Rule-based distributed modular hybrid power system energy management method

Abstract

The invention relates to the technical field of energy management, and particularly discloses a rule-based distributed modular hybrid power system energy management method, which comprises the following steps of: obtaining target power and target torque of a power system under different working conditions; the target power and the target torque are cooperatively distributed through fuzzy control, and torque and power needed by all the power units and the hydraulic motors are obtained; respectively obtaining super capacitor charging and discharging power, battery charging and discharging power, power unit distribution power and battery compensation power by utilizing low-frequency power and high-frequency power distributed to the battery, each power unit and the super capacitor; obtaining an optimal power distribution strategy of the power units by integrating the influence of the environmental change factors on the distribution power of the power units, and obtaining the optimal distribution power of each power unit based on the optimal power distribution strategy of the power units; efficient energy management of the power system under multiple working conditions is achieved so as to adapt to environment change and self-adaptive adjustment of energy output of the power unit.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy management, and more particularly to a rule-based energy management method for a distributed modular hybrid power system. BACKGROUND

[0002] The energy management strategy designed according to the rule determines the operating state of the motor and the engine according to various different driving conditions, and limits the operating parameters of the motor and the engine in a certain region to achieve different expected targets. The energy management strategy adjusts the operating state of the transmission system and the power source according to the determined principle to ensure high efficiency operation in the predetermined region, and the region corresponds to a specific working interval in the engine steady-state MAP graph.

[0003] However, the control of the existing modular power system is relatively complex, but it has high freedom, and each cylinder can work independently, which is a big advantage of the modular power system: in the part-load condition, part of the power unit can be disabled, and the other working units can be kept in a large load and high efficiency operating state, thereby improving the overall efficiency of the system.

[0004] However, due to uneven energy distribution, the cumulative working time of each power unit is quite different, which will lead to a decrease in the durability of the frequently working power unit, affect the coupled work of each unit in the power module, and further cause the overall working efficiency of the modular power system to decrease.

[0005] If the power demand is evenly distributed to all power units, it may lead to low load of all units in the part-load condition, thereby failing to achieve high efficiency operation of the system. In some working points, the power distribution ratio of the minimum number of power units plus one is higher than the power distribution ratio of the minimum number of power units, and the number of working units required for optimal overall efficiency under different demands is different. SUMMARY

[0006] The purpose of the present application is to provide a rule-based energy management method for a distributed modular hybrid power system, which integrates fuzzy control, automatic high-low frequency power matching algorithm and optimal power distribution strategy of power units, realizes efficient energy management of the power system under multiple working conditions, and adapts to the adaptive adjustment of power unit energy output.

[0007] To achieve the above purpose, the present application provides the following technical scheme:

[0008] A rule-based energy management method for a distributed modular hybrid power system, comprising the following steps:

[0009] Obtaining the target power and target torque of the power system under different working conditions;

[0010] The target power and target torque of the power system are allocated by fuzzy control to obtain the torque and power required by each power unit and hydraulic motor in the power system;

[0011] The total demand power of each power unit and hydraulic motor in the power system is decomposed into low-frequency power and high-frequency power by an automatic high-low frequency power matching algorithm, and is allocated to the battery, each power unit and super capacitor to obtain the super capacitor charging and discharging power, the battery charging and discharging power, the power unit allocation power and the battery compensation power, respectively.

[0012] The influence of the comprehensive environmental change factor on the power unit allocation power is obtained to obtain the optimal power allocation strategy of the power unit, and the optimal allocation power of each power unit is obtained based on the optimal power allocation strategy of the power unit to realize adaptive adjustment of the energy output of the power unit under environmental change.

[0013] Further, the total demand power of each power unit and hydraulic motor in the power system is decomposed into low-frequency power and high-frequency power, and is allocated to the battery, each power unit and super capacitor, specifically:

[0014] The low-frequency power is allocated to the battery and each power unit based on the efficiency-optimal low-frequency power allocation strategy.

[0015] The high-frequency power is allocated to the super capacitor based on the high-frequency power rule allocation strategy.

