A control method for preventing overloading of a transformer of an energy storage system

Abstract

The present application relates to power generation, power transformation or power distribution technical field, disclose a kind of control method for preventing overload of energy storage system transformer, comprising: real-time monitoring the total power of three-phase and each phase power of power grid side electric meter, three-phase imbalance degree is calculated based on total power of three-phase when three-phase imbalance phenomenon;When three-phase imbalance degree is greater than preset imbalance threshold, using multi-threshold dynamic adjustment mechanism, according to the dynamic adjustment charging power of energy storage converter according to start threshold and limit charging threshold and forbidden charging threshold;The maximum deviation of each phase power and three-phase average power is overload phase by calculating, and it is reduced load based on ascending and descending load strategy, and it is ascending load to non-overload phase, and the adjusted three-phase power is obtained;Combined with dynamic response mechanism, power distribution instruction is calculated in real time, and three-phase imbalance degree does not exceed safety threshold in cooperation with total power control.It solves the problems of transformer overload, three-phase imbalance and slow response in the prior art, and achieves accurate control of load, high system stability and fast response.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power generation, power transformation or power distribution, in particular to a power supply or distribution circuit device or system. BACKGROUND

[0002] When the energy storage system accesses the power grid, the energy storage power station is charged, the energy storage power station is equivalent to a load operation, the energy storage power station is discharged, which is equivalent to providing a power supply, and the design of the energy storage power station is to utilize the remaining capacity of the existing transformer in the factory (the nominal capacity of the transformer minus the maximum load capacity of the factory). When the energy storage power station is charged, the transformer needs to accommodate not only the inherent load in the factory but also the power of the newly added energy storage power station. Due to the limited capacity of the transformer, two reasons can cause the transformer to be overloaded: when the load fluctuates greatly, it is easy to cause the total power of the transformer to be overloaded; when the load is unbalanced in three phases, it is easy to cause the single-phase overload of the transformer, which can cause local overheating and shorten the service life of the equipment. The traditional overload protection adopts a fixed threshold cut-off method, which affects the continuous operation of the system and the utilization rate of the energy storage equipment, and cannot dynamically balance the relationship between the load power, the energy storage charging power and the capacity of the transformer. In addition, when the three-phase load is unbalanced, the single-phase overload problem is difficult to solve effectively through total power control.

[0003] For example, a Chinese patent with publication number CN119651679A discloses a three-phase imbalance treatment method and a micro energy regulator, which provides the following technical solution. The present application relates to a three-phase imbalance treatment method and a micro energy regulator, which belongs to the technical field of power supply systems. When three-phase imbalance occurs, the three-phase imbalance degree is obtained according to the maximum phase and the power deviation between the phases and the three-phase total power. When the three-phase imbalance degree is greater than the preset imbalance degree, the single-phase power deviation of each phase is calculated according to the three-phase average power and the single-phase power of each phase. The three-phase total power is compared with the preset load rate. When it is greater, the energy storage is discharged, and when it is less, the energy storage is charged. When it is greater or less, the single-phase power deviation needs to be updated according to the corresponding design requirements. The energy storage determines the total discharge power or total power absorption of each phase according to the updated single-phase power deviation, proportional gain and integral gain. Each phase is discharged or absorbs power until the total discharge power or total power absorption of each phase is met. The present application can reduce the occurrence of heavy load or return phenomenon in the transformer area. However, the above-mentioned three-phase imbalance treatment method and micro energy regulator cannot dynamically realize power redistribution, lack a three-level threshold control mechanism, cannot effectively prevent single-phase overload, and the response is not fine enough, which can easily cause local overheating or overload problems of the transformer when the load fluctuates greatly. SUMMARY

[0004] The application solves the problems of easy overload, three-phase imbalance and slow response of the transformer in the prior art, and provides a control method for preventing overload of a transformer of an energy storage system, which achieves the purposes of accurately controlling the load, preventing local overheating, improving system stability and fast response.

[0005] To achieve the above object, the application adopts the following technical scheme:

[0006] The control method for preventing overload of the transformer of the energy storage system comprises the following steps:

[0007] Real-time monitoring of total three-phase power and power of each phase of the power grid side electric meter, and calculation of three-phase imbalance degree based on the total three-phase power when three-phase imbalance occurs;

[0008] When the three-phase imbalance degree is greater than a preset imbalance degree threshold value, a multi-threshold dynamic adjustment mechanism is adopted to dynamically adjust the charging power of the energy storage converter according to a start threshold value and a limit charging threshold value and a forbidden charging threshold value;

[0009] The phase with the largest deviation between the power of each phase and the average three-phase power is determined as an overload phase, and the overload phase is reduced in power based on a load change strategy, and the non-overload phase is increased in power to obtain adjusted three-phase power; and a dynamic response mechanism is used to calculate a power distribution instruction in real time, and the three-phase imbalance degree is controlled to be less than a safety threshold value in cooperation with total power control.

[0010] Through the combination of real-time monitoring, multi-level threshold dynamic adjustment and three-phase power redistribution algorithm, the problems of system interruption and low energy storage utilization caused by traditional fixed threshold cutting are effectively solved, the single-phase overload risk caused by three-phase imbalance is solved, and dual dynamic protection of the total power of the transformer and the single-phase power is realized, thereby improving the continuous operation capability of the system and the service life of the equipment.

