Charging load management method oriented to pulse charging method

Abstract

A charging load management method oriented to a pulse charging method, belonging to the field of lithium-ion batteries. The charging load management method comprises: a staggered sequencing method for different charging pulses, a pulse current waveform adjustment method, a grouping and staggered sequencing method for a series of pulses having equal periods, a grouping and staggered sequencing method for a series of pulses having unequal periods, and an overall coordination method for charging pulses. A method for grouping and temporally sequencing different pulses can reduce the total charging current of all loads and the fluctuation amplitude thereof without reducing an individual charging load. When applied to a battery pack consisting of a plurality of cells, the method of the present invention can suppress the overall charging power fluctuation of the battery pack; and when applied to a plurality of electric vehicles, the method of the present invention can suppress the charging power fluctuation of an electric vehicle charging station, and assist the vehicle charging station in participating in power grid demand-side response, thereby facilitating an enhancement in the stability of a power grid.

Description

A charging load management method for pulse charging method TECHNICAL FIELD

[0001] The application belongs to the technical field of lithium ion batteries, and particularly relates to a charging load management method for a pulse charging method. BACKGROUND

[0002] Lithium ion batteries have been widely applied to various electronic products and electric vehicles. In recent years, with the increasing market penetration rate of electric vehicles, the charging demand of lithium ion batteries is also increasing. The traditional charging methods of lithium ion batteries include constant current charging method, constant voltage charging method and constant voltage-constant current charging method. The constant current charging method uses a constant current for charging, and the charging speed is fast but overcharging may occur. The constant voltage charging method can avoid overcharging, but the charging speed is slow. The constant voltage-constant current charging method combines the advantages of the above two methods, and improves the charging speed under the premise of avoiding overcharging. In addition to the traditional charging methods, the pulse charging method is a fast and efficient charging method, which uses an intermittent way to charge the battery. In one cycle T, the battery is first charged with a large current for a charging time t on , and then stopped for a stop time t off , as shown in FIG. 1. Compared with the traditional charging method, this method can reduce the polarization effect of the battery, prolong the service life of the battery, improve the charging efficiency, and is suitable for high-power charging application scenarios.

[0003] With the rapid increase of the number of electric vehicles, the charging power of each electric vehicle is increasing, and the charging demand of electric vehicles is also rapidly growing. The pulse charging method causes fluctuations in charging current, charging power and even grid voltage due to its intermittency, whether pulse current or pulse voltage is used, which puts pressure on the stability of the power grid. Therefore, when multiple electric vehicles are charged with the pulse charging method, it is necessary to coordinate these charging loads to reduce the impact of their volatility on the power grid. SUMMARY

[0004] To solve the problems in the prior art, the application provides a charging load management method for a pulse charging method.

[0005] The technical scheme adopted by the application to solve the technical problems is as follows:

[0006] The application provides a charging load management method for a pulse charging method, which includes a different charging pulse staggered sequencing method. When the charging pulse is a pulse current, the pulse current is at a low level in the t off period of each cycle T, and the current value is zero. Assuming that the low level interval of pulse current 1 in one cycle T is t off,1 , the high level intervals of pulse currents 2, 3, …, n in one cycle T are ton,2 , t on,3 , …, t on,n ; when formula (1) is established, the high level of these pulse currents is arranged to appear in turn within the low level interval t off,1 , so as to realize that the n pulse currents do not overlap in time and reduce the total current peak value of the n pulse currents.

[0007] Wherein n∈[1,+∞]#(1).

[0008] Further, when the charging pulse is a pulse voltage, the pulse voltage is low within t off in each cycle T, and the voltage value is zero; assuming that the low level interval of the pulse voltage 1 in a cycle T is t off,1 , the high level intervals of the pulse voltages 2, 3, …, n in a cycle T are t on,2 , t on,3 , …, t on,n respectively; when formula (1) is established, the high level of these pulse voltages is arranged to appear in turn within the low level interval t off,1 , so as to realize that the n pulse voltages do not overlap in time and reduce the total voltage peak value of the n pulse voltages.

[0009] Further, the application provides a charging load management method for the pulse charging method, which also comprises a pulse current waveform adjustment method; if the rest single charging pulses do not satisfy formula (1), the charging pulse cannot be staggered and sorted, so the waveform of the pulse current is changed by adjusting the duty ratio and amplitude of the pulse current, while keeping the average charging power in a cycle unchanged; before adjustment, the pulse current amplitude is I m , the cycle is T, the high level interval in a cycle T is t on , the low level interval is t off , and the duty ratio is The calculation formula of the average charging power of the pulse current is:

[0010] Wherein u is the charging voltage; the average charging power in a cycle T is changed by changing the charging voltage, the charging current or the duty ratio; after adjustment, the cycle is unchanged, the high level interval in a cycle T is t' on , the low level interval is t' off , and the duty ratio is The charging voltage and the charging current are adjusted to U' and I' respectively; to keep the average charging power unchanged, only need to ensure that formula (3) is established:

[0011] By adjusting the waveforms of the remaining n pulse currents so that at least one of them satisfies condition (1), the high levels of different charging pulses can be staggered by using the different charging pulse staggering sorting method.

[0012] Furthermore, the charging load management method for pulse charging provided by this invention also includes a method for grouping and staggering a series of pulses with equal periods, if there are N c N charging loads are charged simultaneously, each using a pulse charging method. All charging pulse currents have a period equal to T, but their duty cycles and amplitudes are not all equal. c The peak instantaneous charging current of each charging load i max The sum of all pulse currents; if the charging voltage of the charging load i is u i Then the instantaneous peak charging power p of all charging loads max It is the sum of the instantaneous power of all pulses; assuming the high-level interval t of pulse a within one period T. on,a The low-level interval t of another pulse b is less than or equal to the low level interval of the other pulse b. off,b , a,b∈[1,N c And t on,a ,t off,b If the signal is ∈[0,T], then pulse b can be scheduled to conduct during the off-time of pulse a. By staggering the conduction of pulses a and b in time, the peak value of the total current is reduced, thus lowering the peak value of the total charging power. At this time, the instantaneous total charging current is i. a and i b The larger one rather than i a with i b sum.

[0013] Furthermore, the specific operation steps of the method for grouping and staggering a series of pulses with equal periods are as follows:

[0014] Assume the length of the high-level interval of the pulse used by the i-th load within one cycle is t. on,i The length of the low-level interval is t off,i ; use x i Indicates whether it is in the low-level range t of pulse i. off,i Internal charging of other loads, x i =1 means that x i =0 indicates no; when x i When y = 1, pulse i is considered a "container" of pulses; using y ij Indicates whether or not in t off,i The interval charges the j-th load, y ij =1 indicates that y ij =0 indicates no; when y ijWhen = 1, pulse j is considered an "item" to be loaded into other "containers"; all charging pulses are divided into S groups, S∈[1,N]. c In each group, the pulses appear at high levels sequentially without overlapping. To minimize the peak value of the total charging current of all charging loads, the value of S should be minimized, i.e., the number of "containers" should be minimized. The mathematical expression for this objective is:

[0015] The mathematical representation of its constraints is as follows:

[0016] 1) For any i∈[1,N] c ],have:

[0017] 2) For any j∈[1,N] c ],have:

[0018] Inequality constraint 1) guarantees that within any period, when pulse i acts as a "container", the sum of the high-level intervals of all other pulses loaded into that "container" does not exceed the capacity of the "container", that is, the high levels of each pulse do not overlap in time; when pulse j is arranged in t off,i Conduction, y ij =1, at this time t on,j The left-hand side of the inequality is included; if pulse j is also a "container", then it is not grouped with pulse i. In this case, to avoid including the conduction period t of pulse j... on,j Also included in the calculation is the factor (1-x) on the left-hand side of the inequality. j At this time, x j =1, 1-x j =0,t on,j It is not included in the left-hand side of the inequality; the right-hand side of the inequality includes the factor x. i x , if and only if pulse i is a "container" i =1, its capacity t off,i Only if it is included, otherwise the right side of the inequality is zero, indicating that pulse i is a pulse to be loaded;

[0019] Equality constraint 2) guarantees that pulse j exists and is unique within any period; for any j∈[1,N] c The left side of the equation iterates through i from 1 to N. c All y ij Then sum them; when pulse j is a "container", the sum is 0, meaning pulse j will not be loaded into other "containers" as an "item"; otherwise, the sum is 1, indicating that pulse j will be loaded into another "container" as an "item". Specifically, if pulse j is not grouped with any other pulse, pulse j is considered a "container", i.e., xj = 1;

[0020] Solve equations (6) to (8) to obtain all solutions of x i and y ij , i.e. the grouping result of the pulses; the value of x i indicates which pulses are the "containers" of other pulses, and the sum of all x i is the total number of groups S; all pulses j with the same i value and y ij = 1 are grouped into a group, where j ∈ [1, N c ]; for the pulses in the same group, arrange them to appear in time sequence high level.

[0021] Further, the application provides a charging load management method for the pulse charging method, which further comprises a method for grouping and staggered sequencing a series of pulses with unequal periods, and simultaneously charging N c charging loads by using the pulse charging method, wherein the charging pulse current has unequal periods, and the duty cycle and amplitude are not all equal; if the frequency of any one pulse is not an integer multiple of the frequency of another pulse, then these pulses do not have the possibility of grouping and staggering the high level; if the frequency of at least one pulse is an integer multiple of the frequency of another pulse, and the high level interval of the pulse with lower frequency in one period is not longer than the low level interval of the other pulse with higher frequency, then these pulses can be grouped and the high level interval of each pulse in the group can be staggered for charging, so as to reduce the instantaneous total current and instantaneous total power of all charging loads, while not reducing the charging current and charging power of any charging load.

[0022] Further, the specific operation steps of the method for grouping and staggered sequencing a series of pulses with unequal periods are as follows:

[0023] First, group all charging pulses and respectively group and sequence them; if there are N c charging loads simultaneously charged by using the pulse charging method, the length of the high level interval of the i-th pulse in one period is denoted as t on,i , and the length of the low level interval is denoted as t off,i ; x i is used to represent whether to charge other loads in the low level interval t off,i of the pulse i, x i = 1 represents yes, and x i = 0 represents no; when x i = 1, the low level interval of the pulse i in one period is regarded as the "container" of the high level of another pulse, so that other charging loads can be charged in this interval to reduce the total instantaneous current peak value; y ij is used to represent whether to stagger the high level of the pulse i in t off,i ;interval of the high level of the jth pulse, y ij = 1 means yes, y ij = 0 means no; when y ij = 1, the high level interval of the pulse j is regarded as an "article" to be loaded into other "containers";

[0024] Assuming that all pulse periods are integer multiples of the highest frequency pulse period, consider a time period T, which is the least common multiple of all pulse periods, the T period covers N i low level intervals of the pulse i, which is equivalent to there being N i "containers" in the T period, in which the high level intervals of other qualified pulses are arranged; use variable k to number these low level intervals, and introduce a variable z ijk to represent whether the high level interval of the pulse j is arranged in the kth low level interval of the pulse i, z ijk = 1 means yes, z ijk = 0 means no;

[0025] Divide all charging pulses into S groups, S ∈ [1, N c ], and the pulses in each group appear in high level in turn and do not overlap in time; to make the total charging current peak value of all charging loads as small as possible, the value of S should be as small as possible, that is, the number of "containers" should be as small as possible, and the mathematical expression of this target is:

[0026] The mathematical representation of the constraint conditions is as follows:

[0027] 1) For any k ∈ [1, N i ] and any i ∈ [1, N c ], there is:

[0028] 2) For any k ∈ [1, N i ] and any i, j ∈ [1, N c ] and satisfying N i = N ij N j , there is:

[0029] Where N ij is a positive integer;

[0030] 3) For any j ∈ [1, N c ], there is:

[0031] Inequality constraint condition 1) limits that in a period, the sum of the lengths of the high level intervals of all other pulses in the kth low level interval of the ith pulse does not exceed toff,i ; the factor (1-x j ) on the left of the inequality avoids taking into account the pulse j as a "container" when x j =1, so that 1-x j =0; the factor x i on the right of the inequality, when and only when the pulse i is a "container", i.e. when x i =1, takes into account the width t off,i of the pulse i, otherwise the right of the inequality is zero, indicating that the pulse i is a pulse to be loaded;

[0032] The equality constraint 2) ensures that the period of the pulse j is equal to T j ; this constraint traverses all the low intervals of the pulse i and requires that the sum of z ij in any interval of N ijk consecutive low intervals is equal to 1, so that in any interval of N ij consecutive low intervals of the pulse i there is and only one interval of high level of the pulse j, to ensure that the period of the pulse j is N ij times the period of the pulse i;

[0033] The equality constraint 3) limits that, in a period, if any pulse is not a "container", it must and only with one other pulse be in a group; if a pulse j is not in a group with any other pulse, the pulse j is considered a "container", i.e. x j =1; for any j∈[1,N c ], the left of the equality traverses all the y c when i goes from 1 to N ij and sums them; when the pulse j is a "container", x j =1, 1-x j =0, the sum is zero, meaning that the pulse j will not be loaded as an "item" in other "containers"; otherwise, the sum is 1, indicating that the pulse j will be loaded as an "item" in some "container";

[0034] By solving equations (9) to (12) the values of all x i , y ij and z ijk are obtained, i.e. the grouping of the pulses; x i indicates which pulses are "containers" for other pulses, the sum of x i is the total number of groups S; y ij indicates the grouping arrangement, all the pulses j with the same i and y ij =1 belong to a group, i,j∈[1,N c ]; z ijk indicates the specific arrangement of the intervals: z ijk=1 means arranging the high level interval of pulse j in the kth low level interval of pulse i; z ijk =0 means not arranging it in this way; for all high levels of pulses in the same "container", arrange them to appear in time sequence while ensuring their consistent positions in other "containers"; arrange the first high level interval of pulse j in the kth low level interval of pulse i, which is realized by moving pulse j in time axis, i.e. changing the initial phase of pulse j.

[0035] Further, the method of grouping and sequentially arranging all charging pulses is as follows:

[0036] 1) initialize positive integer variable n, let n=1;

[0037] 2) find the pulse with the highest frequency among all ungrouped charging pulses, and mark it as I n (f n );

[0038] 3) group all pulses with frequency f n integer multiple x into a group, mark it as I nx (f nx ), x is a positive integer;

[0039] 4) group and sequentially arrange all pulses in the group by using the method of grouping and staggered arrangement of a series of pulses with unequal periods;

[0040] 5) if there are still ungrouped pulses, let n=n+1, and repeat the above process until all pulses are grouped and arranged.