[0016] Further, the super capacitor charging and discharging power acquisition method is specifically:

[0017] The capacitor charging and discharging power is obtained by energy management of the super capacitor based on the super capacitor charging and discharging strategy.

[0018] Further, the battery charging and discharging power acquisition method is specifically:

[0019] The battery charging and discharging power is obtained based on the power battery charging and discharging strategy, in combination with the energy storage boundary, temperature, charging and discharging instruction and the allocation power of the battery.

[0020] Further, the power unit optimal power allocation strategy and the battery compensation power acquisition method are specifically:

[0021] The power unit optimal power allocation strategy and the battery compensation power are obtained based on the temperature change compensation strategy, in combination with the allocation power of the power unit, temperature change and altitude.

[0022] Further, the optimal allocation power of each power unit is obtained by optimizing the power unit working time rolling matrix based on the power unit optimal power allocation strategy, including the following steps:

[0023] acquire total demand power of the power system;

[0024] calibrate optimal working efficiency based on conventional characteristics of the power system, and formulate two different rules to obtain two power distribution matrices respectively;

[0025] compare the efficiency of the two power unit working time rolling matrices to obtain the number of power units in working state;

[0026] initialize pre-optimization parameters, determine all optimization directions, update iteration factors for the optimization directions, record the average efficiency of each optimization direction, and judge whether each optimization direction reaches the iteration number; if yes, compare the optimization effect of each optimization direction to determine the optimal optimization path, and if no, constantly update the iteration factors for the optimization directions until the iteration number is reached to obtain the best optimization path;

[0027] initialize main optimization parameters, update the iteration factors and distribution power of each power unit, and judge whether the update process reaches the iteration number; if yes, output the optimal power distribution strategy, and if no, constantly update the iteration factors and distribution power of each power unit until the iteration number is reached to obtain the optimal power distribution power of each power unit.

[0028] According to the specific embodiments provided by the present application, the following technical effects are disclosed:

[0029] The present application can change the power distribution of the power units according to the changes of parameters such as temperature and altitude to achieve the optimal power distribution of each power unit; the energy output of the power units can be adaptively adjusted when the environment changes; the present application has the advantages of simple algorithm, stability, strong practicability, and can be used for real vehicles, etc., avoiding the decrease of the comprehensive working efficiency of the modular power system caused by uneven power distribution of the power units; the working efficiency of the double-iteration optimization algorithm based on rules is high, avoiding the decrease of the comprehensive working efficiency of the modular power system caused by uneven power distribution of the power units. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present application, and those skilled in the art can obtain other drawings according to the provided drawings without creative labor.

[0031] The energy management method of the distributed modular hybrid power system based on rules will be further described below in combination with the drawings;

[0032] Figure 1It is the overall schematic diagram of the rule-based distributed modular hybrid power system energy management method provided by the application.

[0033] Figure 2 It is the flow chart of the optimized distribution power in the rule-based distributed modular hybrid power system energy management method provided by the application.

[0034] Figure 3 It is the high-low frequency decomposition schematic diagram in the rule-based distributed modular hybrid power system energy management method provided by the application. DETAILED DESCRIPTION

[0035] The specific embodiments of the application are further described in detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate the application, but are not used to limit the scope of the application.

[0036] In order to better understand the purpose, structure and function of the application, the application is further described in detail below with reference to the accompanying drawings.

[0037] The application provides a rule-based distributed modular hybrid power system energy management method:

[0038] 1. In view of the cumulative working time imbalance of power distribution to each power unit, a power unit working time proportion balancing strategy is proposed, which avoids the decline of the comprehensive working efficiency of the modular power system caused by uneven power distribution of the power unit.

[0039] 2. On the premise of meeting the total demand power, a rule is established to ensure that more power units work in the high-efficiency interval, and a power distribution matrix is obtained based on the rule. Then an optimization method is proposed to optimize the rule power distribution matrix, so as to realize the real-time working efficiency improvement of the modular power system. The modular power system has the characteristics of multiple power units working together, and the comprehensive working efficiency can be dynamically adjusted by changing the number of power units. For a certain total demand power, as few power units as possible are called, and as many power units as possible in the power units are controlled to work at the optimal working point, so as to realize the real-time working efficiency optimization of the modular power system.