[0011] As a preferred embodiment, the multi-threshold dynamic adjustment mechanism comprises a start threshold value, a limit charging threshold value and a forbidden charging threshold value; the start threshold value is a value below which the energy storage converter is allowed to start charging when the power grid output power is lower than the value; the limit charging threshold value is a value below which the power of the energy storage converter is dynamically adjusted to prevent the power grid output power from exceeding the value; and the forbidden charging threshold value is a value above which the energy storage converter is forced to stop charging when the power grid power exceeds the value.

[0012] The three clear power control points of the start threshold value, the limit charging threshold value and the forbidden charging threshold value are set to build a hierarchical progressive protection mechanism. The start threshold value allows charging to start within a safe range, thereby optimizing the energy storage utilization opportunity; the limit charging threshold value prevents the total power from approaching a dangerous area through dynamic adjustment; and the forbidden charging threshold value provides the final safety guarantee and forces shutdown in an extreme case. Compared with a single threshold value, the mechanism is more accurate in control, which can effectively prevent overload and maximize the reasonable operation time of the energy storage equipment.

[0013] As preferred, the step of dynamically adjusting the energy storage converter power specifically includes: if the grid power is greater than the forbidden charging threshold, the energy storage converter power is set to zero; if the grid power is greater than the start threshold and less than the forbidden charging threshold, the energy storage converter power is the minimum value between the maximum power of the energy storage converter and the limited charging threshold minus the current load power; otherwise, the energy storage converter power is set to the maximum power of the energy storage converter.

[0014] In the dangerous case that the grid power exceeds the forbidden charging threshold, the charging is forcibly stopped to protect the transformer; when the grid power is between the start threshold and the forbidden charging threshold, the charging power is intelligently limited by taking the minimum value to ensure that the grid output power does not exceed the limited charging threshold; when the grid power is lower than the start threshold, the energy storage converter is allowed to charge at full power to fully utilize the valley electricity. This logic ensures that the charging power and the transformer capacity are dynamically balanced under different load conditions.

[0015] As preferred, the identification of the overload phase specifically includes: calculating the deviation of each phase power from the three-phase average power in real time, and identifying the phase with the maximum deviation as the overload phase; and the reducing the power of the phase specifically includes reducing the amount by the amount of power to be transferred multiplied by the load shedding slope coefficient.

[0016] By calculating the deviation of each phase power from the average power and selecting the maximum value, the most serious overload phase can be accurately and quickly located. When reducing the power of the phase, a load shedding slope coefficient is introduced to control the amplitude and speed of power reduction, avoiding stepwise mutation of power, realizing smooth transition of power regulation, reducing the impact on the grid and equipment, and improving the stability of the system.

[0017] As preferred, the multi-threshold dynamic adjustment mechanism includes using a fuzzy logic controller to predictively adjust the power according to the load change trend; specifically, a fuzzy logic controller is used to analyze the load change trend in real time, including the rise rate and fall rate of the load power, to predict future grid power changes; based on the prediction results, the charging power is dynamically adjusted, and when the three-phase unbalance degree is greater than a preset unbalance degree threshold, a power regulation instruction is output by the fuzzy logic controller.

[0018] As preferred, the proportionally increasing the power of the non-overload phase according to the remaining capacity includes: distributing the amount of power to be transferred according to a weight, and the weight is the proportion of the remaining capacity of the non-overload phase to the total remaining capacity of all non-overload phases; and the weight distribution specifically includes that the power of the non-overload phase to be increased is equal to the reduction amount multiplied by the ratio of the remaining capacity of the phase to the sum of the remaining capacities of all non-overload phases.

[0019] The weight strategy of proportionally distributing the power to be transferred according to the residual capacity is adopted, and the power boost is preferentially distributed to the non-overload phase with larger residual capacity. This distribution mode fully utilizes the residual capacity of each phase transformer, effectively avoids the situation that other phases become new overload points when solving single-phase overload problems, ensures the balance and safety of power redistribution, and optimizes the overall utilization rate of transformer capacity.

[0020] As preferred, the input parameters of the fuzzy logic controller include current three-phase load power and historical load change trend and current power, the fuzzy rule is set based on load change rate, if the load rise rate is higher than the preset rise rate threshold, the charging power is reduced; if the load drop rate is lower than the preset drop rate threshold, the charging power is increased; the fuzzy output is obtained through defuzzification processing to obtain a specific power adjustment value; the fuzzy logic controller further includes preferentially reducing the power of the overload phase when the overload risk is high, and performing power redistribution in cooperation with the rise and fall load strategy, and the optimization strategy balances the response speed and stability through a weight factor.

[0021] As preferred, the dynamic response mechanism further includes a safety protection logic, specifically preferentially reducing the power of the overload phase and prohibiting the generation of new overload phases due to power redistribution.

[0022] Preferentially reducing the power of the overload phase ensures that the most urgent risk point can be quickly removed. Explicitly prohibiting the generation of new overload phases during power redistribution is a key safety constraint in the algorithm design, which fundamentally prevents the vicious cycle of "repairing the west wall by demolishing the east wall" and ensures the operation safety of the entire system during and after adjustment.

[0023] As preferred, the total power control specifically includes executing a forced energy storage converter stop charging command when the total power exceeds the forbidden charging threshold; when the total power does not exceed the forbidden charging threshold but there is a single-phase overload, executing a power redistribution command, and the power redistribution specifically includes identifying an overload phase whose power exceeds the single-phase safety margin, reducing the power of the phase, and proportionally increasing the power of the non-overload phase according to the residual capacity.

