Scheduling method of modular power supply system

Abstract

The invention discloses a scheduling method of a modular power supply system, and mainly solves the technical problems of low scheduling operation efficiency and poor reliability of an existing power module. According to the scheduling method based on the energy efficiency and the service life of the collaborative system, operation cost (including electric charge) is reduced, meanwhile, long-term capital expenditure (including module replacement) and maintenance cost are reduced, and minimization of the total cost in the whole life cycle of a modular power system is achieved; and a weight factor during total cost calculation allows an operator to flexibly adjust according to actual conditions, such as time-of-use electricity price, seasonal change, maintenance plan and the like, so that dynamic balance is obtained between the energy efficiency and the service life of the system. Therefore, the system can select the optimal scheduling operation strategy according to the real-time load change and the module health condition, and has high flexibility and self-adaptive capability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of direct current power supply system, and particularly relates to a scheduling method of a modular power supply system. BACKGROUND

[0002] In the application fields of modular direct current power supply system (such as new energy vehicle charging station and the like), such as high-power direct current charging pile, energy storage system, data center power supply and the like, the high-power direct current power supply system generally adopts the architecture of multiple power modules working in parallel, so as to improve the power output and system redundancy.

[0003] However, the present inventors have found in the process of implementing the technical solutions in the embodiments of the present application that the scheduling of the modular high-power direct current power supply system generally adopts a static or simple allocation strategy, for example, a fixed priority order (such as always starting No. 1 first, and then sequentially starting No. 2 and the like) or a simple equal distribution strategy to meet the demand of the load. Such static strategy is difficult to adapt to the changing load and system state, thereby resulting in poor running efficiency and reliability of the system.

[0004] Among them, since the power conversion efficiency of the power module is not a simple linear relationship, the peak efficiency usually appears in a specific medium load interval, that is, light load or full load will cause the power conversion efficiency of the power module to decrease significantly. The simple equal distribution strategy is prone to cause the power modules to work in the low-efficiency light load area for a long time, or cause a small number of modules to work in the low-efficiency full load area, thereby causing significant waste of electric energy and additional heat loss, and increasing the operation cost of the system. In addition, the strategy of fixed priority order will cause the cumulative running time, fan wear and tear, and thermal cycle number of the power devices of the high-priority module to be much higher than those of other power modules. Such uneven use will accelerate the aging and premature failure of part of the modules, thereby reducing the mean time between failures and reliability of the entire system, and increasing the maintenance cost of the system.

[0005] The information disclosed in this BACKGROUND section is only for the purpose of enriching the understanding of the background of the present disclosure and should not be considered as recognition or implicit admission that such information constitutes prior art to the present disclosure. SUMMARY

[0006] In view of at least one of the above technical problems, the present disclosure provides a scheduling method of a modular power supply system, mainly solving the technical problem of low running efficiency and poor reliability of the existing power module scheduling.

[0007] According to one aspect of the present disclosure, a scheduling method of a modular power supply system is provided, which comprises the following steps: (1) Establish energy efficiency curve models of power modules under different load rates and module health models including cumulative running time and cumulative start-stop times, and store the two models in the system controller. (2) Obtain the total power required for the charging gun P t And query the number of all currently available power modules in the system. N and the cumulative running time of each power module in the module health model T i ,in i = 1 … N ; (3) Normalize the cumulative running time of each power module to obtain the normalized value of the cumulative running time of each power module. t i ,in i = 1 … N ; (4) Based on the total power required by the charging gun P t and the rated power of the power module P r Calculate the initial value for optimization. m ,in, m To meet m × P r > P t The minimum value and m ≤ N ; (5) Calculate the load factor Z = P t / ( m × P r After multiplying by 100%, the power module corresponding to the energy efficiency curve model is queried based on the load rate. Z The corresponding efficiency H Corresponding computational efficiency cost Ce =1-( H / H max ),in H max Efficiency in the energy efficiency curve model H The maximum value; select after sorting. m The power module with the shortest cumulative running time was selected, and its corresponding normalized value was obtained. t i , with that m The average value of the normalized values ​​of each power module is used as the lifespan cost. Cl ; (6) Calculate the total cost when corresponding start m power modules TC m = w 1 × Ce + w 2 × Cl , wherein w 1 is the energy efficiency weight, w 2 is the life weight; (7) Let m = m + 1 , cycle steps (5) and (6) until m = N ; compare the total cost when starting different numbers of power modules, and obtain the minimum total cost corresponding to starting j power modules TC min ; obtain the minimum total cost TC min corresponding to the optimal power module set containing j power modules; (8) After obtaining the load power module set that has been served for the load, compare the load power module set with the optimal power module set, and adjust the state of the corresponding power module according to the comparison result to make only the power modules in the optimal power module set participate in the load service at the corresponding power; (9) The system controller updates the module health model and the real-time state corresponding to each power module; (10) Repeat steps (2)-(9) when the period is fixed and / or the load demand changes.

