃. The CBMUs summarize the data and calculate SOC and SOH every 5 seconds, with an SOC error of less than 8%. When the SOC is below 20%, a discharge limit command is sent to the coordination control center; when it is above 80%, a charging limit command is sent. The PCS adopts a three-level topology design with a conversion efficiency of 97.5%. It integrates V / F and PQ dual-mode switching functions. In off-grid mode, the bus voltage is maintained at 380±10%V and the frequency at 50±0.2Hz through V / F control. In grid-connected mode, active and reactive power are adjusted according to commands from the coordination control center.

[0050] Diesel generator set group: 3 500kW diesel generator sets are divided into 2 main units and 1 standby unit. The coordination and control center determines the start-up and shutdown of the diesel generators based on load forecasting and energy storage SOC. Load forecasting adopts a combined forecasting model, inputting the park's production plan and meteorological data for the past 3 months. Monday to Friday are production days with a load of 1800kW, and Saturday and Sunday are rest days with a load of 800kW. The average irradiance is 400W / m² in the rainy season and 800W / m² in the dry season. The forecasting error is controlled within ±8%. For example, on a production day from 8:00 AM to 6:00 PM, the predicted load is 1800kW, the predicted photovoltaic output is 900kW, the energy storage SOC is 60%, and the load gap is 900kW. Two main diesel generators with a total capacity of 1000kW are started, each carrying a load of 450kW, resulting in a load rate of 90%. Simultaneously, the energy storage system is instructed to discharge at a power of 300kW to supplement the remaining 300kW gap, reducing the diesel generator load rate to the optimal range of 60%. If the main diesel generator fails, the backup diesel generator starts within 10 seconds to ensure uninterrupted power supply to the load. After the diesel generator starts, the system collects its three-phase voltage, frequency, and phase angle. The voltage deviation must be controlled within 400±5%V (corresponding to 10KV bus voltage of 10±5KV), the phase angle difference must be 120±2 degrees, the negative sequence voltage imbalance must be less than 2%, and the total harmonic distortion of the voltage must be less than 5%. Once these requirements are met, the power supply switch is closed; otherwise, the excitation current and speed controller parameters are adjusted through the diesel generator controller until the standards are met.

[0051] Coordination and Control Center and Scheduling Strategy: Based on an LSTM neural network-trained prediction model, inputting photovoltaic (PV) output data, load data, and meteorological data from the past two months, with PV output data every 15 minutes, load data every 10 minutes, and meteorological data every hour, the model training error is less than 5%. It can predict PV output and load demand every 15 minutes for the next 24 hours. For example, predicting a peak PV output of 700kW for a single subsystem at 12 PM the next day, with a load demand of 600kW, generates an energy storage charging and discharging plan: from 1 AM to 5 AM (out of grid power), the energy storage charges at 300kW; from 12 PM to 2 PM (PV peak), the energy storage charges at 100kW; from 7 PM to 10 PM (load peak), the energy storage discharges at 400kW. Simultaneously, the coordination and control center uses formulas... Calculate the total predicted photovoltaic output for 24 hours, where P pv_pred Contribute to the 24-hour total photovoltaic forecast. for The predicted irradiance at any given time is 800 W / m² during the day (12 hours) and 0 W / m² at night (12 hours). The photovoltaic module's photoelectric conversion efficiency is 23.5%, S is the area of ​​a single photovoltaic subarray, 670kWp / 235W / m²≈2851m², and α is the photovoltaic module's temperature coefficient -0.45% / ℃. for The deviation of the photovoltaic module temperature from the standard test temperature of 25℃ is measured at any given time, averaging 20℃ during the day and 0℃ at night. Substituting the parameters, the calculation yields: the integral term for the 12-hour daytime period is 800×0.235×2851×(1-0.0045×20)×3600×12=800×0.235×2851×0.91×3600×12≈800×0.235×2851×0.91=478442.84, multiplied by 3600×12, we get 478442.84×43200≈2.067× The total predicted output of photovoltaic power for a single subsystem over 24 hours is approximately 20,670 kWh. The integral term for the nighttime 12-hour period is 0. The total predicted output of photovoltaic power for the three subsystems is approximately 62,010 kWh. The deviation from the actual operating data is less than 8%, providing a reliable basis for scheduling strategies.