[0024] When the total power is severely out of limit, the highest level of protection measure, i.e. forced shutdown, is preferentially executed. When the total power is not severely out of limit but there is a local single-phase overload, a refined power redistribution strategy is started for internal adjustment. This hierarchical processing mechanism not only ensures safety in extreme cases, but also solves local imbalance problems within a relatively safe total power range, improving the efficiency and intelligence level of control.

[0025] As preferred, the ascending and descending load strategy comprises an ascending load strategy, specifically, the power amount to be transferred is distributed according to weights, and the weight is the proportion of the residual capacity of the non-overloaded phase to the sum of the residual capacities of all non-overloaded phases; specifically, the power of the non-overloaded phase to be raised is equal to the power amount to be transferred multiplied by the ratio of the residual capacity of the phase to the sum of the residual capacities of all non-overloaded phases.

[0026] As preferred, the ascending and descending load strategy comprises a descending load strategy, specifically, the power of the overloaded phase is reduced, and the reduction amount is equal to the power amount to be transferred multiplied by a descending slope coefficient.

[0027] As preferred, the three-phase unbalance degree specifically comprises obtaining the maximum power deviation between phases, dividing the quotient by the total power of three phases, and taking the average of multiple calculations as the final three-phase unbalance degree.

[0028] The most serious unbalance condition is highlighted, and dividing by the total power makes the measurement result comparable. Taking the average of multiple calculations effectively filters out the influence of instantaneous fluctuations or measurement noise, making the evaluation of the three-phase unbalance degree more stable, reliable and accurate, providing a solid basis for subsequent control decisions.

[0029] As preferred, after the identification of the overloaded phase, the power amount to be transferred is calculated, specifically, the power amount to be transferred is equal to the power of the overloaded phase minus the average power of three phases minus a safety margin.

[0030] The actual power of the overloaded phase, the average power level of the system and a preset safety margin are explicitly considered. The introduction of the safety margin provides necessary buffer space for system operation, avoiding frequent adjustments near the critical point, making the calculated power amount to be transferred more reasonable and having safety redundancy, guiding the subsequent power redistribution more accurate and effective.

[0031] As preferred, the real-time calculation of the power distribution instruction adopts a sliding window real-time calculation, the window length is 0.5-2 seconds, and the power distribution instruction is updated every 50-200 milliseconds.

[0032] The sliding window mechanism is adopted and real-time calculation is performed in a short window length, and the power distribution instruction is updated at a high frequency. The sliding window calculation can smooth the short-time power fluctuation and avoid the algorithm being too sensitive to noise. The update frequency of 50-200 milliseconds ensures that the control instruction can quickly respond to changes in load and power grid state, greatly improving the dynamic response speed and real-time control of the system, thereby more effectively responding to sudden load fluctuations and preventing overload.

[0033] Compared with the prior art, the present application has the following beneficial effects.

[0034] 1. The present invention effectively prevents transformer overall overload risk by monitoring three-phase power and total power in real time and setting multi-level threshold dynamic control of energy storage charging. The core lies in intelligent identification of single-phase overload and accurate calculation of power to be transferred, significantly improving three-phase imbalance problem and avoiding local overheating by reducing overload phase power and increasing non-overload phase power according to remaining capacity weight. Combined with safety protection logic, the adjustment process will not produce new overload phase, eliminating single-phase overload hidden danger from the root, greatly prolonging the service life of key equipment such as transformer.

[0035] 2. The present invention significantly improves the continuous operation capacity of the system and the utilization rate of the energy storage device. The three-level threshold mechanism allows the energy storage system to start charging or dynamically adjust power under wider load conditions on the premise of safety, maximizing the use of valley period for charging. The power redistribution algorithm solves the three-phase imbalance while reasonably distributing the charging and discharging power of the energy storage, fully utilizing the remaining capacity of each phase of the transformer, avoiding the waste of idle capacity, and improving the overall energy utilization efficiency.

[0036] 3. The present invention has excellent dynamic response capability based on real-time calculation and high-frequency power instruction update, and can quickly track load fluctuations and grid state changes. Its total power control and single-phase power redistribution mechanism work together, with clear priority management. When the total power is severely over-limited, it is forced to shut down to provide the highest protection, and when there is local single-phase overload, it starts fine adjustment. This fast, smooth and intelligent power distribution effectively suppresses the impact of power mutation on the power grid, maintains three-phase balance and system voltage stability, and ensures the reliable and stable operation of industrial and commercial power and the energy storage system itself. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 The present invention is a kind of energy storage system transformer anti-overload control method.

[0038] Figure 2 The present invention is a kind of energy storage system transformer anti-overload control method.

[0039] Figure 3 The present invention is a kind of energy storage system transformer anti-overload control method.

[0040] Figure 4 The present invention is a kind of energy storage system transformer anti-overload control method. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical scheme and advantages of the present disclosure clearer, the following will further describe the embodiments of the present disclosure in detail with reference to the drawings. The proportions of various components are not drawn according to the true proportions, and the proportions and sizes shown in the drawings should not limit the essential technical scheme of the present disclosure. These embodiments do not describe all the details, nor limit the present disclosure to the specific embodiments described.