[0008] In some embodiments of the present disclosure, in the step (1), the energy efficiency curve model includes the efficiency of the power module corresponding to a limited load rate, and is stored in the non-volatile memory of the system controller in the form of a lookup table.

[0009] In some embodiments of the present disclosure, in the step (5), the lookup table corresponding to the energy efficiency curve model uses the table lookup method and the linear interpolation method to correspondingly query the efficiency of the power module at any load rate Z . H .

[0010] In some embodiments of the present disclosure, in the step (3), let the maximum value of the cumulative running time of each power module be T max , then t i =T i / T max wherein i = 1 … N .

[0011] In some embodiments of the present disclosure, in the step (8), the load power module set and the optimal power module set are differentially compared, and according to the comparison result, power modules belonging to the load power module set but not belonging to the optimal power module set are powered off, power modules not belonging to the load power module set but belonging to the optimal power module set are started, and output parameters of power modules belonging to both the load power module set and the optimal power module set are adjusted.

[0012] In some embodiments of the present disclosure, the system controller is a CAN controller, each power module is connected to the CAN controller through a corresponding CAN bus, and the adjustment instruction corresponding to the comparison result of the load power module set and the optimal power module set is converted into a corresponding CAN message and sent to the corresponding power module.

[0013] In some embodiments of the present disclosure, in the step (9), the power module sends real-time state information including online status, last heartbeat time, charging status, module number, module power, voltage, current, and switch state to the system controller through the CAN bus.

[0014] In some embodiments of the present disclosure, the modular power supply system further comprises a plurality of relays corresponding to each power module and controlled by the system controller through GPIO.

[0015] One or more technical solutions provided in the embodiments of the present application have at least any of the following technical effects or advantages: 1. Through the system energy efficiency optimization scheduling, the power module can be long-term operated in the high-efficiency interval, effectively reducing the invalid power loss and heat dissipation cost, and providing the profitability of the charging station in the actual application scenario.

[0016] 2. Through the life balance scheduling method, the power module with the shortest cumulative running time is dynamically selected, effectively avoiding the premature aging of part of the modules, significantly improving the average failure-free time of the system, and being beneficial to prolonging the life cycle of the equipment.

[0017] 3. The scheduling method based on the energy efficiency and lifespan of the collaborative system reduces both operating costs (including electricity costs) and long-term capital expenditures (including module replacement) and maintenance costs, thereby minimizing the total cost of the modular power system throughout its entire life cycle. This enables the system to select the optimal scheduling and operation strategy based on real-time load changes and module health status, exhibiting high flexibility and adaptability.

[0018] 4. The weighting factors in total cost calculation allow operators to flexibly adjust them according to actual conditions such as time-of-use pricing, seasonal changes, and maintenance plans, thereby achieving a dynamic balance between system energy efficiency and lifespan. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system principle of a modular power supply system in one embodiment of this application. Detailed Implementation

[0020] The programs involved or relied upon in the following embodiments are all conventional or simple programs in this technical field. Those skilled in the art can make conventional choices or adaptive adjustments according to specific application scenarios.

[0021] To better understand the technical solution of this application, the above technical solution will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] Existing modular power systems employ static or simple allocation strategies, leading to power modules operating in inefficient states and increasing system operating costs. Furthermore, uneven workload allocation strategies accelerate the aging and failure of some modules, resulting in reduced system reliability and increased maintenance costs. To address this, this paper discloses a scheduling method for modular power systems. By comprehensively considering system energy efficiency and lifespan, this method flexibly schedules each power module based on external conditions (such as electricity prices) or operational objectives (such as prioritizing energy conservation). Specifically, this scheduling method includes the following steps: (1) Establish energy efficiency curve models of power modules under different load rates and module health models including cumulative running time and cumulative start-stop times, and store the two models in the system controller.

