Orderly charging method, device, computer-readable storage medium, and computer equipment

Abstract

The present invention discloses a method, apparatus, computer-readable storage medium, and computer equipment for orderly charging. The method comprises: determining a predicted charging duration for a target device based on the type of the target device; obtaining the charging power of the target device's battery as it varies with the charging duration and state of charge; determining an initial charging strategy based on the charging power and the predicted charging duration; and iteratively optimizing the initial charging strategy within a preset number of iterations, with the goal of minimizing the sum of operating cost and charging dissatisfaction, based on a preset set of constraints and an adaptive inertia weight, to obtain a target charging strategy. The present invention addresses the technical issues of disordered charging behavior, poor charging performance, and increased battery life loss caused by charging behavior when charging electric agricultural implements.

Description

Technical Field

[0001] The present invention relates to the field of electric energy, and in particular to an orderly charging method, device, computer-readable storage medium and computer equipment. Background Art

[0002] As the modernization of agricultural facilities continues to advance, centralized, refined, and intelligent agricultural park planting models will gradually replace traditional, inefficient, and extensive planting methods. At the same time, mobile electric tools, offering numerous advantages such as environmental friendliness, portability, and affordability, are increasingly replacing traditional fuel- and manual-powered tools. These include compact and convenient three-wheeled electric vehicles, energy-saving and environmentally friendly electric pickup trucks, electric forklifts, small electric bulldozers, efficient and portable electric sprayers, and electric irrigation machines. Mobile agricultural machinery and equipment, including high-standard electrified farmland, are gaining widespread adoption, in line with the nation's green energy transition. However, the widespread use of electric agricultural tools also presents challenges in effectively managing their charging.

[0003] Therefore, in the related art, there are technical problems such as disordered charging behavior, poor charging effect, and increased battery life loss caused by charging behavior when charging electric agricultural tools.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] Embodiments of the present invention provide an orderly charging method, apparatus, computer-readable storage medium, and computer equipment to at least solve the technical problems of disordered charging behavior, poor charging effect, and increased battery life loss caused by charging behavior when charging electric agricultural implements.

[0006] According to one aspect of an embodiment of the present invention, an orderly charging method is provided, comprising: determining a predicted charging time of a target device based on the type of the target device; obtaining the charging power of a battery of the target device as it changes with the charging time and the state of charge; determining an initial charging strategy based on the charging power and the predicted charging time; and iteratively optimizing the initial charging strategy according to an adaptive inertia weight, based on a preset set of constraints and with the goal of minimizing the sum of the operating cost and the degree of charging dissatisfaction, to obtain a target charging strategy within a preset number of iterations.

[0007] Optionally, based on the type of the target device, the predicted charging time of the target device is determined, including: based on the type of the target device, obtaining actual charging data within a third predetermined time range before the charging date, wherein the actual charging data includes actual charging time and actual charging frequency; determining the actual charging probability distribution corresponding to the actual charging data; performing Gaussian distribution fitting on the actual charging probability distribution to obtain a fitting result; and generating a predicted charging time based on the fitting result.

[0008] Optionally, obtaining the charging power of the battery of the target device that changes with the charging time and state of charge includes: obtaining the current state of charge of the battery; when the current state of charge is less than a preset state of charge, determining the charging power to be a first power that changes with the charging time in a constant current charging mode; when the current state of charge is greater than or equal to a preset state of charge, determining the charging power to be a second power that changes with the charging time in a constant voltage charging mode.

[0009] Optionally, the preset constraint condition group includes at least one of the following: charging power and state of charge constraint conditions, charging time constraint conditions, and battery charging and discharging life cost constraint conditions.

[0010] Optionally, within a preset number of iterations, with the goal of minimizing the sum of the operating cost and the charging dissatisfaction, based on a preset constraint condition group and in accordance with the adaptive inertia weight, the initial charging strategy is iteratively optimized to obtain a target charging strategy, including: determining the operating cost and charging dissatisfaction corresponding to the initial charging strategy; screening the initial charging strategy based on the operating cost and charging dissatisfaction corresponding to the initial charging strategy to obtain a first candidate charging strategy; subjecting the preset constraint condition group to satisfaction, iteratively optimizing the first candidate charging strategy in accordance with a preset update method and adaptive inertia weight within a preset number of iterations with the goal of minimizing the sum of the operating cost and charging dissatisfaction to obtain a second candidate charging strategy; determining an optimization evaluation factor for the second candidate charging strategy; and determining the target charging strategy based on the second candidate charging strategy and the optimization evaluation factor.