[0042] Referring to Figures 1-4 As shown in the figure, a control method for preventing overload of a transformer of an energy storage system comprises the following steps:

[0043] Real-time monitoring of the total power and the power of each phase of the power meter on the grid side, and calculation of the three-phase imbalance degree based on the total power when three-phase imbalance occurs;

[0044] When the three-phase imbalance degree is greater than the preset imbalance threshold, a multi-threshold dynamic adjustment mechanism is used to dynamically adjust the charging power of the energy storage converter according to the start threshold and the limit charging threshold and the forbidden charging threshold;

[0045] The phase with the largest deviation between the power of each phase and the average power of three phases is determined as the overload phase, and the load of the overload phase is reduced and the load of the non-overload phase is increased based on the load increase and decrease strategy to obtain the adjusted three-phase power; and the power distribution instruction is calculated in real time based on the dynamic response mechanism, and the three-phase imbalance degree is controlled to be less than the safety threshold in cooperation with the total power control.

[0046] In order to solve the problem of transformer overload, the energy storage EMS communicates with the anti-overload power meter through the ModbusRTU protocol, reads the three-phase load and single-phase load data, and communicates with the energy storage PCS through the ModbusTCP protocol, controls the power of the energy storage PCS in a separate phase, and when the load fluctuation is large, the present application uses a multi-threshold dynamic adjustment of the energy storage charging power to dynamically balance the relationship between the energy storage charging power and the transformer capacity.

[0047] As Figure 1 and Figure 3 As shown in an embodiment, Figure 1 The present application is a control method for preventing overload of a transformer of an energy storage system, Figure 3 The present application is a total power control coordination flowchart of a control method for preventing overload of a transformer of an energy storage system. First, the present application monitors the total power of three phases and the power of each phase of the power meter on the grid side in real time. If three-phase imbalance is detected, the current three-phase imbalance degree is calculated based on the total power of three phases.

[0048] Then, it is determined whether the three-phase imbalance degree is greater than the preset safety threshold. If it is greater, a multi-threshold dynamic adjustment mechanism is triggered. The mechanism uses three key thresholds:

[0049] Start threshold: when the grid output power is lower than this value, the energy storage converter is allowed to start charging.

[0050] Charge limit threshold: dynamically adjusts the target upper limit of the energy storage inverter power, ensures that the grid output power does not exceed this value.

[0051] Charge prohibition threshold: when the grid power exceeds this value, the energy storage inverter is forced to stop charging.

[0052] Then, according to the current grid power value, dynamically adjust the total charging power of the energy storage inverter:

[0053] If the grid power is greater than the charge prohibition threshold, set the energy storage inverter power to zero, and force the charging to stop.

[0054] If the grid power is greater than the start threshold but less than or equal to the charge prohibition threshold, set the energy storage inverter power to the smaller of the energy storage inverter maximum power and the charge limit threshold minus the current load power.

[0055] If the grid power is less than or equal to the start threshold, set the energy storage inverter power to the energy storage inverter maximum power.

[0056] At the same time, in order to solve the problem of single-phase overload caused by three-phase imbalance:

[0057] Identify the overload phase: real-time calculation of the deviation of each phase power from the three-phase average power, and identify the phase with the largest deviation as the overload phase whose power exceeds the single-phase safety margin.

[0058] Calculate the transfer amount: subtract the sum of the overload phase power and a safety margin value from the three-phase average power to get the amount of power that needs to be transferred.

[0059] Reduce the power of the overload phase: reduce the power of the identified overload phase by an amount equal to the amount of power that needs to be transferred multiplied by a load reduction slope coefficient.

[0060] Increase the power of non-overload phases: proportionally increase the power of non-overload phases according to the remaining capacity. Specifically, calculate the remaining capacity of each non-overload phase (transformer single-phase rated capacity minus the current total power of the phase), and then distribute the transferred power to each non-overload phase according to the proportion of each non-overload phase's remaining capacity to the total of all non-overload phase's remaining capacities. That is, the power of each non-overload phase is increased by an amount equal to the reduction amount multiplied by the ratio of the phase's remaining capacity to the sum of all non-overload phase's remaining capacities.

[0061] In the process of performing power dynamic adjustment and redistribution, the system combines a dynamic response mechanism and uses a sliding window to calculate power allocation instructions in real time, with a typical window length of 0.5 seconds to 2 seconds, and the instructions are updated every 50 milliseconds to 200 milliseconds. At the same time, follow the safety protection logic, preferentially reduce the power of the overload phase, and strictly prohibit any new overload phase from being generated due to power redistribution.

[0062] Finally, the above dynamic power regulation (three-level threshold control) works in coordination with single-phase power redistribution. When the total three-phase power exceeds the forbidden charging threshold, the total power control takes priority, and the forced energy storage converter stop charging command is executed. When the total three-phase power does not exceed the forbidden charging threshold but the monitoring identifies that there is a single-phase overload, the single-phase power redistribution command is executed in priority, the overload phase is identified, its power is reduced, and the power of the non-overload phase is increased in proportion to the remaining capacity, to ensure that the final three-phase unbalance does not exceed the safety threshold. The specific calculation method of the three-phase unbalance is to obtain the power deviation between the maximum phase and the phase, then divide the average value of the total three-phase power to obtain the quotient, and take the average value of multiple calculations as the final three-phase unbalance.