[0040] 3. The application uses a double-iteration optimization algorithm for control, realizing the optimal control of the comprehensive efficiency distribution of the power system in all working conditions.

[0041] The application provides the following specific embodiments:

[0042] As shown in Figure 1 The rule-based distributed modular hybrid power system energy management method provided by the embodiment of the application includes the following steps:

[0043] Obtaining target power and target torque of the power system under different working conditions;

[0044] Coordinately distributing the target power and target torque of the power system by fuzzy control to obtain the torque and power required by each power unit and hydraulic motor in the power system;

[0045] Using an automatic high-low frequency power matching algorithm to decompose the total required power of each power unit and hydraulic motor in the power system into low frequency power and high frequency power, and distribute them to the battery, each power unit and super capacitor, to obtain the super capacitor charging and discharging power, the battery charging and discharging power, the power unit distribution power and the battery compensation power, respectively;

[0046] Obtaining the optimal power distribution strategy of the power unit by comprehensively considering the influence of environmental changes on the power unit distribution power, and optimizing the power unit working time rolling matrix based on the optimal power distribution strategy of the power unit to obtain the optimal distribution power of each power unit, so as to realize adaptive adjustment of the energy output of the power unit under environmental changes.

[0047] As shown in Figure 3 , the total required power of each power unit and hydraulic motor is decomposed into low frequency power and high frequency power, and is distributed to the battery, each power unit and super capacitor, specifically:

[0048] Based on the low frequency power distribution strategy optimized in terms of efficiency, the low frequency power is distributed to the battery and each power unit;

[0049] Based on the high frequency power rule distribution strategy, the high frequency power is distributed to the super capacitor.

[0050] The method for obtaining the super capacitor charging and discharging power, specifically:

[0051] Based on the super capacitor charging and discharging strategy, the energy of the super capacitor is managed to obtain the capacitor charging and discharging power.

[0052] The method for obtaining the battery charging and discharging power, specifically:

[0053] Based on the power battery charging and discharging strategy, and combined with the energy storage boundary, temperature, charging and discharging instructions and the distribution power of the battery, the battery charging and discharging power is obtained.

[0054] In this embodiment, based on the power battery charging and discharging strategy, and combined with the energy storage boundary, temperature, charging and discharging instructions and the allocated power of the battery, the battery charging and discharging power is obtained, specifically: the working boundary of the SOC value of the power battery and the super capacitor is set to 40%-90% in advance; the charging and discharging strategy is set: when SOC≤40%, the power battery and the super capacitor start charging, and when the SOC of the energy storage unit is greater than or equal to 90%, the power battery and the super capacitor stop charging and start discharging; based on the rule-based algorithm of the preset power unit target power, the obtained target demand power is allocated and calculated according to the high and low optimal power calculation points. The output power of the power unit is only an integer multiple of the optimal efficiency point output power. Since the power unit has a problem of power output surplus or deficit when facing different demand powers, the power battery needs to perform "peak clipping and valley filling" at this time. The extra output power of the power unit is used to charge the power battery, and the vacancy of the output power of the power unit is supplemented by the power battery. Since the power unit target power is based on the rule-based algorithm and is divided into two ways of high and low frequency optimal power matching, the total efficiency of the power unit is optimal, and the final result obtained by comparing which power allocation algorithm is better, so that the comprehensive efficiency of the power unit under different demand powers is always optimal, wherein the way to ensure the optimal efficiency of the power unit is to ensure that the power unit always works at the highest efficiency point, otherwise the power unit is prohibited from working. The steady-state demand is borne by the power battery and the power unit, and the power unit needs to work at the optimal efficiency point. The transient fluctuation is borne by the super capacitor, reducing the loss of frequent charging and discharging of the battery.

[0055] The process of obtaining the battery charging and discharging power is described as follows:

[0056] The charging limit of the battery: if SOC≥90%, the battery charging power Pbat_chg=0, and the excess power is handled by the super capacitor or other means.

[0057] The discharge limit of the battery: if SOC≤40%, the battery discharging power Pbat_dis=0, and the insufficient power is adjusted or limited by the power unit.