[0052] 10kV Intelligent Busbar and STS Switching Cabinet: The 10kV intelligent busbar system adopts a segmented design. Each subsystem is connected to the 380V low-voltage busbar through a 10kV / 0.4kV isolation transformer. The three low-voltage busbars are connected in parallel to supply power to the park load. The busbar system is equipped with intelligent protection devices. When a short-circuit fault occurs in a section of the busbar, the faulty section is quickly disconnected without affecting the power supply to other sections. The STS switching cabinet has a capacity of 1MW and is used for grid-connected / off-grid switching of each subsystem. It supports both active and passive switching: During active switching, the coordination control center issues an off-grid command, and the PCS switches to off-grid operation within 0.2 seconds. The STS simultaneously disconnects the mains power switch, and there is no power interruption during the switching process. During passive switching, the sampling board collects the mains voltage in real time. When the mains voltage drops to 320V, which is 15% lower than 380V, the control board confirms the mains power failure within 10ms of half a cycle, immediately controls the STS to disconnect the mains power switch, and commands the PCS to switch to off-grid mode. The entire switching process takes 18ms, ensuring that the production line equipment does not stop. The STS switching cabinet also has a synchronous grid connection function. During grid connection, it detects the phase difference between the mains voltage (e.g., mains phase 0°) and the PCS output phase (e.g., PCS output phase 10°). The coordination control center commands the PCS to adjust the output phase until the phase difference with the mains is less than 5° and the voltage deviation is less than 2%. Then, the grid connection switch is closed to achieve seamless grid connection.

[0053] Load Distribution and Dynamic Load Simulation: The load distribution unit combines hardware contactors with software algorithms. Primary production loads are directly connected to the critical load circuit of the 10KV intelligent busbar, which has priority power supply. Secondary auxiliary loads are connected to ordinary load circuits, which can be switched on and off according to the energy supply status. When the total system power supply is below 1200kW (primary load power), the ordinary load circuit automatically disconnects to prioritize power supply to the production line. When the power supply is between 1200-2000kW (primary + secondary load total power), some ordinary load circuits are closed (e.g., dormitory air conditioners operate at half capacity). When the power supply exceeds 2000kW, all ordinary load circuits are closed. The dynamic load simulator simulates the ramp load during the startup of a machining plant production line (e.g., increasing from 300kW to 800kW at a rate of 50kW / s) and the fluctuating load during processing (e.g., ±80kW at a rate of 8kW / s). The coordinated control center uses the formula dP / dt=k1(P) to control the load. pv -Pl)+k2(SOC max -SOC)-k3(SOC-SOC min Adjust the rate of change of energy storage charging and discharging power, where k1 = 0.04 kW / (s·kW), P pv =600kW is the photovoltaic output of a single subsystem, Pl=800kW is the load of a single subsystem, k2=0.015kW / (s·%), SOC max=85%, SOC=70%, k3=0.03kW / (s·%), SOC min =25%. Substituting into the calculation, we get dP / dt=0.04×(600-800)+0.015×(85-70)-0.03×(70-25)=-8+0.225-1.35=-9.125kW / s, which means that the energy storage increases the discharge power at a rate of 9.125kW / s, from the initial 100kW to 300kW, to make up for the gap between photovoltaic output and load and maintain the system power balance.

[0054] Safety Interlocking and Cloud Monitoring: The safety interlocking module monitors the voltage, frequency, and equipment temperature of each subsystem. Voltage standards include 10±5%KV and 380±10%V; the frequency standard is 50±0.2Hz; and equipment temperatures include the IGBT temperature of the PCS and the engine oil temperature of the diesel generator. The IGBT temperature threshold for the PCS is 85℃, and the engine oil temperature threshold is 95℃. When the IGBT temperature of a PCS reaches 86℃, an alarm signal is immediately sent to the coordination control center, and the PCS is instructed to reduce its load by 20%. If the temperature continues to rise to 88℃, the power supply to that PCS is cut off to prevent equipment damage. The safety interlocking module communicates with the controllers of the three subsystems via a CAN bus at a communication rate of 250kbps, ensuring real-time transmission of protection commands. The cloud-based monitoring module collects real-time operational data from the three subsystems, generating daily, monthly, and annual reports, such as monthly total photovoltaic power generation, total diesel fuel consumption, and total energy storage charging and discharging. It supports remote control functions, allowing managers to adjust energy storage charging and discharging thresholds in the cloud, such as adjusting the SOC charging limit from 85% to 80%, and diesel generator start-up and shutdown parameters, such as adjusting the diesel generator start-up threshold from SOC<25% to SOC<20%. When a new production line is added to the park, increasing the load by 300kW, the cloud-based monitoring module can remotely upgrade the strategy library of the coordination control center without on-site modifications, thus improving system scalability.