[0063] The energy storage system transformer overload prevention control method provided by the application realizes the deep integration of the innovative three-level threshold dynamic adjustment mechanism and the intelligent single-phase power redistribution algorithm, and provides a revolutionary solution for the safe and stable operation of the energy storage system connected to the power grid. The core value lies in fundamentally solving the industry pain point that the traditional overload protection method is difficult to balance the safety of the transformer and the efficient use of the energy storage device. The application first constructs a dynamic protection system with the synergistic effect of the start threshold, the limit charging threshold and the forbidden charging threshold. When the power grid output power is lower than the start threshold, the energy storage converter is allowed to charge at the maximum capacity, fully tapping the economic benefits of low-price valley electricity storage; when the power rises between the start threshold and the limit charging threshold, the system automatically adjusts the power of the energy storage converter, accurately controls the power grid output power close to but never exceeding the limit charging threshold, and ensures that the total power of the transformer operates within the safety margin; once the power grid power breaks through the forbidden charging threshold, the energy storage converter is immediately forced to stop, providing the highest level of instantaneous overload protection for the transformer. This hierarchical response mechanism not only completely avoids the frequent interruptions caused by the traditional fixed threshold cut-off, significantly improves the continuous operation time and charge-discharge cycle number of the energy storage system, but also maximizes the utilization of the remaining capacity potential of the transformer through fine power dynamic management, so that the energy storage device realizes almost extreme utilization rate within the framework of absolute safety, bringing higher return on investment for users.

[0064] More importantly, the application solves the problem of three-phase load imbalance that has long plagued transformers through a single-phase power redistribution algorithm. By monitoring grid-side meter data in real time, the application continuously calculates the deviation of each phase power from the three-phase average power and accurately locates the overload phase that exceeds the single-phase safety margin. The core of the algorithm is an intelligent power transfer strategy: on the one hand, it scientifically reduces the output power of the energy storage converter in the overload phase according to the load shedding slope coefficient, achieving rapid and flexible reduction of overload current; on the other hand, it innovatively dynamically increases the power input of non-overload phases in proportion to the remaining capacity, with the distribution weight strictly following the ratio of the remaining capacity of each non-overload phase to the total remaining capacity of all non-overload phases. This directional power scheduling ensures that every kilowatt of power transferred from the overload phase is accurately injected into the non-overload phase with the most abundant carrying capacity, eliminating the risk of local overheating and avoiding a new round of imbalance caused by "patching the east wall with the west wall". With the built-in safety protection logic of forced priority load shedding in the overload phase and strict prohibition of new overload phase generation, the algorithm can continuously and stably control the three-phase imbalance to below 15% of the safety threshold, effectively suppressing the rapid increase of transformer iron loss and accelerated insulation aging caused by single-phase overload, and significantly extending the service life of key power equipment.

[0065] The outstanding performance of the control method of the application also lies in its millisecond-level dynamic response capability. It uses a sliding window technique with continuous refresh within a specific time interval for real-time calculation, with a window length of usually between half a second and two seconds, and power distribution instructions are updated at a high speed of 50 milliseconds to 200 milliseconds per cycle. This high-frequency closed-loop control enables the system to instantly capture the instantaneous fluctuations of the load and subtle changes in the grid state, whether it is facing power surges caused by sudden start-stop of factory equipment or seasonal sharp increase in air conditioning load of commercial complexes, and can achieve instantaneous dynamic balance of transformer load. It is particularly worth emphasizing that the system realizes seamless cooperation and intelligent priority management of total power control and single-phase power redistribution. When the total three-phase power exceeds the limit, the total power protection mechanism has the highest priority and immediately executes the global shutdown instruction; when the total power is not over-limit but single-phase overload is detected, the single-phase power redistribution algorithm is started first to eliminate the hidden danger in its infancy through intelligent scheduling of inter-phase power. This hierarchical and cooperative protection strategy builds a three-dimensional protection network that covers the whole and the local, and takes into account both steady state and transient state.

[0066] Overall, the technical benefits brought by the present application are comprehensive. In terms of safety, it builds a double insurance consisting of total power three-level threshold protection and single-phase dynamic redistribution, almost eliminating the risk of transformer overload damage and the resulting fire risk; in terms of operating efficiency, it discards the simple and rough shutdown protection, significantly improves the charge-discharge operation time of the energy storage device and the capacity utilization rate of the transformer through flexible adjustment and intelligent distribution of power; in terms of economy, extending the life of the transformer reduces the cost of equipment replacement, increasing the throughput of energy storage increases the peak-valley price difference income, and reducing the frequency of operation and maintenance intervention saves labor costs; in terms of system adaptability, its dynamic threshold setting and parameter adjustable characteristics can be flexibly adapted to different capacity transformers and diversified load scenarios. Therefore, this technology not only provides essential safety protection for the current large-scale development of energy storage, but also optimizes the economic efficiency of system operation through intelligent energy management, and has great practical value and broad application prospects for promoting the high-quality development of the energy storage industry and building a highly flexible new power system.

[0067] In another embodiment, the present application proposes a multi-threshold dynamic adjustment based energy storage system transformer overload prevention control method, which monitors the three-phase total power and each phase power of the power grid side electric meter in real time through EMS, intelligently adjusts the charging power of PCS, and realizes the dynamic balance of the transformer load. Under the total power three-level threshold control mechanism, the single-phase power redistribution algorithm is nested, that is, the total power is dynamically changed, and the single-phase power redistribution is intelligently controlled for dynamic adjustment of three-phase power. The core goal is to control the three-phase unbalance degree within the safety threshold, usually ≤15%, by adjusting the output power of the energy storage PCS in each phase in real time.