[0058] The total charging / discharging power of the battery needs to integrate low-frequency allocation, high-frequency compensation and dynamic surplus / deficit adjustment, and the formula is: bat_total Pbat_total=Pbat_low+Pbat_high+ΔPbat bat_low bat_high

[0059] Let the total demand power Ptotal be decomposed into low-frequency power Plow and high-frequency power Phigh, that is: 总 Ptotal=Plow+Phigh low high 总 low high

[0060] The low-frequency power allocated to the battery is P​​​​​​​bat_low ; Need to meet:

[0061] Where i is the sum of the optimal efficiency point power of all working power units; if P bat_low > 0, the battery discharges to supplement the low-frequency power gap; if P bat_low < 0, the battery charges to absorb the excess power of the power unit. Even if the battery SOC is close to the boundary, the power unit must be allowed to work at the optimal efficiency point, and the battery only compensates within the allowed range. By comparing the efficiency of the two power distribution schemes (such as whether to call additional power units), the scheme that minimizes the battery charging and discharging power and maximizes the efficiency of the power unit is selected.

[0062] The power unit optimal power distribution strategy and the method for obtaining the battery compensation power are as follows:

[0063] Based on the temperature change compensation strategy, and combined with the distribution power of the power unit, the temperature change and the altitude, the power unit optimal power distribution strategy and the battery compensation power are obtained respectively;

[0064] As shown in Figure 2 The power unit optimal power distribution strategy is used to optimize the rolling matrix of the working time of the power unit to obtain the optimal distribution power of each power unit, including the following steps:

[0065] Obtain the total demand power of the power system;

[0066] Based on the conventional characteristics of the power system, the best working efficiency is calibrated, and two different rules are developed to obtain two power distribution matrices respectively;

[0067] Compare the efficiency of the two power unit working time rolling matrices to obtain the number of power units in the working state;

[0068] Initialize the pre-optimization parameters to clearly define all optimization directions, update the iteration factor for the optimization direction, record the average efficiency of each optimization direction, and determine whether each optimization direction reaches the iteration number; if so, compare the optimization effect of each optimization direction to determine the optimal optimization path, and if not, constantly update the iteration factor for the optimization direction until the iteration number is reached to obtain the best optimization path;

[0069] Initialize the main optimization parameters, update the iteration factor and distribution power of each power unit, and determine whether the update process reaches the iteration number; if so, output the best power distribution strategy, and if not, constantly update the iteration factor and distribution power of each power unit until the iteration number is reached to obtain the optimal power distribution power of each power unit.

[0070] The following detailed description is made to the automatic high-low frequency power matching algorithm:

[0071] Wavelet transform can extract signal information in time domain and frequency domain, and can be decomposed according to different phases and scales. The wavelet with localization characteristics is particularly suitable for analyzing transient signals. Since the total demand power is a discrete signal, the discrete wavelet transform is used to decompose the discrete demand power signal (one-dimensional waveform) into different decomposition layers. Therefore, the wavelet transform can decompose the discrete demand power signal into high-frequency transient components and low-frequency components. Therefore, the one-dimensional load demand power signal is decomposed into signals of different frequency bands by using wavelet transform, and the transient demand power is extracted and allocated to the super capacitor, and the signal decomposition and reconstruction expression is:

[0072]

[0073]

[0074] In the formula, x(t) is the original signal of demand power; W is the wavelet coefficient; t is the time of wavelet transform; a is the scale factor; y is the translation factor; u=k2 j ; j, k represent the power value, both are integers; ψ is the mother function, which is a Haar wavelet function.

[0075] The mother function is a Haar wavelet function, in which the wavelet transform is equal to the inverse transform, which can greatly simplify the wavelet algorithm and improve the efficiency of code execution. The expression of the Haar wavelet function is:

[0076]

[0077] Two-channel filters are used to design the decomposition and reconstruction filter banks based on Haar wavelet transform. Through the z-transform of the low-pass filter H0(z) and the high-pass filter H1(z), the high-frequency components and the low-frequency components in the original signal are extracted respectively. The process of decomposition and reconstruction of the signal x(t) by using 5-order Haar wavelet transform is shown in Figure 3 .