[0023] In order to consider the energy efficiency and service life of each power module in the scheduling, in the embodiment, firstly, an energy efficiency curve model of the power module in the system is established, and the energy efficiency curve model is used to describe the electric energy conversion efficiency of the power module under different load rates, so as to reflect the energy efficiency of the power module through the electric energy conversion efficiency. Specifically, in the embodiment, the efficiency of the power module under a plurality of different integer load rates is obtained through experiments, and Table 1 is shown as follows. In order to facilitate the use of the energy efficiency curve model, and considering that the efficiency of the same power module under the corresponding load rate is basically stable in its life cycle, the energy efficiency curve model is stored in the SPI Flash in the system controller in the form of a lookup table in the embodiment, so as to avoid power loss.

[0024] .

[0025] In addition, a corresponding module health model is established for each power module, and the module health model at least includes the cumulative running time of the corresponding power module. In the embodiment, the module health model includes not only the cumulative running time of the power module, but also the cumulative start-stop times of the power module, so as to comprehensively reflect the service life state of the corresponding power module. Similarly, in order to avoid power loss of the module health model information, the module health model is also saved to the non-volatile memory of the system in the embodiment, so as to realize data persistence after power failure.

[0026] In addition, in the embodiment, in order to realize the scheduling of the modular power supply system, at the hardware level, the modular power supply system includes a system controller, and the system controller specifically adopts a CAN controller in the embodiment, and is connected with each power module through a CAN bus, so that the independent control of each power module can be realized based on the CAN bus. In addition, in the embodiment, considering that the power module is often controlled by a semiconductor switch, and there may be a small leakage current in the semiconductor device in the off state of the module, which may cause safety risks. In order to ensure the safety of the modular power supply system, the modular power supply system further includes a relay corresponding to each power module, and the CAN controller can control each relay through a GPIO, so as to provide electrical isolation by using the isolation characteristics of the relay, and cut off the main power link of the power module which does not participate in the load service, thereby disconnecting the corresponding power module from the high-voltage line, reducing the power consumption of the module in standby state, and being beneficial to reduce the temperature rise and power consumption of the system, and improve the service life and efficiency of the power module.

[0027] (2) Obtain the total power required by the charging gun P t ; and query the number of all available power modules in the system N and the cumulative running time of each power module in the module health model T i , whereini 1 N

[0028] First, the total power required by the charging gun is obtained before management scheduling P t Figure 1 In this embodiment, the modular power supply system mainly includes a business application layer, a policy decision layer, a service management layer, a hardware abstraction driving layer, and a hardware layer. The business application layer in this example includes charging control and a cloud platform, which serves as the demand source of the system. After the vehicle battery pipeline system interacts, the power demand for charging is determined, or the remote start-stop instruction is issued by the cloud platform, so as to issue the power target to the policy decision layer, that is, the total power required by the output charging gun P t If the system needs to output 45KW power.

[0029] While obtaining the total power required by the charging gun, the number of power modules currently available for load service and the corresponding cumulative running time of the system also need to be obtained, thereby serving the subsequent scheduling decision. In this embodiment, after the policy decision layer obtains the power target, it issues a query instruction to the service management layer to query the real-time state of the system, including the number of power modules available for load service and the cumulative running time of the corresponding power modules. The service management layer maintains a central data structure for describing the current accurate state of all hardware in the system, especially the power modules, which includes the real-time voltage, current, fault state, and cumulative running time of the power modules, thereby providing the policy decision layer with the data information required for scheduling decisions. In this embodiment, the number of all available power modules N and the cumulative running time of each power module in the module health model T i i 1 N

[0030] (3) The cumulative running time of each power module is normalized to obtain the normalized value of the cumulative running time of each power module t i i 1 N