[0068] Total power three-level threshold control mechanism:

[0069] Start threshold (e.g. 145kW): allows PCS to start charging when the grid output power is below this value;

[0070] Charging limit threshold (e.g. 170kW): dynamically adjusts PCS power to make grid output power not exceed this value;

[0071] Charging prohibition threshold (e.g. 200kW): forces PCS to stop charging when the grid power exceeds this value.

[0072] Dynamic adjustment algorithm, control curve diagram reference Figure 4 as shown:

[0073] IF grid power > charging prohibition threshold THEN

[0074] PCS power = 0

[0075] ELSE IF grid power > start threshold THEN

[0076] PCS power = min(PCS max power, limit charge threshold - current load power)

[0077] ELSE

[0078] PCS power = PCS max power

[0079] END IF.

[0080] Single-phase power redistribution algorithm mechanism:

[0081] The basic principle of the algorithm is:

[0082] 1. Input parameters:

[0083] Real-time monitoring of three-phase power values; single-phase rated capacity of transformer; current PCS output power in each phase; load power.

[0084] 2. Control objectives:

[0085] Calculate the three-phase imbalance: get the maximum phase-to-phase power deviation, divide by the total three-phase power to get the quotient, and take the average of multiple calculations as the final three-phase imbalance.

[0086] 2.2.1, algorithm implementation steps

[0087] Step 1: Overload phase identification, specific program (python) as follows:

[0088] def identify_overloaded_phase():

[0089] phase_powers = [Pa, Pb, Pc] # Get three-phase power values (load + PCS)

[0090] avg_power = sum(phase_powers) / 3 # Calculate average power

[0091] imbalance = [abs(p - avg_power) for p in phase_powers] # Calculate each phase imbalance

[0092] overload_phase = np.argmax(imbalance) # Find the phase corresponding to the maximum value

[0093] return overload_phase # Return the phase number with the largest imbalance, 0 for A phase, 1 for B phase, and 2 for C phase.

[0094] Step 2: Power difference calculation

[0095] Calculate the amount of power to be transferred: subtract the sum of the average power and the safety margin from the overload phase power. (For example: if the B phase power is 120 kW, the average power is 100 kW, and the safety margin is 10 kW, then 10 kW needs to be transferred).

[0096] Step 3: Power redistribution strategy

[0097] 1. Load shedding strategy:

[0098] Reduce the PCS overload phase power: the reduction amount is equal to the amount of power to be transferred multiplied by the load shedding slope coefficient.

[0099] 2. Load boosting strategy:

[0100] Boost the PCS non-overload phase power according to the weight: the weight is the proportion of the remaining capacity of the non-overload phase to the sum of the remaining capacities of all non-overload phases; the weight distribution is as follows: the power of the non-overload phase to be boosted is equal to the reduction amount multiplied by the ratio of the remaining capacity of the phase to the sum of the remaining capacities of all non-overload phases (preferentially boost the phase with low load rate). Physical meaning: the greater the remaining capacity of the phase, the higher the proportion of power increment allocated to it.

[0101] Step 4: Dynamic limiting protection, refer to Figure 2 for details.

[0102] The technical features of the present application are as follows: it has dynamic responsiveness: real-time calculation is performed using a sliding window (typical window 1 second), and the power distribution instruction is updated every 100 ms; it also has safety protection logic: the overload phase is preferentially reduced in power, the non-overload phase is boosted in power according to the capacity proportion, and "splitting east to fill west" to cause new overload phase is prohibited. It also includes coordination with total power control: when the total power exceeds the forbidden charging threshold, the forced energy storage converter stop charging command is executed; when the total power does not exceed the forbidden charging threshold but there is a single-phase overload, the power redistribution command is executed, and the power redistribution is specifically to identify the overload phase whose power exceeds the single-phase safety margin, reduce the power of the phase, and boost the power of the non-overload phase according to the remaining capacity proportion.

[0103] In one embodiment, the scenario is a certain shopping mall energy storage system (transformer 400 kVA)

[0104] Initial state:

[0105] A phase: load 80 kW + PCS 60 kW = 140 kW;

[0106] B phase: load 50 kW + PCS 30 kW = 80 kW;

[0107] C phase: load 60 kW + PCS 40 kW = 100 kW.

[0108] Algorithm execution:

[0109] 1. Identify A-phase overload;

[0110] 2. Calculate the amount to transfer: ΔP = 140 - (106.7 + 10) ≈ 23.3 kW;

[0111] 3. Reduce A-phase PCS power: 60 → 40 kW (down 20 kW);

[0112] 4. Increase B / C-phase PCS power:

[0113] B-phase: 30 → 40 kW (+10 kW)

[0114] C-phase: 40 → 50 kW (+10 kW).

[0115] Final state:

[0116] A-phase: 80 + 40 = 120 kW (safe);

[0117] B-phase: 50 + 40 = 90 kW;

[0118] C-phase: 60 + 50 = 110 kW;

[0119] (Unbalance degree from 32% to 12%).