[0041] Further, the charging load management method for the pulse charging method provided by the present application further comprises a charging pulse conduction timing overall coordination method, which first determines which grouping and arrangement method to use according to the frequency of the pulse; if all charging pulses have the same frequency, the method of grouping and staggered arrangement of pulses with equal periods is used to group and arrange all charging pulses; if not all charging pulses have the same frequency, the method of grouping and staggered arrangement of a series of pulses with unequal periods is used to group and arrange all charging pulses.

[0042] Further, after determining the grouping and arrangement, the total charging power P Σ of all loads is calculated; if there is an upper limit P max for the total charging power of all loads, it is determined whether to charge all loads simultaneously according to the size relationship between P max and P Σ .

[0043] If P Σ ≥ Pmax Then, all vehicles are grouped according to the charging plan, and further divided into non-delayable and delayable groups based on whether their charging plans can be delayed. First, the total charging power P of all vehicles in the non-delayable group is calculated. Σ If P Σ >P max Therefore, it is necessary to limit the total charging power of all loads within the group; while keeping the charging voltage constant, this can be achieved by reducing the charging current I of all vehicles. i To reduce the total charging power P Σ The reduced current The calculation formula is:

[0044] Without changing the amplitude of the charging current of each vehicle, by reducing the duty cycle d of each pulse. i To reduce the total charging power P Σ Reduced duty cycle The calculation formula is:

[0045] If the method of reducing the duty cycle is used to reduce the total charging power P, Σ After calculating the duty cycle Then, the order of the car charging pulses is rearranged using either the method of grouping and staggering a series of pulses with equal periods or the method of grouping and staggering a series of pulses with unequal periods, thereby performing orderly charging.

[0046] If P Σ <P max The charging station selects a portion of the vehicles within the deferred charging group for charging; for the deferred charging group, charging is based on the vehicle charging priority c. i Sort, c i ∈[0,1], the larger the value, the higher the priority; if it is not possible to charge all cars in the group that can be postponed, then charge the m cars with higher priority in the group first, where m should satisfy equation (15):

[0047] Among them, P i This represents the charging power of the i-th load; for lower priority vehicles, in P... max Stop charging before the charging power of another vehicle is increased or reduced.

[0048] The beneficial effects of this invention are:

[0049] The pulse charging method can reduce the polarization effect of the lithium ion battery, prolong the service life of the battery, and is suitable for high-power charging application scenarios. For this pulse charging method, the application provides a method for grouping and time sequencing different pulses, which can reduce the total charging current of all loads without reducing the single charging load, and reduce the fluctuation amplitude. Secondly, the application also provides a method for arranging the charging sequence of all loads according to the power upper limit. If the method of the application is applied to a battery pack composed of multiple battery cells, the total charging power fluctuation of the battery pack can be suppressed; if the method of the application is applied to multiple electric vehicles, the charging power fluctuation of the electric vehicle charging station can be suppressed, the electric vehicle charging station can participate in the demand side response of the power grid, and the stability of the power grid can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0050] Fig. 1 is a schematic diagram of the current waveform of the pulse charging method.

[0051] Fig. 2 is a schematic diagram of n charging pulses staggered sequencing.

[0052] Fig. 3 is a schematic diagram of charging pulse waveform adjustment.

[0053] Fig. 4 is a schematic diagram of a group of pulse charging currents with the same period.

[0054] Fig. 5 is a schematic diagram of two pulses staggered on.

[0055] Fig. 6 is a schematic diagram of a series of period equal pulse grouping analogy packing problem.

[0056] Fig. 7 is a schematic diagram of a group of pulse currents with different periods.

[0057] Fig. 8 is a schematic diagram of a series of period unequal pulse grouping analogy packing problem.

[0058] Fig. 9 is a schematic diagram of the low level interval number of the pulse.

[0059] Fig. 10 is a schematic diagram of pulse grouping and sequencing.

[0060] Fig. 11 is a schematic diagram of the grouping and sequencing method of the pulse with unequal period.

[0061] Fig. 12 is a schematic diagram of the overall coordination method of the charging pulse on-off timing.

[0062] Fig. 13 is a waveform diagram of a series of equal period pulses before and after grouping and sequencing in embodiment 1.

[0063] Fig. 14 is a total current waveform diagram of a series of equal period pulses before and after grouping and sequencing in embodiment 1.

[0064] Figure 15 is a waveform diagram of a series of equal period unequal pulse grouping before and after sorting in Example 2.

[0065] Figure 16 is a total current waveform diagram of a series of unequal period pulse grouping before and after sorting in Example 2. DETAILED DESCRIPTION

[0066] The application will be further described in detail below with reference to the accompanying drawings.

[0067] The application provides a charging load management method for pulse charging method, which specifically comprises the following steps:

[0068] 1. Different charging pulse staggered sorting method

[0069] For a plurality of loads simultaneously charged by the pulse charging method, the application proposes a method of grouping and sorting the charging loads to reduce the total instantaneous power and power fluctuation of charging.

[0070] The charging pulse to which the application is directed can be pulse current or pulse voltage. For ease of introduction, only pulse current will be described below. The pulse current shown in Figure 1 has a low level in part of the period T, i.e. t off period, and the current value is zero. For this type of charging current, the application proposes a different charging pulse staggered sorting method, which arranges the high level of other charging pulse currents to appear in this period to reduce the total instantaneous current of all charging loads. To be general, it is assumed that the low level interval of pulse current 1 in a period T is t off,1 , and there are pulse currents 2, 3, …, n, whose high level intervals in a period T are t on,2 , t on,3 , …, t on,n When the condition formula (1) is established, the high levels of these pulse currents can be arranged to appear in the low level interval t off,1 in turn, as shown in Figure 2.

[0071] where n∈[1, +∞] (1)

[0072] The different charging pulse staggered sorting method can realize that n pulse currents do not overlap in time, thereby reducing the total current peak value of the n pulse currents. At this time, the maximum instantaneous value of the total current does not exceed the maximum value of the n pulse currents.

[0073] 2. Pulse current waveform adjustment method

[0074] For a single load, the application proposes a method of changing the duty cycle of the charging pulse without changing the average power of the charging, to cooperate with the subsequent grouping and sorting with other charging loads.

[0075] If none of the remaining single charging pulses satisfy equation (1), the charging pulses cannot be staggered. The present application proposes a pulse current waveform adjustment method, which adjusts the duty cycle and amplitude of the pulse current to change its waveform while keeping its average charging power unchanged in a period. Before adjustment, the pulse current waveform is shown in Figure 3, with an amplitude of I m , a period of T, a high level interval of t on , a low level interval of t off , and a duty cycle of The average charging power of the pulse current can be calculated by equation (2):

[0076] where u is the charging voltage. As can be seen, changing the charging voltage, charging current or duty cycle can change the average charging power in a period T. If the charging voltage and charging current are adjusted while the duty cycle of the pulse current is changed, the average charging power can also be unchanged, as shown in one case of the adjusted waveform in Figure 3. After adjustment, the period is unchanged, the high level interval in a period T is t' on , the low level interval is t' off , the duty cycle is The charging voltage and charging current are adjusted to U' and I' respectively. To keep the average charging power unchanged before and after adjustment, it is only necessary to ensure that equation (3) is established:

[0077] By adjusting the waveforms of the remaining n pulse currents so that at least one of them satisfies condition equation (1), the above step 1 proposes a different charging pulse staggered sequencing method to stagger the high levels of different charging pulses.