[0031] In order to efficiently participate in the subsequent scheduling decision of the cumulative running time of each power module in the module health model, in this embodiment, the cumulative running time of each power module is normalized. Specifically, the maximum value of the cumulative running time of each power module is T max ​​​​​​​​​​​​then t i = T i / T max wherein i = 1 … N For example, in the present embodiment, referring to Figure 1 , the system sends a request to the service management layer through the policy decision layer to query the real-time state of the system, and the service management layer feeds back the real-time state of the system to the policy decision layer after reading the stored data. According to the query, it is found that in the present example, the system can participate in load service with a total of 5 available power modules M1-M5, and the corresponding running time in the module health model is 1000 hours for M1, 1200 hours for M2, 3000 hours for M3, 3200 hours for M4, and 5000 hours for M5. Thus, the maximum value of the cumulative running time of each power module is 5000 hours corresponding to M5, i.e. T max = 5000, and thus the normalized result of the cumulative running time of the power module M1 is 1000 / 5000, i.e. t 1 = 0.2, and similarly, the normalized results of the cumulative running time of the power modules M2-M5 are 0.24, 0.6, 0.64, and 1.0, respectively. t 2 t 5

[0032] (4) According to the total power required by the charging gun P t and the rated power of the power module P r , the optimization starting value m。

[0033] In the present embodiment, the total power required by the current system charging gun P t is 45 KW, and the rated power of each power module P r is 30 KW; therefore, a single power module cannot meet the total power required by the charging gun, and thus the scheduling scheme involving only a single power module can be omitted. Similarly, to improve the efficiency of scheduling decisions, in the present embodiment, the optimization starting value of the scheduling scheme is calculated as m ; wherein, m is the minimum value satisfying m × P r > P t and m ≤​​N， Its implied physical meaning is that the system needs at least m The participation of multiple power modules may meet the total power requirements of the current system. This can shield fewer than [number missing] modules. m A scheduling scheme involving multiple power modules. For example, in this embodiment, the total power required by the charging gun... P t It is 45KW, and the rated power of each power module is... P r For a power of 30KW, the requirement can be met when m=2. m × P r > P t Therefore, in this example, the participation of two power modules is taken as the starting point for optimizing the scheduling scheme. By traversing various scheduling schemes under the condition of m=2…N, the optimal scheme that takes into account both system energy efficiency and module lifespan is found.

[0034] (5) Calculate the load factor Z = P t / ( m × P r After multiplying by 100%, the power module's load factor is queried based on the energy efficiency curve model. Z The corresponding efficiency H Corresponding computational efficiency cost Ce =1-( H / H max ),in H max Efficiency in the energy efficiency curve model H The maximum value; select after sorting. m The power module with the shortest cumulative running time was selected, and its corresponding normalized value was obtained. t i , with that m The average value of the normalized values ​​of each power module is used as the lifespan cost. Cl。

[0035] Determine the total power required by the charging gun P t and the rated power of each power module. P r Next, the load of a single power module is calculated. In this embodiment, the initial value of m is 2, so two power modules are started at this time, and the total power required by the charging gun is... P t Rated power of a single power module: 45KW P r The value is 30KW, which corresponds to the load rate of this power module. Z = P t / (m × P r ) × 100% = 45 / (2 × 30) × 100% = 75%. Then, based on the energy efficiency curve model, we can look up the efficiency of the power module at a load rate of 75%. H However, the amount of data in the energy efficiency curve model is limited, making it impossible to accurately record the efficiency corresponding to the load rate at any given point. Therefore, when it is not possible to directly obtain the efficiency corresponding to the current load rate from the data in the energy efficiency curve model by looking up a table, a linear interpolation method is used to query the power module at any load rate. Z The corresponding efficiency H Specifically, in this embodiment, referring to Table 1, there is no direct efficiency value corresponding to a 75% load rate. The closest values ​​are 70% and 80% load rates. The efficiency corresponding to a 70% load rate is 94.5%, and the efficiency corresponding to an 80% load rate is 94.0%. Therefore, according to the conventional linear interpolation formula, the power module efficiency corresponding to a 75% load rate is 94.25%. Furthermore, the efficiency cost is calculated. Ce ,in Ce =1-( H / H max ), H max Efficiency in the energy efficiency curve model H The maximum value. In this embodiment, it can be found by looking up Table 1 that... H max The efficiency rate is 95.8%, therefore the efficiency cost is... Ce =1-(94.25% / 95.8%)≈0.0162.