[0078] It should be noted that after the processing of the automatic high-low frequency power matching algorithm, the demand power high-frequency component and the demand power low-frequency component are obtained respectively. The power distribution is carried out for the purpose of optimizing the comprehensive efficiency of the distributed modular hybrid power system, and the optimal power distribution strategy of the power unit is invented. The strategy mainly includes two parts of algorithm, one part is the unit algorithm, and the other part is the optimization algorithm.

[0079] For the sub-unit algorithm, based on the power-efficiency relationship of the power unit, it is concluded that the more the number of power units called in the power distribution, the lower the efficiency of the power unit will be, because when the power distribution value decreases, the power unit will constantly approach the low-efficiency area due to the low power value. Therefore, in order to avoid the power unit approaching the low-efficiency area, when the power distribution is performed, it is necessary to ensure that the power is provided by using as few power units as possible.

[0080] The sub-unit algorithm can ensure that the power unit works in the high-efficiency area as much as possible, and then further optimization needs to be performed in the high-efficiency area to further improve the working efficiency of the power unit. According to the first power distribution result obtained through the sub-unit algorithm, the optimal efficiency point of the power unit output power at 25 KW can be retained, and the next part of the iterative optimization can be performed on the non-optimal efficiency point.

[0081] The first step of the iterative algorithm high-efficiency area optimization is to find the power unit that is not allocated to the optimal efficiency point after the sub-unit algorithm, and the initial optimization step is obtained by calculation. It mainly includes two cases, the first case is that when the target power is greater than the product of the optimal efficiency point and the number of power units allocated, the initial optimization step is the maximum output power of the power unit minus the value that does not reach the optimal efficiency point; the second case is that the initial optimization step is 2 by default, and the expression is as follows.

[0082]

[0083] In the formula, p is the target demand power; max(P) represents the maximum output power of the power unit, which is 30 KW in the present application; u represents the number of power units used; opt represents the power corresponding to the optimal efficiency point of the power unit, which is 25 KW in the present application; x mn represents the allocated power that does not reach the optimal efficiency point; s represents the initial optimization step.

[0084] After obtaining the initial optimization step, optimization is performed, and the initial optimization step is assigned to one of the values that do not reach the optimal efficiency point, which is defined as the guide seed; the remaining values that do not reach the optimal efficiency point are generated in the form of a random number (the random number is less than the initial optimization step). Taking the value assigned by the initial optimization step as the core, the efficiency of the power module is compared after each iteration, i.e. optimization according to the step. If the efficiency increases, the optimization continues along the step; if the efficiency decreases, the optimization step is reduced by a step correction factor and the optimization is performed in the opposite direction. Until the iteration step termination value is reached, the optimization ends.

[0085]

[0086] In the formula, S mnrepresents the optimization step matrix; δ represents the step correction factor; r mn (m, n cannot be 1 at the same time) is a random factor distributed in [0, 1] as an absolute value. There are two restrictions in the formula, that is, the sum of the optimization step matrix is zero, so that the sum of the energy matrix after each iteration is equal to the demand power value, and the maximum value of the optimization step matrix is the guide seed value after iteration.

[0087]

[0088] In the formula, X mn represents the energy matrix after each iteration.

[0089] The mentioned automatic high-low frequency power matching algorithm and the optimal power distribution strategy of the power unit are combined to build a joint control model. The equivalent fuel minimum consumption theory which is well-known in the field of energy management strategy is adopted, and after a plurality of test groups, the relationship between the equivalent factor and the battery SOC is fitted.

[0090]

[0091] In the formula, λ cha , λ dis respectively represent the charge and discharge equivalent factors; 1.7 and 1.8 simulate the charge equivalent factors when the SOC seriously deviates from the target value; 1.2 and 1.3 simulate the discharge equivalent factors when the SOC seriously deviates from the target value.