[0055] Dynamic adjustment of diesel generator power: In the multi-mode joint commissioning process, the load-bearing power of the diesel generator set is determined by the formula... Real-time adjustment, where P deg For the real-time output power of the diesel generator, P load Let the total load power of the current subsystem be, for example, 600kW, β be the diesel generator load rate optimization coefficient, with a value of 0.7, and P bat The real-time output power of the energy storage system is represented by positive for discharging and negative for charging, for example, an energy storage system discharging 120kW. Substituting the parameters, we calculate P. deg =600×0.7-120=420-120=300kW, the diesel generator's rated power is 500kW, and the load rate is 60%, which is within the optimal range; when the load power drops to 400kW, P deg=400×0.7-80=280-80=200kW, load rate 40%, at this time the coordination control center commands the energy storage system to reduce the discharge power to 40kW, P deg =400×0.7-40=280-40=240kW, load rate 48%, still close to the optimal range, avoid low load operation of diesel generator.

[0056] (III) System Operation Performance Data Characterization Table 2 Comparison of System Operation Performance with Existing Diesel Generator Power Supply System in Example 2: Table 2 shows that the system of this invention has significant advantages over existing diesel generator power supply systems in terms of economy, environmental protection, and power quality. The unit cost of electricity is reduced from $0.8 / kWh to $0.45 / kWh due to the system's significant increase in the proportion of photovoltaic power from 0% to 45%, reducing diesel fuel consumption and lowering operation and maintenance costs; grid dependence is increased from 30% to 75% by using grid-connected / off-grid switching and energy storage to smooth out fluctuations, fully utilizing grid resources and reducing diesel generator start-up; the power quality compliance rate is improved from 80% to 99%, with PCS V / F control and diesel generator parameter adjustment ensuring stable output voltage and frequency, meeting the power quality requirements of production line equipment; and the annual maintenance cost of the diesel generator is reduced from $80,000 to $30,000. The system reduces the number of diesel generator start-ups and shutdowns, optimizes load rates, and lowers mechanical wear and maintenance requirements. Photovoltaic power accounts for 45% of the total, fully utilizing the park's abundant solar resources and reducing fossil fuel consumption. Load interruption time has decreased from an average of 120 hours per year to 4 hours, with energy storage backup, diesel generator emergency response, and rapid switching functions significantly reducing power outages. Total carbon emissions have decreased from an average of 1,500 tons of CO2 per year to 600 tons of CO2, with photovoltaic power replacing diesel generators reducing carbon emissions and meeting environmental protection requirements. The system deployment cycle has been shortened from 45 days to 20 days, with containerized integrated design and modular installation adapting to the park's rapid production needs.

[0057] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A hybrid energy storage power supply system for diesel engine replacement, characterized in that, Comprise: Photovoltaic array module: photovoltaic module configuration 10KV group string inverter, each inverter access 20-30 groups of components, integrated photovoltaic- energy storage combined MPPT control function, through energy storage coordination regulation when light mutation grid voltage fluctuation; Lithium iron phosphate energy storage module: configuration of bidirectional energy storage converter PCS, integrated V / F and PQ dual mode switching function, total capacity according to 1.5 times daily peak valley difference + 2 hours emergency power design, using active balancing technology to balance the battery; Diesel generator group: the diesel generator group as a bottom emergency power supply, according to the load gap hierarchical start, through load prediction algorithm to adjust the unit output in advance, maintain the unit load rate in 60%-80% interval; Charging unit: unit for 10KV / 380V charging array, adapt to the charging demand of mining area, support flexible adjustment power supply mode, according to the state of energy supply and demand dynamic adjustment of charging power; Coordination control center: the coordination control center is RB-I5martEMS coordination center, based on LSTM neural network to realize the two-dimensional prediction of photovoltaic output and load demand, 24 hours in advance to plan the coordinated scheduling strategy of photovoltaic- energy storage- diesel generator, through TSN time sensitive network to realize the data interaction and instruction issuing of each unit; 10KV bus module: the 10KV bus realizes the integrated management and control of photovoltaic array system, lithium iron phosphate energy storage system, diesel generator group and charging unit, reduces the transmission loss of long distance power transmission, and reduces the loss by 80% compared with 0.4KV low voltage transmission under the same power; Load distribution unit: the load distribution unit divides the load into primary production load and secondary auxiliary load, dynamically allocates power resources according to load priority and energy supply state, and realizes continuous power supply of load; Monitoring module: the monitoring module includes weather monitoring module and equipment state monitoring module, the weather monitoring module collects irradiance, temperature and wind speed data in real time, the equipment state monitoring module collects voltage, current, temperature and SOC operating parameters of each unit, and provides data support for the coordination control center; each module is deeply coupled through the 10KV bus module to build the operation mode of photovoltaic priority consumption, energy storage fluctuation suppression, diesel emergency backup and charging flexible adjustment.