[0036] Since the discussion here focuses on activating only two power modules, m In the case of 2, considering the lifespan of the power modules, the two power modules with the shortest cumulative operating time are selected to participate in the load service under this scheme. Specifically, based on the cumulative operating time data of the five available power modules recorded in step (3), after sorting, it can be seen that power modules M1 and M2 are the two available power modules with the shortest cumulative operating time. Therefore, power modules M1 and M2 are selected to participate in the load service under this scheme, and then the average value of the normalized values ​​of the two power modules is calculated as the lifespan cost. Cl Specifically, as can be seen from step (3), the normalized result of the cumulative running time of power module M1 is... t 1 The value is 0.2, which is the normalized result of the cumulative running time of power module M2. t 2 The lifespan cost is 0.24. Cl =avg(0.2, 0.24) = 0.22.

[0037] (6) Calculate the corresponding startup m Total cost per power module TC m = w 1 × Ce + w 2 × Cl .

[0038] Based on step (5) m =2 efficiency cost Ce =0.0162, lifespan cost Cl =0.22, calculate startup m Total cost when using 2 power modules TC m = w 1 × Ce + w 2 × Cl ;,in w 1 As an energy efficiency weight, w 2 The weight is based on lifespan, and thus, by configuring weighting factors, dispatch bias can be adjusted. For example, during peak electricity price periods, an energy efficiency weight can be configured. w 1 Greater than lifespan weight w 2 Prioritize energy saving; additionally, configure lifespan weighting. w 2 Greater than energy efficiency weight w 1 In this case, priority is given to ensuring the lifespan of system modules. In this embodiment, taking the scheduling principle that considers system energy efficiency and system lifespan as equally important as an option, the energy efficiency weight is assigned as a weighting factor. w 1 and lifespan weight w 2 All of these are stored as system parameters, and energy efficiency weights are set and adjusted. w 1 and lifespan weight w 2 Both are 0.5. At this point, start. m Total cost when using 2 power modules TC 2 : TC 2 = w 1 × Ce + w 2 ×Cl= 0.5×0.0162+0.5×0.22=0.1181.

[0039] (7) Order m = m + 1 Repeat steps (5) and (6) until... m = N Compare the total costs when starting different numbers of power modules to obtain the startup cost. j Minimum total cost corresponding to each power module TC min To obtain the minimum total cost TC min The corresponding includes j The optimal set of power modules for each power module; To obtain the total cost of the system when starting the remaining number of power modules. TC m In this example, the calculation yields... m =2 represents the total cost when activating two power modules. TC 2 Afterwards, m = m + 1 Repeat steps (5) and (6) until... m = N This involves iterating through m from 2 to N, thereby obtaining the total cost corresponding to starting different numbers of power modules. TC m By comparing the total costs of each feasible solution, an appropriate initiation can be made. j The total cost is minimized when there are only one power module, and the minimum total cost is... TC min Therefore, the set of power modules corresponding to the minimum total cost is taken as the optimal scheduling scheme, that is, the global optimal solution is found. In this example, the above cyclic traversal process is implemented through the policy decision layer.

[0040] For example, in this embodiment, the result is obtained after the first loop. m= 2. The total cost when starting two power modules. TC 2 Before the second loop, increment m by 1, that is, during the second loop... m= 3 means starting three power modules; at this time, the load rate corresponding to a single power module is... Z = P t / ( m × P r )×100%=45 / (3×30)×100%=50%; The power module's load rate can be obtained by querying the energy efficiency curve model. Z Efficiency corresponding to 50% H It is 95.2%; thereforem= Efficiency cost at 3 o'clock Ce =1-(95.2% / 95.8%)≈0.0063; For lifespan cost, select the three modules M1, M2 and M3 with the shortest cumulative operating time. From step (3), it can be seen that the normalized result of the cumulative operating time of power module M1 is... t 1 The value is 0.2, which is the normalized result of the cumulative running time of power module M2. t 2 The value is 0.24, which is the normalized result of the cumulative running time of power module M3. t 3 The lifespan cost is 0.6. Cl =avg(0.2, 0.24, 0.6) ≈ 0.347. Maintain energy efficiency weighting. w 1 and lifespan weight w 2 Both remain unchanged at 0.5. m =3 represents the total cost when activating three power modules. TC 3 for: TC 3 = w 1 × Ce + w 2 × Cl= 0.5×0.0063+0.5×0.347=0.1767.