[0078] 3. A method of grouping and staggering a series of pulse currents with equal periods (grouping and staggering method 1)

[0079] For the case of charging pulse currents with the same period for all charging loads, the present application proposes a method of grouping and staggering all charging loads, thereby reducing the total charging current peak value and total power peak value of all loads.

[0080] If there are N c charging loads charging at the same time, each load is charged by the pulse charging method shown in Figure 1, and the periods of all charging pulse currents are equal to T, but the duty cycles and amplitudes are not all equal, as shown in Figure 4. If all loads are charged by the timing shown in Figure 4, the instantaneous charging current peak value i c of the N max charging loads is the sum of all pulse currents, i.e.:

[0081] If the charging voltage of the charging load i is u i , then the instantaneous charging power peak value p max of all charging loads is the sum of all instantaneous power, i.e.:

[0082] If the high level interval of at least one pulse current is less than or equal to the low level interval of at least another pulse current in a cycle T, the total instantaneous charging current peak value and the total instantaneous charging power peak value of all charging loads can be reduced by adjusting the timing of all charging pulses without reducing the charging current and the charging power of any charging load. For the convenience of understanding and without loss of generality, it is assumed that the high level interval t on,a of pulse a is less than or equal to the low level interval t off,b of pulse b in a cycle T, where a, b ∈ [1, N c ] and t on,a , t off,b ∈ [0, T], then pulse b can be arranged to be turned on in the off period of pulse a, as shown in Fig. 5. By staggering the turn-on of pulse a and pulse b in time, the total current peak value of them is reduced, and thus the total charging power peak value of them is reduced. At this time, the total instantaneous charging current of them is the larger one of i a and i b (i.e. i a ) rather than the sum of i a and i b .

[0083] Generally, if the sum of the high level intervals of multiple pulses in a cycle is less than or equal to the low level interval of some other pulse, the high levels of these pulses can be arranged to appear in turn in a cycle, thereby reducing the total instantaneous current and the total instantaneous power. The present application proposes a method of grouping a series of charging pulses with equal cycles and staggering the high levels of them in time, which can reduce the total current peak value and the total power peak value of all charging loads. The specific principles and implementation steps are as follows:

[0084] It is assumed that the high level interval length of the pulse used by the i-th load in a cycle is t on,i , and the low level interval length is t off,i . x i is used to represent whether to charge other loads in the low level interval t off,i of pulse i, x i = 1 represents yes, and x i = 0 represents no. When x i = 1, pulse i can be regarded as a “container” of a pulse. y ij is used to represent whether to charge other loads in t off,iThe interval charges the j-th load, y ij =1 indicates that y ij =0 indicates no. When y ij When = 1, pulse j can be considered as an "item" to be loaded into other "containers". The problem addressed by this invention becomes a variant of the bin packing problem, namely: how to arrange all pulses into groups so that some pulses are merged into other pulses, so that the total number of groups is minimized, as shown in Figure 6. If all charging pulses can be divided into S groups, S∈[1,N] c In each group, the pulses appear sequentially at high levels without overlapping. To minimize the peak value of the total charging current for all charging loads, the value of S should be minimized, i.e., the number of "containers" should be minimized. The mathematical expression for this objective is as follows:

[0085] The mathematical representation of its constraints is as follows:

[0086] For any i∈[1,N] c ],have:

[0087] For any j∈[1,N] c ],have:

[0088] Inequality constraint 1 (Equation (7)) guarantees that within any given period, when pulse i acts as a "container," the sum of the high-level intervals of all other pulses loaded into that "container" (the left side of the inequality) does not exceed the capacity of the "container" (the right side of the inequality), meaning that the high levels of each pulse do not overlap in time. When pulse j is arranged in t off,i Conduction, y ij =1, at this time t on,j It is included in the left-hand side of the inequality. If pulse j is also a "container," then it is not grouped with pulse i. In this case, to avoid including the conduction period t of pulse j... on,j Also included in the calculation is the factor (1-x) on the left-hand side of the inequality. j At this time x j =1, 1-x j =0,t on,j It is not included in the left-hand side of the inequality. The right-hand side of the inequality contains the factor x. i x , if and only if pulse i is a "container" i =1, its capacity t off,i Only if it is included, otherwise the right side of the inequality is zero, indicating that pulse i is a pulse to be loaded.

[0089] Equality constraint 2 (Equation (8)) guarantees that pulse j exists and is unique within any period. For any j∈[1,N] ci from 1 to N c ij and summing them up. When pulse j is a "container", the sum is 0, meaning that pulse j will not be loaded as an "article" in other "containers"; otherwise, the sum is 1, indicating that pulse j will be loaded as an "article" in some other "container". In particular, if pulse j is not grouped with any other pulse, the method also regards pulse j as a "container", i.e. j x

[0090] By solving the mathematical model described by equations (6) to (8), the solutions of all x i and y ij are obtained, i.e. the grouping result of the pulses is obtained. The value of x i indicates which pulse is the "container" of other pulses, and the sum of all x i is the total number of groups S. All pulses j with the same i value and y ij = 1 are grouped into a group, where j ∈ [1, N c ]. For the pulses in the same group, their high levels are arranged to appear in sequence in time. For example: when x 12 = 1, y 13 = 1, y 14 = 0 and y c = 1, the high levels of pulse 2 and pulse 4 are arranged to appear in sequence in the low level interval of pulse 1, and the high level of pulse 3 is not arranged in this time period.

[0091] The arrangement of the high level interval of pulse j in the low level interval of pulse i can be achieved by moving pulse j in the time axis, i.e. changing the initial phase of pulse j. Assuming that the initial phase of pulse i is then the time when the first low level interval of pulse i appears is which can be set as the initial phase of pulse j. If the high level of pulse k is arranged to appear after the end of the high level of pulse j, the earliest possible time when it appears is which can be set as the initial phase of pulse k, and the initial phases of other pulses are set in the same way.

[0092] 4. A method for grouping and staggered sequencing of a series of pulses with unequal periods (grouping and sequencing method 2)

[0093] For the case where the charging pulse currents of all charging loads have different periods, the present application also proposes a method for grouping and sequencing them.

[0094] In this case, the pulse charging method is used to charge N cWhen multiple charging loads are charging simultaneously, the periods of the charging pulse currents are not all consistent, and their duty cycles and amplitudes are not all equal, as shown in Figure 7. If the frequency of any one pulse is not an integer multiple of the frequency of any other pulse, then these pulses cannot be grouped and their high-level intervals staggered. If the frequency of at least one pulse is an integer multiple of the frequency of another pulse, and the high-level interval of the lower-frequency pulse within a period is not longer than the low-level interval of the higher-frequency pulse, then these pulses can be grouped and their high-level intervals staggered within each group for charging. This reduces the total instantaneous charging current and total instantaneous charging power of all charging loads without reducing the charging current and charging power of any single charging load.

[0095] The problem addressed by this invention is: how to group pulses and stagger the pulses within each group in terms of timing, so that the high-level intervals of some pulses are located within the low-level intervals of others, and the high levels do not overlap in time, thereby reducing the total instantaneous charging current of all charging loads. This problem can be viewed as a variant of the binning problem, where the low-level interval of each pulse has the potential to accommodate the high levels of other pulses, i.e., it can become a "container" to hold the high levels of other pulses, as shown in Figure 8.