2. The hybrid energy storage power supply system for diesel replacement of claim 1, wherein, Also include: Dynamic load simulator: the simulator is a programmed electronic load that supports step / slope load changes, simulates concentrated high load during mine production periods and daily electricity consumption level steady fluctuation scenarios; the coordination control center dynamically adjusts the energy storage system charge and discharge power change rate based on load simulation data and real-time data collected by the monitoring module through the formula dP / dt=k1(P pv -Pl)+k2(SOC max -SOC)-k3(SOC-SOC min ); wherein P is the real-time charge and discharge power of the energy storage system, t is time, k1 is the photovoltaic output response coefficient, P pv is the real-time output of the photovoltaic array, Pl is the real-time power of the load, k2 is the upper limit adjustment coefficient of the energy storage, SOC max is the maximum remaining capacity percentage of the energy storage system, SOC is the real-time remaining capacity percentage of the energy storage system, k3 is the lower limit adjustment coefficient of the energy storage, SOC min is the minimum remaining capacity percentage of the energy storage system, and real-time matching of energy storage charge and discharge, photovoltaic output and load fluctuation is achieved through dynamic adjustment.

3. The hybrid energy storage power supply system for diesel replacement of claim 1, wherein, Also include: Safety interlocking module: real-time monitoring of system voltage, frequency and temperature parameters, triggering three-stage protection mechanism of early warning, load reduction and shutdown; when the system voltage deviates from 10±5%kV, the frequency deviates from 50±0.2Hz or the equipment temperature exceeds 85℃, the early warning signal is first sent and reported to the coordination control center, if the parameters continue to deteriorate for more than 0.3 seconds, the load is automatically reduced by 30%, if the load is still not restored to normal after load reduction, the power supply of the fault unit is cut off within 0.5 seconds; the safety interlocking module and the coordination control center realize data interaction through CAN / Modbus dual communication protocol.

4. The hybrid energy storage power supply system for diesel replacement of claim 1, wherein, The 10KV-level string inverter of the photovoltaic array system integrates a photovoltaic-storage combined MPPT control algorithm, which tracks the maximum power point of the photovoltaic module and the charging and discharging state of the storage system, so that the utilization rate of photovoltaic output is increased to 95%; the bidirectional storage converter PCS adopts an improved droop control logic, which dynamically adjusts the output frequency through the formula ; wherein f is the output frequency of the PCS, f ref is the rated output frequency 50Hz, SOC ref is the target residual capacity percentage 80%, k sf is the SOC-frequency correlation coefficient, tau is the integral variable, P ref is the rated output power of the PCS, P is the real-time output power of the PCS, k pf is the power-frequency correlation coefficient, when SOC<20%, the output frequency is automatically increased by 0.5Hz through integral accumulation to trigger peak clipping, when SOC>80%, the frequency is reduced by 0.3Hz to absorb surplus power, and the system frequency is maintained stable.

5. The hybrid energy storage power supply system for diesel replacement of claim 1, wherein, The load prediction algorithm of the coordination control center adopts a combined prediction model based on historical load data and real-time monitoring parameters. By analyzing mine production plans, weather trends, and equipment operating state factors, the future 24-hour load demand is predicted. Based on the load prediction results, the coordination control center implements a hierarchical start-up strategy for the diesel generator set group. When the predicted load gap is less than 30% of the rated capacity, one unit is started; when the gap is between 30% and 60%, two units are started; and when the gap is greater than 60%, all units are started. At the same time, the energy storage system adjusts the charging and discharging power to supplement the remaining load.

6. The hybrid stored energy power supply system for diesel engine replacement of claim 1, wherein, The 10KV bus module is configured with an off-grid switching cabinet (STS). It adopts fast switch, self-detection, logic control, and integrated design of external communication, supports active switching and passive switching modes. When active switching, the coordination control center issues off-grid instructions, and the PCS switches to off-grid operation within 0.2 seconds, and the STS synchronously disconnects the fast switch to realize switching. When passive switching, the sampling board collects real-time grid voltage information. When voltage drop is detected, the control board confirms power failure within half a cycle and immediately controls the fast switch to break and issues switching instructions to the PCS to realize zero flash interruption of sensitive loads such as charging piles. The off-grid switching cabinet also has a synchronous grid connection function. When connected to the grid, it detects the grid voltage phase and amplitude, adjusts the PCS output parameters through the coordination control center, and completes grid connection after matching with the grid.