[0041] Similarly, during the third loop, m is incremented by 1, so m=3, until m=N. In this example, there are a total of 5 usable power modules, i.e., N=5. After the loop ends, a total cost of 4 is obtained. TC 2 ~ TC 5 Thus, the minimum total cost can be obtained through comparison. TC min The scheduling is based on the set of power modules corresponding to the minimum total cost, ensuring that the power modules in this set output the corresponding power while shutting down the remaining modules. For example, [the following is an example of scheduling:] TC 2 and TC 3 After comparison, it can be found that TC 2 < TC 3 This means TC 2 The corresponding scheduling scheme will be superior to TC 3The corresponding scheduling scheme, although starting three power modules has higher individual efficiency, but due to the need to start a higher degree of labor module, resulting in the life cost of the system is significantly increased, so in the example set under the weight, will consider starting two modules (M1 and M2) is a better choice than starting three power modules.

[0042] (8) After obtaining the current load power module set that has served the load, the load power module set is compared with the optimal power module set, and the state of the corresponding power module is adjusted according to the comparison result to only make the power module in the optimal power module set participate in the load service at the corresponding power.

[0043] After obtaining the optimal power module combination scheme considering system energy efficiency and life and corresponding minimum total cost based on the multi-objective (including energy efficiency and life) optimization algorithm in the strategy decision layer, the combination scheduling scheme is issued to the service management layer for execution operation. Specifically, after obtaining the scheduling decision result, the service management layer converts it into specific module corresponding detailed standard control instructions (such as setting the voltage of power module M3 to 500V), and then the hardware abstraction driver layer converts the control quality into hardware recognizable signals in the hardware layer after obtaining the standard control instructions, and packs them into CAN messages of the corresponding protocol according to the specific model of the power module, and sends them to the corresponding hardware module through the CAN bus. In addition, in this embodiment, the hardware abstraction driver layer is also responsible for parsing the bottom layer interrupts and response data output by the hardware layer into a standard format and reporting them to the service management layer for identification and storage.

[0044] However, after obtaining the optimal power module set, considering that the current system may have been serving the load, this power module set optimization may be an adjustment operation due to changes in load demand during load service. To avoid interruption of the load service process, in this embodiment, the current load power module set that has served the load is compared with the optimal power module set. Specifically, a list of power modules that need to be started, closed or adjusted parameters is established compared with the optimal power module set, thereby converting the complex system change problem into a data comparison and plan list generation problem, thereby realizing smooth and efficient system update and only processing the functional modules that need to be changed. Among them, the service management layer issues a series of specific and atomic control instructions to the hardware abstraction driver layer based on the list after comparison, so that the adjustment scheduling instruction is executed asynchronously by the hardware abstraction driver layer, that is, the control instruction enters the corresponding queue or task pool in the hardware abstraction driver layer for execution, avoiding the interruption of the scheduling process due to the waiting of the service management layer for the instruction response, and ensuring the reliability of the scheduling.

[0045] According to the list of power modules that need to be started, closed or adjusted obtained after comparison, the power modules that belong to the load power module set and do not belong to the optimal power module set are adjusted, the power modules that do not belong to the load power module set and belong to the optimal power module set are closed, and the output parameters of the power modules that belong to both the load power module set and the optimal power module set are adjusted, thereby avoiding full adjustment of all power modules in the system, and helping to improve the scheduling efficiency.

[0046] (9) The system controller updates the module health model and the real-time state of each power module.

[0047] After the modules in the hardware layer execute the corresponding control instructions, the real-time state information of the modules, such as online status, last heartbeat time, charging status, module number, module power, voltage, current, switch state and the like, is reported to the service management layer through the CAN bus, and the service management layer updates and maintains the central data structure accordingly, so as to provide the latest and reliable data basis for the decision-making of the next scheduling period.

[0048] (10) Steps (2)-(9) are repeated when the periodicity and / or load demand changes.

[0049] Thus, a complete scheduling closed-loop operation of sensing, decision-making, execution and feedback is completed, the system enters a stable running state, and waits for the triggering of the next scheduling event. In the embodiment, considering that the demand of the load may change in the process of providing load service for the load, in order to respond to the change of the load, steps (2)-(9) are repeated when the periodicity and / or load demand changes.