[0120] In another embodiment, a fuzzy logic controller is used instead of fixed thresholds, specifically:

[0121] The input variables are real-time three-phase load power change rate, historical load trend, and current three-phase unbalance degree deviation. The membership function is divided into {negative large, negative small, zero, positive small, positive large} for the change rate, and a triangular function is used to cover the set interval. The unbalance degree deviation is divided into {low, medium, high}, and a trapezoidal function is used to cover the set interval.

[0122] A fuzzy rule base is constructed, for example:

[0123] When the load change trend is "positive large" and the unbalance degree deviation is "high", the power adjustment instruction is "rapidly reduce charging power";

[0124] When the load change trend is "positive small" and the unbalance degree deviation is "medium", the power adjustment instruction is "slowly reduce charging power";

[0125] When the load change trend is "zero" and the unbalance degree deviation is "low", the power adjustment instruction is "maintain current power";

[0126] When the load change trend is "negative and large" and the imbalance deviation is "high", the power adjustment command is "gradually increase the power of the non-overload phase".

[0127] This process continues with 25 rules, prioritizing responses to combinations with high overload risk. The center-of-gravity method is used to calculate precise power adjustment values, with the output range between the positive and negative values ​​of the maximum power of a single phase in the PCS. Based on this, a dynamic prediction mechanism can be implemented. In this embodiment, the power demand for the next 2 seconds is extrapolated based on the load change rate of the previous 5 seconds. If the predicted overload probability is >80%, a load reduction command is triggered in advance.

[0128] This invention also features an overload risk classification response, categorized into high risk (imbalance > 25%), medium risk (imbalance 15%~25%), and low risk (imbalance < 15%). In the high-risk scenario, the fuzzy controller immediately outputs a negative power adjustment value, and the load reduction strategy employs an overload phase load reduction slope of approximately 1 for full-speed load reduction. In the medium-risk scenario, the fuzzy controller outputs a negative power adjustment value, and in this embodiment, the load reduction strategy employs an overload phase load reduction slope of 0.7. In the low-risk scenario, only the fuzzy controller is activated for maintenance or fine-tuning.

[0129] The optimization strategy balances response speed and stability through weighting factors, specifically by multiplying the stability factor by the stability parameter and then adding the response speed factor by the fast response parameter.

[0130] The stability parameter is calculated as follows: First, calculate the rated capacity of a single phase of the transformer minus the total load power of that phase to obtain a difference (the remaining capacity of that phase). Then, calculate the sum of the differences between the rated capacity of a single phase of the transformer and the total load power of that phase for all non-overloaded phases (the sum of the remaining capacities of all non-overloaded phases). Divide the difference (the remaining capacity of that phase) by the sum of the differences (the sum of the remaining capacities of all non-overloaded phases) to obtain a ratio. Then, multiply the total power to be transferred by the obtained ratio. Finally, add the current charging power of the energy storage converter in that phase. The fast response parameter is calculated as follows: First, calculate the load descent slope coefficient multiplied by the total power to be transferred. Then, subtract the calculated result of the load descent slope coefficient multiplied by the total power to be transferred from the current charging power of the energy storage converter in that phase.

[0131] like Figure 2 In one embodiment shown, Figure 2 This is a flowchart illustrating the dynamic limiting protection of a transformer overload prevention control method for an energy storage system according to the present invention. The specific program (Python) is as follows:

[0132] # Ensure that each phase does not exceed the limit after adjustment

[0133] for phase in [a, b, c]:

[0134] P_pcs[phase] = clamp(P_pcs[phase], 0, P_pcs_max) # limit in 0~PCS max power

[0135] P_total[phase] = P_load[phase]+ P_pcs[phase] # add load power and PCS charging power

[0136] assert P_total[phase]<= 1.05 * P_rated # check total power not exceed 105% rated power, leave 5% margin.

[0137] As Figure 4 shown in an embodiment, Figure 4 is a control curve diagram of the control method of the energy storage system transformer overload prevention of the application. Among them, the unbolded curve (the curve with a triangle shape with the time axis) is the total power of the three-phase anti-backflow ammeter during charging. Another bolded curve, the time axis from left to right, the first straight line indicates that the anti-overload ammeter power P is less than the starting threshold value, and the energy storage system starts charging; then the first endpoint of the second downward broken line indicates that the anti-overload ammeter power P is greater than the limited charging threshold value, and the energy storage system starts to reduce the charging power; the second endpoint indicates that the anti-overload ammeter power P is greater than the forbidden charging threshold value, and the energy storage system prohibits charging; the next straight line with zero grid power indicates that the anti-overload ammeter power P is less than the starting threshold value, and the energy storage system starts charging.

[0138] In summary, the energy storage system transformer overload prevention control method proposed by the application realizes the dual goals of safe operation of the transformer and efficient use of the energy storage through the synergistic effect of the three-level threshold dynamic adjustment mechanism and the single-phase power redistribution algorithm. The method first monitors the three-phase total power and single-phase power of the grid-side ammeter in real time. When it is detected that the three-phase imbalance exceeds the safety threshold, a multi-level regulation consisting of a starting threshold, a limited charging threshold and a forbidden charging threshold is triggered: if the grid power is below the starting threshold, the energy storage converter is allowed to charge at full power; if the power is between the starting threshold and the forbidden charging threshold, the energy storage converter power is dynamically adjusted to make the grid output not exceed the limited charging threshold; if it exceeds the forbidden charging threshold, it is forced to stop. For the three-phase imbalance problem, the system accurately identifies the overloaded phase whose power exceeds the safety margin, reduces the power of this phase by a load reduction slope coefficient, and distributes an equal amount of power to the non-overloaded phases according to the proportion of the remaining capacity of each non-overloaded phase to the total remaining capacity of all non-overloaded phases. The whole process adopts a 50-200 millisecond level dynamic response mechanism, calculates the power command in real time through a sliding window, and implements double safety protection: when the total power exceeds the forbidden charging threshold, the global shutdown is executed first; when there is a single-phase overload, the power redistribution is started first, strictly avoiding the generation of a new overloaded phase.