[0096] This invention proposes a method for grouping a series of charging pulses with unequal periods and staggering the high-level pulses within the same group, which can reduce the total peak current and total peak power of all charging loads. The specific principle and implementation steps are as follows:

[0097] If there is N c A load is simultaneously charged using the pulse charging method. The length of the high-level interval of the i-th pulse within one cycle is denoted as t. on,i The length of the low-level interval is denoted as t. off,i Use x i Indicates whether it is in the low-level range t of pulse i. off,i Internal charging of other loads, x i =1 means that x i =0 indicates no. When x i When y = 1, the low-level interval of pulse i within one cycle can be regarded as a "container" for the high-level interval of another pulse. This interval can then be used to charge other charging loads, thereby reducing the total instantaneous current peak. Using y... ij Indicates whether it is in pulse i at time t off,i The interval is arranged to represent the high-level interval of the j-th pulse, y ij =1 indicates that y ij =0 indicates no. When y ij When = 1, the high-level range of pulse j can be regarded as an "item" to be loaded into other "containers".

[0098] Unlike the case of grouping a series of pulses with equal periods, in order to make the high level intervals of two pulses with different periods appear in time one after another, one of the pulse periods must be an integer multiple of the other, otherwise there will be a time when the high level intervals of the two pulses overlap in time. Therefore, the grouping problem in this scenario only considers a series of pulses, where all the pulse periods are integer multiples of the highest frequency pulse period.

[0099] Consider a time period T, whose size is the least common multiple of all the pulse periods. Obviously, the T period covers N i low level intervals of pulse i, which is equivalent to there being N i "containers" in the T period, in which the high level intervals of other qualified pulses can be placed. Use variable k to number these low level intervals, as shown in Figure 9, and introduce a variable z ijk to represent whether the high level interval of pulse j is arranged in the kth low level interval of pulse i, z ijk = 1 means yes, and z ijk = 0 means no.

[0100] If all the charging pulses can be divided into S groups, S ∈ [1, N c ], and the pulses in each group appear in high level one after another and do not overlap in time. To make the total charging current peak of all charging loads as small as possible, the value of S should be as small as possible, that is, the number of "containers" should be as small as possible. The mathematical expression of this goal is as follows:

[0101] The mathematical representation of the constraint conditions is as follows:

[0102] 1) For any k ∈ [1, N i ] and any i ∈ [1, N c ], there is:

[0103] 2) For any k ∈ [1, N i ] and any i, j ∈ [1, N c ] and satisfying N i = N ij N j , there is:

[0104] where N ij is a positive integer.

[0105] 3) For any j ∈ [1, N c ], there is:

[0106] Inequality constraint 1) Restricts the sum of the lengths of the high-level intervals of all other pulses within the k-th low-level interval of the i-th pulse in one cycle to not exceed t. off,i The factor on the left side of the inequality (1-x) j This avoids including the pulse j, which acts as a "container," in the calculation; in this case, x... j =1, therefore 1-x j = 0. The factor x on the right side of the inequality. i If and only if pulse i is a "container", i.e., x i When = 1, its width t off,i Only if it is included, otherwise the right side of the inequality is zero, indicating that pulse i is a pulse to be loaded.

[0107] Equality constraint 2) Ensure that the period of pulse j is equal to T. j This constraint iterates through all low-level intervals of pulse i, and requires that N consecutive intervals be... ij z within a low-level interval ijk The sum is 1, so in any consecutive N of pulse i ij There is one and only one high-level interval for the j pulse in each low-level interval, thus ensuring that the period of the j pulse is N times that of the i pulse. ij times.

[0108] Equality constraint 3) stipulates that within a period, if any pulse is not a "container," it must and only once be grouped with a single other pulse. Specifically, if a pulse j is not grouped with any other pulse, this method also considers pulse j as a "container," i.e., x... j =1. For any j∈[1,N] c The left side of the equation iterates through i from 1 to N. c All y ij And sum them. When pulse j is a "container", x j =1, 1-x j =0, the sum is 0, which means that pulse j will not be loaded into other "containers" as an "item"; otherwise, the sum is 1, which means that pulse j will be loaded into other "containers" as an "item".

[0109] All x are obtained by solving the mathematical models of equations (9) to (12). i y ij and z ijk The value of x will give the pulse grouping result. i The value of x indicates which pulses are "containers" for other pulses. i The sum of these is the total number of groups, S. ij The grouping arrangement is specified, all groups with the same i value and y ijPulse j with z c =1 is grouped into a set, where i, j ∈ [1, N ijk The specific arrangement of the interval is indicated: z ijk =1 indicates that the high level interval of pulse j can be arranged in the kth low level interval of pulse i; z ijk =0 indicates that it cannot be arranged. For all high levels of pulses in the same "container", they can be arranged to appear in time in turn, but their positions in other "containers" must be consistent. For example: when x1=1, y 12 =1, y 13 =1, y 14 =1, z 121 =z 123 =1, z 122 =z 124 =0, z 131 =z 134 =1, z 132 =z 133 =0, z 141 =z 143 =1, z 142 =z 144 =0, the high levels of pulse 2 and pulse 4 are arranged in the low level intervals 1 and 3 of pulse 1 instead of intervals 2 and 4, as shown in Figure 10; and the high level intervals of each pulse are consistent in the positions of the low level intervals of pulse 1, that is, the positions of the high level intervals of pulses 2, 3, and 4 in this interval are consistent with their positions in interval 1.

[0110] Arranging the first high level interval of pulse j in the kth low level interval of pulse i can be achieved by moving pulse j on the time axis, that is, changing the initial phase of pulse j. Assuming the initial phase of pulse i is , then the time when the kth low level interval of pulse i appears is At this time, the initial phase of pulse j can be set as If the first high level of a pulse is arranged to appear after the end of the high level of pulse j, the earliest possible time of its appearance is At this time, the initial phase of pulse k can be set as The initial phases of other pulses are similar.

[0111] Since grouping and sorting method 2 requires that the periods of all pulses to be grouped are integer multiples of the period of the pulse with the highest frequency, before using grouping and sorting method 2, all charging pulses need to be grouped and sorted respectively. The specific operation steps are shown in Figure 11:

[0112] 1) Initialize a positive integer variable n, and let n=1;

[0113] 2) Find the highest frequency pulse among all the ungrouped charging pulses, denoted as I n (f n );

[0114] 3) Group all the pulses with frequency f n integer multiple x, denoted as I nx (f nx ), where x is a positive integer;

[0115] 4) Group and orderly control all the pulses in the group using the designed optimal grouping and sorting method 2;

[0116] 5) If there are still pulses ungrouped, let n = n + 1, repeat the above process until all the pulses are grouped and arranged.

[0117] 5. A method for overall coordination of charging pulses

[0118] Considering the upper limit of total charging power, the present application proposes a method for overall coordination of charging sequence of all charging loads and orderly arrangement of charging pulse on-off timing, so as to ensure that the total power does not exceed the upper limit, or to cooperate with the demand side response of the power grid.