[0092] The power corresponding to the optimal efficiency point of the engine MAP diagram in the power unit is 25KW, so the energy storage unit can maintain the optimal efficiency point of the power module as much as possible. On this basis, the power module collaborative control strategy based on the optimal efficiency is proposed, and since the working efficiency of the battery is higher than that of the power unit when the SOC is reasonable, the battery is used as an auxiliary energy device in the strategy. The control strategy can calculate the corresponding power of the two power unit optimal efficiency points closest to the demand power, and the battery corresponds to a charging power and a discharging power at this time. By comparing the total efficiency of the power system of the two schemes, the scheme with high efficiency is selected for control.

[0093] The above description of the disclosed embodiments enables one of ordinary skill in the art to make or use the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A rule-based distributed modular hybrid power system energy management method, characterized in that: The following steps are involved: Obtain the target power and target torque of the power system under different working conditions; Fuzzy control is used to coordinately distribute the target power and target torque of the power system, and the torque and power required by each power unit and hydraulic motor in the power system are obtained; An automated high- and low-frequency power matching algorithm is used to decompose the total power demand of each power unit and hydraulic motor in the power system into low-frequency power and high-frequency power, and distribute it to the battery, each power unit, and supercapacitor, obtaining the supercapacitor charging and discharging power, battery charging and discharging power, power unit distribution power, and battery compensation power, respectively. The optimal power distribution strategy of the power unit is obtained by comprehensively considering the influence of environmental change factors on the power distribution of the power unit. Based on the optimal power distribution strategy of the power unit, the rolling matrix of the power unit working time is optimized to obtain the optimal distribution power of each power unit, so as to realize adaptive adjustment of the energy output of the power unit under environmental changes.

2. The rule-based distributed modular hybrid power system energy management method according to claim 1, characterized in that: The total power requirement of each power unit and hydraulic motor in the power system is decomposed into low-frequency power and high-frequency power, and distributed to the battery, each power unit and supercapacitor, specifically: Based on the most efficient low-frequency power allocation strategy, low-frequency power is allocated to the battery and each power unit; Based on the high-frequency power rule allocation strategy, high-frequency power is allocated to the supercapacitor.

3. The rule-based distributed modular hybrid power system energy management method according to claim 1, characterized in that: The method for obtaining the supercapacitor charging and discharging power is as follows: Based on the supercapacitor charging and discharging strategy, the supercapacitor is energy managed to obtain the capacitor charging and discharging power.

4. The rule-based distributed modular hybrid power system energy management method according to claim 1, characterized in that: The method for obtaining the battery charge and discharge power is specifically as follows: Based on the power battery charging and discharging strategy, and combined with the energy storage boundary, temperature, charging and discharging instructions and the battery's allocated power, the battery charging and discharging power is obtained.

5. The rule-based distributed modular hybrid power system energy management method according to claim 1, characterized in that: The optimal power distribution strategy for the power unit and the method for obtaining the battery compensation power are specifically as follows: Based on the temperature change compensation strategy and combined with the power distribution of the power unit, temperature change and altitude, the optimal power distribution strategy of the power unit and the battery compensation power are obtained respectively.

6. The rule-based distributed modular hybrid power system energy management method according to claim 1, characterized in that: The method of optimizing the rolling matrix of the power unit working time based on the power unit optimal power allocation strategy to obtain the optimal allocated power of each power unit includes the following steps: Obtain the total required power of the power system; The optimal working efficiency is calibrated based on the conventional characteristics of the power system, and two different rules are formulated to obtain two power distribution matrices respectively; Compare the efficiency of the rolling matrix of the working time of the two power units to obtain the number of power units in working state; Initialize the pre-optimization parameters, identify all optimization directions, update the iteration factor for each optimization direction, record the average efficiency of each optimization direction, and determine whether each optimization direction has reached the number of iterations. If so, compare the optimization effects of each optimization direction to determine the optimal optimization path. If not, continuously update the iteration factor for each optimization direction until the number of iterations is reached and the optimal optimization path is obtained. Initialize the main optimization parameters, update the iteration factor and allocated power of each power unit, and determine whether the update process reaches the number of iterations; if so, output the optimal power allocation strategy; if not, continuously update the iteration factor and allocated power of each power unit until the number of iterations is reached, and obtain the optimal power allocation power of each power unit.

Images