[0139] The present application is not limited to the above-described embodiments, and any changes in shape or material composition are permissible as long as the structure design provided by the present application is adopted.

Claims

1. A control method for preventing overloading of a transformer of an energy storage system, characterized by, The application relates to a three-phase power distribution method and device for a three-phase power distribution system. The method comprises the following steps: real-time monitoring of the total three-phase power and the power of each phase of a power grid side electric meter, and calculation of the three-phase unbalance degree based on the total three-phase power when a three-phase unbalance phenomenon occurs; when the three-phase unbalance degree is greater than a preset unbalance degree threshold value, a multi-threshold dynamic adjustment mechanism is adopted to dynamically adjust the charging power of a storage energy converter according to a starting threshold value and a limited charging threshold value and a forbidden charging threshold value; the starting threshold value is a value below which the storage energy converter is allowed to start charging when the power grid output power is lower than the value; the limited charging threshold value is a value below which the power grid output power is dynamically adjusted to be not higher than the value; the forbidden charging threshold value is a value above which the storage energy converter is forced to stop charging when the power grid power is higher than the value; if the power grid power is greater than the forbidden charging threshold value, the storage energy converter power is set to zero; if the power grid power is greater than the starting threshold value and smaller than the forbidden charging threshold value, the storage energy converter power is the minimum value of the storage energy converter maximum power or the limited charging threshold value minus the current load power; otherwise, the storage energy converter power is set to the storage energy converter maximum power. The method further comprises the following steps: calculating the maximum deviation of each phase power from the three-phase average power to obtain an overload phase; based on a load raising and lowering strategy, the overload phase is lowered and the non-overload phases are raised to obtain adjusted three-phase power; and combining a dynamic response mechanism to real-time calculate a power distribution instruction, and cooperating with the total power control to make the three-phase unbalance degree not exceed a safety threshold value. The multi-threshold dynamic adjustment mechanism comprises a starting threshold value, a limited charging threshold value and a forbidden charging threshold value. The multi-threshold dynamic adjustment mechanism comprises a fuzzy logic controller which is used to predictively adjust the power according to the load change trend; specifically, the fuzzy logic controller is used to analyze the load change trend in real time, including the load power rising rate and the load power falling rate, to predict the future power grid power change; based on the prediction result, the charging power is dynamically adjusted, and when the three-phase unbalance degree is greater than the preset unbalance degree threshold value, the fuzzy logic controller outputs a power adjustment instruction.

2. The control method of claim 1, wherein, The input parameters of the fuzzy logic controller include the current three-phase load power, the historical load change trend and the current power, the fuzzy rule is set based on the load change rate, if the load rising rate is higher than a preset rising rate threshold value, the charging power is reduced; if the load falling rate is lower than a preset falling rate threshold value, the charging power is increased; the fuzzy output is obtained through defuzzification processing to obtain a specific power adjustment value; the fuzzy logic controller further comprises preferentially reducing the overload phase power when the overload risk is high, and cooperating with the load raising and lowering strategy to perform power redistribution, and the optimization strategy balances the response speed and stability through a weight factor.

3. The control method of claim 1, wherein, The load raising and lowering strategy comprises a load lowering strategy, specifically, the overload phase power is reduced, and the reduction amount is equal to the power amount to be transferred multiplied by a load lowering slope coefficient.

4. The control method of claim 3, wherein, The load raising and lowering strategy comprises a load raising strategy, specifically, the power amount to be transferred is distributed according to a weight, and the weight is the proportion of the residual capacity of the non-overload phase to the total residual capacity of all non-overload phases; the weight distribution is specifically that the power of the non-overload phase to be raised is equal to the reduction amount multiplied by the ratio of the residual capacity of the non-overload phase to the sum of the residual capacities of all non-overload phases.

5. The control method of claim 4, wherein, The calculation of the three-phase unbalance degree specifically comprises the following steps: obtaining the maximum power deviation between the maximum phase and the phases, and dividing the maximum power deviation by the total three-phase power to obtain the three-phase unbalance degree.

6. The control method of overloading prevention of an energy storage system transformer according to claim 5, characterized in that, ​ 7. The control method of overloading prevention of an energy storage system transformer according to claim 1, characterized in that, ​ 8. The control method of overloading prevention of an energy storage system transformer according to claim 1, characterized in that, After the overload phase is identified, the amount of power to be transferred is calculated, specifically, the amount of power to be transferred is equal to the overload phase power minus the three-phase average power minus a safety margin.

9. The control method of claim 8, wherein, The real-time power distribution instruction is calculated using a sliding window, and the window length is 0.5-2 seconds, and the power distribution instruction is updated every 50-200 milliseconds.

Images