7. The hybrid stored energy power supply system for diesel engine replacement of claim 1, wherein, It also includes: Cloud monitoring module: The cloud monitoring module and the coordination control center realize data interaction through Ethernet, real-time display of system operation state, energy flow, and device parameter information, support remote control and fault diagnosis; the cloud monitoring module has a built-in mine scene strategy library, including daily power supply, high load production, and emergency power supply preset modes, which can be automatically adapted or manually switched according to user needs; when a fault occurs, the cloud monitoring module locates the fault type and position through the fault diagnosis algorithm, generates a maintenance guidance scheme, and pushes it to the management personnel; at the same time, the cloud monitoring module records system operation data, providing historical data support for strategy optimization of the coordination control center.

8. An energy management method for the hybrid energy storage power supply system for diesel engine replacement according to any one of claims 1-7, characterized in that, It includes: Pre-calibration of equipment parameters: establish photovoltaic MPPT efficiency model, correct temperature / irradiance coefficient to make error less than 2%, calibrate SOC-OCV curve of energy storage system by using piecewise linear fitting and R 2 >0.99, set parameters of diesel generator speed regulator and excitation system to make frequency fluctuation less than 0.5 Hz; Joint debugging: when the photovoltaic output P pv ≥80% load demand enters the photovoltaic dominant mode, the energy storage system is charged, and the diesel generator set is stopped, the target is photovoltaic utilization rate >95%, and energy storage charging efficiency >92%; when P pv <50% load and energy storage SOC <30% enters the diesel auxiliary mode, the diesel generator starts and carries the load to 70%, the energy storage system supplements the remaining power, the target is diesel load rate 40%-60%, and fuel consumption rate <220g / kWh; when the city power is simulated to be out of power, it enters the off-grid switching mode, the system switches to the off-grid mode within 0.2 seconds, the target is switching success rate >99%, and frequency stabilization time <1 second; Dynamic strategy optimization: based on historical operation data, train LSTM neural network, input meteorological monitoring data, load historical data, and device operating parameters, predict future 24-hour photovoltaic output and load demand, generate energy storage charging and discharging plan, optimize diesel generator set start-stop times to ≤3 times per day; Off-grid switching control: when connected to the grid, the coordination control center prioritizes photovoltaic and grid power to ensure load power consumption, and stores excess power in the energy storage system. When photovoltaic output is at its peak and the energy storage is full, the 10KV transmission line is controlled to deliver no more than 50kW to the grid; when off-grid, the energy storage system acts as a backup power source, combined with diesel generator sets to ensure power supply; Load dynamic allocation: according to load priority, primary production load is given priority to obtain power supply, and when energy supply is insufficient, secondary auxiliary load power supply is reduced through load distribution unit; Device state closed-loop management: the monitoring module collects the operation parameters of each unit in real time, uploads them to the coordination control hub and the cloud monitoring module, and triggers the protection mechanism of the safety interlocking module when the parameters exceed the preset threshold. At the same time, the coordination control hub adjusts the operation strategy, forming a closed-loop management of monitoring-analysis-control-feedback.

9. The energy management method for hybrid energy storage power supply system as claimed in claim 8, wherein, In the dynamic strategy optimization step, the photovoltaic output prediction adopts integral algorithm to realize quantization of irradiance cumulative effect, and the formula is used to calculate the photovoltaic prediction output in the future 24 hours; wherein P pv_pred is the 24-hour photovoltaic total prediction output, I(τ) is the prediction irradiance at τ time, η is the photovoltaic component photoelectric conversion efficiency, S is the total area of the photovoltaic array, α is the photovoltaic component temperature coefficient, ΔT(τ) is the deviation of the photovoltaic component temperature at τ time from the standard test temperature, and the integral operation is used to fully consider the comprehensive influence of irradiance and temperature at different time periods on the photovoltaic output.

10. The energy management method for hybrid energy storage power supply system as claimed in claim 8, wherein, In the joint debugging step, the load power of the diesel generating set is optimally matched through a dynamic adjustment algorithm, and when the load power changes, the coordination control center adjusts the load power according to the formula The output power of the diesel generating set is adjusted in real time; wherein P deg is the real-time output power of the diesel generating set, P load is the current total load power, β is the load rate optimization coefficient of the diesel generator, P bat is the real-time output power of the energy storage system, and the diesel generating set is always operated in the optimal load rate interval through dynamic adjustment; meanwhile, in the off-grid switching mode, the V / F control parameters of the energy storage system PCS and the excitation parameters of the diesel generating set are synchronously adjusted through the coordination control center, so that the voltage fluctuation control in the switching process is within ±5%.

Images