[0119] The first step of the method is to determine which grouping and sorting method to use according to the frequency of the pulses. If all the charging pulses have the same frequency, the grouping and sorting method 1 proposed in the present application can be used to group and sort all the charging pulses; if not all the charging pulses have the same frequency, the grouping and sorting method 2 proposed in the present application needs to be used to group and sort all the charging pulses. As shown in FIG. 12, after determining the grouping and sorting, the total charging power P Σ of all the loads can be calculated. If there is an upper limit P max for the total charging power of all the loads, such as participation in the demand side response of the power grid, etc., it is necessary to determine whether to charge all the loads at the same time according to the size relationship between P max and P Σ .

[0120] If P Σ <P max , orderly charging is performed using the grouping and sorting method; if P Σ ≥ P max , all the cars need to be grouped according to the charging plan, for which the cars can be divided into non-postponable group and postponable group according to whether their charging plan can be postponed.

[0121] First, calculate the total charging power P Σ of all the cars in the non-postponable group, if P Σ >P max, the total charging power of all loads in the group needs to be limited. With the charging voltage unchanged, the total charging power P i can be reduced by reducing the charging current I Σ of all vehicles, i.e. The reduced current I

[0122] Without changing the amplitude of the charging current of each vehicle, the total charging power P i can also be reduced by reducing the duty cycle d Σ of each pulse, i.e. The formula for calculating the reduced duty cycle d

[0123] If the method of reducing the duty cycle d Σ is used to reduce the total charging power P , after the duty cycle d is calculated, the grouping sorting method 1 or the grouping sorting method 2 proposed in the present application can be used to rearrange the order of the charging pulses of the vehicles and thus perform orderly charging.

[0124] If P Σ <P max , the charging station can also select to delay the charging of some vehicles in the group.

[0125] For the delayable group, the vehicles are sorted according to the charging priority c i , where c i ∈ [0, 1], and the larger the value, the higher the priority. If it is not possible to charge all vehicles in the delayable group, the m vehicles with higher priority in the group are charged first, where m should satisfy formula (15):

[0126] where P i represents the charging power of the i-th load.

[0127] For the vehicles with lower priority, the charging can be suspended before P max increases or other vehicles reduce the charging power.

[0128] In embodiment 1, all charging pulse periods are equal

[0129] Consider 10 loads charging simultaneously by pulse charging method, all pulse current amplitudes and periods are equal, but initial phase and duty cycle are inconsistent, as shown in Table 1. When the initial phase of each pulse is randomly distributed, the waveform is shown as a solid line in FIG. 13, and the total current after superposition of each pulse current is shown as a solid line in FIG. 14. The total current after superposition of each pulse current when randomly arranged is shown as a solid line in FIG. 14, the maximum value is 100 A, the minimum value after stabilization is 20 A, and the fluctuation amplitude is 80 A. As can be seen, when all loads are charged in disorder, the total current peak value and fluctuation amplitude are large.

[0130] Table 1 Pulse current parameters of a series of equal periods

[0131] The pulses are grouped and sorted by the grouping and sorting method 1 of the application, and the calculation results are shown in Table 2. At this time, x2=x3=x8=x9=x 10 =1, which means that the low level interval of pulses 2, 3, 8, 9 and 10 can be used as a "container" for the high level interval of other pulses, and the high level interval of other pulses can be arranged in the low level interval of pulses 2, 3, 8, 9 and 10. Specifically, since y 27 =y 56 =y 84 =y 91 =1, the high level interval of pulse 7 can be arranged in the low level interval of pulse 2, the high level interval of pulse 6 can be arranged in the low level interval of pulse 5, the high level interval of pulse 4 can be arranged in the low level interval of pulse 8, and the high level interval of pulse 1 can be arranged in the low level interval of pulse 9. Thus, the initial phase of pulses 1, 4, 6 and 7 needs to be adjusted, and the values after grouping and sorting are shown in the last row of Table 1. It needs to be specially pointed out that since for any j∈[1,10], the value of y 10j is 0, even if x 10 =1, the high level interval of other pulses is not arranged in the low level interval of pulse 10, and the low level interval of pulse 10 can be regarded as an empty "container".

[0132] Table 2 Calculation results of grouping and sorting pulses by grouping and sorting method 1 of the application

[0133] Due to the adjustment of the initial phase, the waveforms of the pulses 1, 4, 6 and 7 are shifted on the time axis after grouping and sorting, as shown by the dashed lines in FIG. 13. The total current of the superposition of the current of each pulse after grouping and sorting is shown by the dashed line in FIG. 14. The peak value is reduced from 100 A to 60 A, a reduction of 40%, and the fluctuation amplitude is reduced from 80 A to 10 A, only 12.5% of the original value. It can be seen that the method of the present application can reduce the total current and the fluctuation amplitude of the charging load of multiple equal-period pulses.

[0134] Example 2 All charging pulse periods are not equal

[0135] It is considered that 10 loads are simultaneously charged by the pulse charging method, and the amplitudes of the pulse currents are equal, but the periods, initial phases and duty cycles are inconsistent, as shown in Table 3. When the initial phases of the pulses are randomly distributed, the waveform is shown by the solid line in FIG. 15, and the total current of the superposition of the pulse currents is shown by the solid line in FIG. 16. The total current of the superposition of the pulse currents when randomly arranged is shown by the solid line in FIG. 16. The maximum value is 90 A, the minimum value after stabilization is 10 A, and the fluctuation amplitude is 80 A. It can be seen that when all the loads are charged in disorder, the peak value and the fluctuation amplitude of the total current are large.

[0136] Table 3 A series of pulse current parameters with unequal periods

[0137] The grouping and sorting method 2 of the present application is used to group and sort the pulses, and the calculation results are shown in Table 4. The values in the brackets in the table represent the value of k when z ijk = 1 appears for the first time, indicating that the first high-level interval of the pulse j appears in the kth low-level interval of the pulse i. In this example, x1 = x2 = x4 = x6 = x 10 = 1, meaning that the low-level intervals of the pulses 1, 2, 4, 6 and 10 can be used as "containers" for the high-level intervals of other pulses, and the high-level intervals of other pulses can be arranged in the low-level intervals of the pulses 1, 2, 4, 6 and 10. Specifically, since y 19 = y 27 = y 43 = y 68 = y 105 = 1, the high-level interval of the pulse 9 can be arranged in the low-level interval of the pulse 9, the high-level interval of the pulse 7 can be arranged in the low-level interval of the pulse 2, the high-level interval of the pulse 3 can be arranged in the low-level interval of the pulse 4, the high-level interval of the pulse 8 can be arranged in the low-level interval of the pulse 6, and the high-level interval of the pulse 5 can be arranged in the low-level interval of the pulse 10. Therefore, the initial phases of the pulses 3, 5, 7, 8 and 9 need to be adjusted, and the values after grouping and sorting are shown in the last row of Table 3.

[0138] Table 4: calculation results of grouping and sequencing the pulses by using the grouping and sequencing method 2 of the present application

[0139] Due to the adjustment of the initial phase, the waveforms of the grouped and sequenced pulses 3, 5, 7, 8 and 9 are all shifted on the time axis, as shown by the dashed lines in Fig. 15. The total current of the superposition of the currents of the grouped and sequenced pulses is shown by the dashed line in Fig. 16. The peak value of the total current is reduced from 90A to 60A, with a reduction of about 33.33%, and the fluctuation amplitude is reduced from 80A to 20A, which is only 25% of the original value. It can be seen that the method of the present application can reduce the total current and the fluctuation amplitude of the charging load of multiple unequal period pulses.