[0050] Although some preferred embodiments of the present application have been described, those skilled in the art can make further changes and modifications to these embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including all the preferred embodiments and all the changes and modifications falling within the scope of the present application.

[0051] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A scheduling method for a modular power system, characterized in that, Includes the following steps: (1) Establish energy efficiency curve models of power modules under different load rates and module health models including cumulative running time and cumulative start-stop times, and store the two models in the system controller. (2) Obtain the total power required for the charging gun P t And query the number of all currently available power modules in the system. N and the cumulative running time of each power module in the module health model T i ,in i = 1 … N ; (3) Normalize the cumulative running time of each power module to obtain the normalized value of the cumulative running time of each power module. t i ,in i = 1 … N ; (4) Based on the total power required by the charging gun P t and the rated power of the power module P r Calculate the initial value for optimization. m ,in, m To meet m × P r > P t The minimum value and m ≤ N ; (5) Calculate the load factor ζ=P t / ( m × P r After multiplying by 100%, the power module corresponding to the energy efficiency curve model is queried based on the load rate. ζ The corresponding efficiency η Corresponding computational efficiency cost Ce =1-( η / η max ),in η max Efficiency in the energy efficiency curve model η The maximum value; select after sorting. m The power module with the shortest cumulative running time was selected, and its corresponding normalized value was obtained. t i , with that m The average value of the normalized values ​​of each power module is used as the lifespan cost. Cl ; (6) Calculate the corresponding startup m Total cost when using a single power module TC m = w 1 × Ce + w 2 × Cl ,in w 1 As an energy efficiency weight, w 2 Lifetime weighting; (7) Order m = m + 1 Repeat steps (5) and (6) until... m=N Compare the total costs when starting different numbers of power modules to obtain the startup cost. j Minimum total cost corresponding to each power module TC min To obtain the minimum total cost TC min The corresponding includes j The optimal set of power modules for each power module; (8) After obtaining the set of load power modules currently serving the load, compare the set of load power modules with the set of optimal power modules, and adjust the state of the corresponding power modules according to the comparison result so that only the power modules in the set of optimal power modules participate in the load service with the corresponding power. (9) The system controller updates the module health model and the real-time status of each power module; (10) Repeat steps (2) to (9) when the periodicity and / or load demand changes.

2. The scheduling method according to claim 1, characterized in that, In step (1), the energy efficiency curve model includes the efficiency of power modules corresponding to a finite number of load rates, and is stored in the non-volatile memory of the system controller in the form of a lookup table.

3. The scheduling method according to claim 2, characterized in that, In step (5), the lookup table corresponding to the energy efficiency curve model is used to query the power module at any load rate using a lookup table method and a linear interpolation method. ζ The corresponding efficiency η .

4. The scheduling method according to claim 1, characterized in that, In step (3), the maximum cumulative running time corresponding to each power module is set to... T max ,but t i = T i / T max ,in i = 1 … N .

5. The scheduling method according to claim 1, characterized in that, In step (8), the load power module set is compared with the optimal power module set by differential comparison, and the output parameters of the power modules that belong to the load power module set but not to the optimal power module set are adjusted according to the comparison results: the power modules that belong to the load power module set but not to the optimal power module set are turned off, the power modules that do not belong to the load power module set but belong to the optimal power module set are turned on, and the power modules that belong to both the load power module set and the optimal power module set are adjusted by differential comparison.

6. The scheduling method according to claim 1 or 5, characterized in that, The system controller is a CAN controller, and each power module is connected to the CAN controller via a corresponding CAN bus communication connection. The adjustment command corresponding to the comparison result between the load power module set and the optimal power module set is converted into a corresponding CAN message and sent to the corresponding power module.

7. The scheduling method according to claim 6, characterized in that, In step (9), the power module sends real-time status information to the system controller via the CAN bus, including online status, last heartbeat time, charging status, number of modules, module power, voltage, current, and switch status.

8. The scheduling method according to claim 1 or 6, characterized in that, The modular power system also includes several relays that correspond one-to-one with each power module and are controlled by the system controller via GPIO.

Images