[0140] The charging load management method of the present application for the pulse charging method can also be used for managing a series of events with periodic occurrence, such as:

[0141] 1) several different sizes of articles are conveyed by a conveyor belt, and the size and interval of each type of article are the same, but the size and interval of different types of articles are different. By using the technical solution of the present application, different articles can also be mixedly conveyed on the same conveyor belt, and the required number of conveyor belts is minimized without changing the original interval of the same type of articles.

[0142] 2) several different signals are conveyed through a channel, and the duty cycle and period of each type of signal are the same, but the duty cycle and period of different types of signals are different. By using the technical solution of the present application, different signals can also be mixedly conveyed on the same channel, and the required number of channels is minimized without changing the information of the same type of signal.

[0143] The above only describes the preferred embodiments of the present application, and it should be noted that for those skilled in the art, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A charge load management method for a pulse-charging method, characterized by, The method comprises a different charging pulse staggered arrangement method, when the charging pulse is a pulse current, the pulse current is low in t off period T, and the current value is zero; assuming that the low interval of pulse current 1 in a period T is t off,1 , the high interval of pulse current 2, 3, …, n in a period T is t on,2 , t on,3 , …, t on,n ; when formula (1) is established, the high intervals of the pulse currents are arranged to appear in turn in the low interval t off,1 , so that the n pulse currents do not overlap in time, and the total current peak value of the n pulse currents is reduced. Wherein n∈[1,+∞]#(1).

2. The charge load management method according to claim 1, characterized by, When the charging pulse is a pulse voltage, the pulse voltage is low in the time interval t off in each cycle T, and the voltage value is zero; assuming that the low interval of the pulse voltage 1 in one cycle T is t off,1 , the high interval of the pulse voltage 2, 3, …, n in one cycle T is t on,2 , t on,3 , …, t on,n , respectively; when formula (1) is established, the high intervals of these pulse voltages are arranged to appear in turn in the low interval t off,1 , so that the n pulse voltages do not overlap in time, and the total voltage peak value of the n pulse voltages is reduced.

3. The charge load management method for the pulse-charge method according to claim 1, characterized by, The application also provides a pulse current waveform adjusting method. If the remaining single charging pulses do not satisfy the formula (1), the charging pulse staggered sequence cannot be realized. Therefore, the pulse current waveform is adjusted by changing the duty ratio and amplitude, while the average charging power in a cycle is kept unchanged. Before the adjustment, the pulse current amplitude is I m , the cycle is T, the high level interval in a cycle T is t on , the low level interval is t off , and the duty ratio is The average charging power of the pulse current is calculated by the following formula: Wherein u is the charging voltage, the average charging power in one period T is changed by changing the charging voltage, the charging current or the duty cycle; after adjustment, the period is unchanged, the high level interval in one period T is t' on , the low level interval is t' off , and the duty cycle is The charging voltage and charging current are adjusted to U' and I', respectively; to keep the average charging power unchanged before and after, only need to ensure that equation (3) is established: By adjusting the remaining n pulse current waveform, so that at least one of the conditions (1) to meet the different charging pulse staggered sequencing method to achieve different charging pulse high level staggered.

4. The charge load management method for the pulse-charge method according to claim 1, characterized by, Also included is a method of grouping and staggering a series of equal period pulses, if there are N c charging loads being charged simultaneously, each load being charged using the pulse charging method, all charging pulse currents having a period equal to T, but not all having equal duty cycles and amplitudes, then the instantaneous charging current peak i c for the N max charging loads is the sum of all the pulse currents. If the charging voltage of the charging load i is u i , then the instantaneous charging power peak value p max of all charging loads is the sum of all pulse instantaneous powers; assuming that the high level interval t on,a of a pulse a in a period T is less than or equal to the low level interval t off,b of another pulse b, a, b ∈ [1, N c ] and t on,a , t off,b ∈ [0, T], then pulse b can be arranged to be turned on in the off period of pulse a, the total current peak value is reduced by staggering the turn-on of pulse a and pulse b in time, the charging total power peak value is reduced, at this time the instantaneous charging total current is the larger one of i a and i b rather than the sum of i a and i b .

5. The charge load management method for the pulse-charge method according to claim 4, characterized by, The specific operation steps of the method of grouping and staggered sequencing of a series of pulses with equal periods are as follows: Assume that the length of the high level interval of the pulse used by the i-th load in a cycle is t on,i , the length of the low level interval is t off,i ; x i represents whether to charge other loads in the low level interval t off,i of the pulse i, x i = 1 means yes, x i = 0 means no; when x i = 1, the pulse i is regarded as a "container” of a pulse; y ij represents whether to charge the j-th load in the t off,i interval, y ij = 1 means yes, y ij = 0 means no; when y ij = 1, the pulse j is regarded as an "item” to be loaded into other "containers”; all charging pulses are divided into S groups, S ∈ [1, N c ], the pulses in each group appear high level in turn and do not overlap, to make the total current peak of all charging loads as small as possible, the value of S should be as small as possible, that is, the number of "containers” should be as small as possible, the mathematical expression of the target is: The mathematical representation of the constraint condition is as follows: 1) for any i e [1, N c ], there is: 2) for any j e [1, N c ], there is: Inequality constraint 1) guarantees that within any period, when pulse i acts as a "container", the sum of the high-level intervals of all other pulses loaded into that "container" does not exceed the capacity of the "container", that is, the high levels of each pulse do not overlap in time; when pulse j is arranged in t off,i Conduction, y ij =1, at this time t on,j The left-hand side of the inequality is included; if pulse j is also a "container", then it is not grouped with pulse i. In this case, to avoid including the conduction period t of pulse j... on,j Also included in the calculation is the factor (1-x) on the left-hand side of the inequality. j At this time, x j =1, 1-x j =0,t on,j It is not included in the left-hand side of the inequality; the right-hand side of the inequality includes the factor x. i x , if and only if pulse i is a "container" i =1, its capacity t off,i Only if it is included, otherwise the right side of the inequality is zero, indicating that pulse i is a pulse to be loaded; Equation constraint 2) guarantees that for any one period, impulse j exists and is unique; for any j e [1, N c ], the left-hand side traverses all y c 's from 1 to N ij and sums them up; when impulse j is a "bin", the sum is 0, meaning that impulse j will not be loaded as an "item" in some other "bin"; otherwise, the sum is 1, indicating that impulse j will be loaded as an "item" in some other "bin". In particular, if impulse j is not grouped with any other impulse, then impulse j is considered a "bin", i.e., x j = 1; All x i and y ij solutions, i.e. grouping results of the pulses, are obtained by solving equations (6) to (8); the value of x i indicates which pulses are "containers" of other pulses, and the sum of all x i is the total number of groups S; all pulses j with the same i value and y ij = 1 are grouped into a group, where j ∈ [1, N c ]; for the pulses in the same group, arrange them to appear in time sequence high.

6. The charge load management method for the pulse-oriented charging method according to claim 4, characterized by, Also included is a method of grouping and staggering a series of periodic unequal pulses to simultaneously charge N c charge loads using pulse charging method, wherein the charging pulse currents are not all of the same period and the duty cycles and amplitudes are not all equal. If the frequency of any one pulse is not an integer multiple of the frequency of any other pulse, then these pulses cannot be grouped and have their high periods staggered. If the frequency of at least one pulse is an integer multiple of the frequency of another pulse and the high period of the lower frequency pulse is not longer than the low period of the higher frequency pulse, then these pulses can be grouped and have their high periods staggered to reduce the instantaneous total charging current and instantaneous total charging power of all the charge loads while not reducing the charging current and charging power of any one charge load.

7. The charge load management method for a pulse-charge method according to claim 6, characterized by, The specific operation steps of the method of grouping and staggered sequencing of a series of pulses with unequal periods are as follows: First, group all charging pulses and sort each group; if there are N c A load is simultaneously charged using the pulse charging method. The length of the high-level interval of the i-th pulse within one cycle is denoted as t. on,i The length of the low-level interval is denoted as t. off,i ; use x i Indicates whether it is in the low-level range t of pulse i. off,i Internal charging of other loads, x i =1 means that x i =0 indicates no; when x i When y = 1, the low-level interval of pulse i within one cycle is regarded as a "container" for the high-level interval of another pulse. This interval can then be used to charge other charging loads to reduce the total instantaneous current peak. ij Indicates whether it is in pulse i at time t off,i The interval is arranged to represent the high-level interval of the j-th pulse, y ij =1 indicates that y ij =0 indicates no; when y ij When = 1, the high-level range of pulse j is regarded as an "item" to be loaded into other "containers"; Assuming that all pulse periods are integer multiples of the highest frequency pulse period, consider a time period T, the size of which is the least common multiple of all pulse periods, the T period encompasses N i low intervals of pulse i, equivalent to there being N i "vessels" in the T period, within which to place the high intervals of other eligible pulses; These low level intervals are numbered using a variable k, while a variable z ijk is introduced to indicate whether or not the high level interval of pulse j is arranged in the kth low level interval of pulse i, z ijk = 1 means yes, z ijk = 0 means no; Divide all charging pulses into S groups, S∈[1,N c ], and the pulses in each group appear high level in turn and do not overlap in time; to make the total charging current peak value of all charging loads as small as possible, the value of S should be as small as possible, that is, the number of "containers" should be as small as possible, and the mathematical expression of this target is: The mathematical representation of the constraint condition is as follows: 1) for any k e [1, N i ] and any i e [1, N c ], there is: 2) for any k e [1, N i ] and any i,j e [1, N c ] satisfying N i = N ij N j , we have: wherein N is a positive integer; and ij is a positive integer; and 3) for any j e [1, N c ] has: Inequality constraint 1) limits the sum of the lengths of the high level intervals of all other pulses within a cycle within the kth low level interval of the ith pulse to not exceed t off,i ; the factor (1 - x j ) on the left side of the inequality prevents the inclusion of pulse j as a "container" in the calculation, when x j = 1, so that 1 - x j = 0; the factor x i on the right side of the inequality, when and only when pulse i is a "container", i.e., when x i = 1, its width t off,i is counted, otherwise the right side of the inequality is zero, indicating that pulse i is a pulse to be loaded. Equation constraint 2) ensures that the period of pulse j is equal to T j ; this constraint traverses all low intervals of pulse i and requires that the sum of z ij over any interval of N ijk consecutive low intervals of pulse i is 1, so that there is and only one high interval of pulse j in any interval of N ij consecutive low intervals of pulse i to ensure that the period of pulse j is N ij times the period of pulse i; Equation constraint 3) limits that, within a cycle, if any one pulse is not a "container", it must and only be grouped with one and only one other pulse; if a pulse j is not grouped with any other pulse, pulse j is considered a "container", i.e. x j = 1 ; for any j e [1, N c ], the left-hand side sums over all y c 's as i goes from 1 to N ij and sums them up; when pulse j is a "container", x j = 1, 1 - x j = 0, the sum is 0, meaning that pulse j is not loaded as an "item" in another "container"; otherwise, the sum is 1, indicating that pulse j is loaded as an "item" in another "container"; All values of x i , y ij and z ijk are obtained by solving equations (9) to (12), i.e. the grouping result of the pulses; x i indicates which pulses are the "containers" of other pulses, the sum of x i is the total number of groups S; y ij indicates the grouping arrangement, all pulses j with the same i value and y ij = 1 belong to a group, i, j ∈ [1, N c ]; z ijk indicates the specific arrangement of the intervals: z ijk = 1 indicates that the high level interval of pulse j is arranged in the kth low level interval of pulse i; z ijk = 0 indicates that it cannot be arranged in this way; for all high levels of pulses in the same "container", they are arranged to appear in time in turn, while ensuring that their positions in other "containers" are consistent; the first high level interval of pulse j is arranged in the kth low level interval of pulse i, which is realized by moving pulse j in the time axis, i.e. changing the initial phase of pulse j.

8. The charge load management method for the pulse-charge method according to claim 7, characterized by, The method of grouping and sequencing all charging pulses is as follows: 1) Initialize the positive integer variable n, let n = 1; 2) Find the highest frequency pulse among all the ungrouped charging pulses, call it I n (f n ); 3) all pulses with frequency f n are grouped into a set, denoted by I nx (f nx ), x is a positive integer; 4) Group and sequence all pulses in the group using the method of grouping and staggered sequencing of a series of pulses with unequal periods; 5) If there are still pulses to be grouped, let n = n + 1, repeat the above process until all pulses are grouped and arranged.

9. The charge load management method for the pulse-charge method according to claim 6, characterized by, Also includes a charging pulse conduction timing coordination method, first according to the frequency of the pulse to determine which grouping and sequencing method to use; if all the charging pulse frequency is consistent, then use the method of grouping and staggered sequencing of a series of pulses with equal periods to group and sequence all charging pulses; if not all the charging pulse frequency is equal, then need to use the method of grouping and staggered sequencing of a series of pulses with unequal periods to group and sequence all charging pulses.

10. The charge load management method for a pulse-charge method according to claim 9, wherein After the grouping and sequencing are determined, the total charging power P of all loads is calculated Σ If the total charging power P of all loads has an upper limit P max , then according to the size relationship between P max and P Σ , it is determined whether to charge all loads at the same time; If P Σ ≥ P max , all the cars are grouped according to the charging plan, and the cars are divided into a non-postponable group and a postponable group according to whether their charging plans can be postponed; first, the total charging power P Σ of all the cars in the non-postponable group is calculated, and if P Σ >P max , the total charging power of all the loads in the group needs to be limited; in the case of unchanged charging voltage, the total charging power P i is reduced by reducing the charging current I Σ of all the cars, and the reduced current I The calculation formula is: Without changing the amplitude of the charging current of each vehicle, by reducing the duty cycle d i of each pulse to reduce the total charging power P Σ , the reduced duty cycle d The calculation formula is: If the method of reducing the duty cycle is used to reduce the total charging power P Σ After the duty cycle is calculated Then, use the method of grouping and staggered sequencing of a series of pulses with equal periods or the method of grouping and staggered sequencing of a series of pulses with unequal periods to rearrange the order of the car charging pulse and perform orderly charging; If P Σ max , the charging station selects some cars in the postponable group to charge; for the postponable group, the cars are sorted according to the charging priority c i , c i ∈[0,1], the greater the value, the higher the priority; if it is not possible to charge all cars in the postponable group, the first m cars with higher priority in the group are charged, where m should satisfy formula (15): ​ where P i represents the charging power of the ith load; for lower priority cars, P max suspends charging until the charging power is increased or other cars decrease the charging power.

Images