[0011] Optionally, the method for determining the operating cost and the degree of charging dissatisfaction includes: determining the theoretical charging time of the target device based on the type of the target device; predicting the photovoltaic power generation power that changes with time on the charging day based on the historical light data within the first predetermined time range before the charging day; predicting the basic load that changes with time on the charging day based on the historical load curve within the second predetermined time range before the charging day; determining the equivalent cycle life of the battery in this charging based on the charge and discharge depth of the battery; determining the battery life loss cost based on the equivalent cycle life; determining the operating cost and charging dissatisfaction of the initial charging strategy based on the photovoltaic power generation power, basic load, theoretical charging time, predicted charging time, and charging power.

[0012] Optionally, the value of the adaptive inertia weight is determined by the current iteration number in the iterative optimization process.

[0013] According to another aspect of an embodiment of the present invention, an orderly charging device is also provided, including: a first determination module, used to determine the theoretical charging time and predicted charging time of a target device based on the type of the target device; an acquisition module, used to obtain the charging power of the battery of the target device as the charging time and the state of charge change; a second determination module, used to determine an initial charging strategy based on the charging power and the predicted charging time; an optimization module, used to iteratively optimize the initial charging strategy based on a preset constraint group and an adaptive inertia weight within a preset number of iterations, with the goal of minimizing the sum of the operating cost and the degree of charging dissatisfaction, to obtain a target charging strategy.

[0014] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is further provided, the computer-readable storage medium including a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned orderly charging methods.

[0015] According to another aspect of an embodiment of the present invention, a computer device is provided, comprising: a memory and a processor, wherein the memory stores a computer program; and the processor is configured to execute the computer program stored in the memory, wherein when the computer program is executed, the processor executes any one of the above-mentioned orderly charging methods.

[0016] In an embodiment of the present invention, a method is adopted in which new energy consumption, user charging behavior, peak-valley electricity price cost, battery life loss and charging efficiency characteristics are comprehensively considered when formulating a charging strategy for electric agricultural implements. The theoretical charging time and predicted charging time of the target device are determined according to the type of the target device, wherein the theoretical charging time is the shortest time that the target device can theoretically complete charging, and the predicted charging time is the predicted time that the target device is allowed to charge. The charging power used when charging the battery of the target device is determined by the charging time and charge state of the battery. Thereafter, an initial charging strategy is generated based on the determined dynamically changing charging power and the predicted charging time, wherein the initial charging strategy is a charging power composition scheme, including the charging power value corresponding to each moment within the predicted charging time, that is, the charging power value at each moment. At what charging power should the battery be charged at all times? On the basis of the initial charging strategy, the embodiment of the present invention uses iterative calculation to, under the premise of satisfying a preset constraint group, take the sum of the operating cost and the charging dissatisfaction corresponding to the charging strategy as the minimum, and optimize the initial charging strategy according to the adaptive inertia weight to obtain a target charging strategy with better charging effect, thereby achieving the purpose of orderly charging of the target device based on the target charging strategy, and determining a charging solution that is in line with the maximization of charging satisfaction and charging benefits, on a long time scale, and in a reasonable and orderly manner, thereby achieving the technical effect of optimizing the charging management of electric agricultural implements to support efficient and reliable agricultural production, thereby solving the technical problems of disordered charging behavior, poor charging effect, and increased battery life loss due to charging behavior when charging electric agricultural implements. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0018] Figure 1 is a flow chart of an orderly charging method provided according to an embodiment of the present invention;

[0019] Figure 2 This is a flow chart of a method for orderly charging of mobile agricultural facilities based on power usage behavior and battery characteristics according to an optional embodiment of the present invention;

[0020] Figure 3 is a schematic diagram of historical data provided according to an optional embodiment of the present invention;

[0021] Figure 4 is a schematic diagram of an optimal particle fit curve provided according to an optional embodiment of the present invention;

[0022] Figure 5is a power stacking diagram of an optimal ordered charging solution obtained after convergence of an algorithm provided in an optional embodiment of the present invention;

[0023] Figure 6 4 is a structural block diagram of an orderly charging device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] In the existing technology, the charging solution for electric agricultural tools cannot reasonably allocate charging ports and charging time, resulting in difficulty in charging agricultural tools and insufficient battery power, affecting normal agricultural production and farmer satisfaction. It also cannot carefully consider the difference in peak and valley electricity prices, which further increases charging costs. In addition, the existing charging solution cannot effectively combine the level of new energy power generation such as photovoltaics and the energy consumption level that meets the requirements of basic agriculture to reduce the new energy absorption rate. At the same time, it is also impossible to effectively track the state of charge. High-frequency charging and low-charging depth charging behavior will also lead to increased battery life loss.

[0027] To address the above issues, the existing technology has yet to propose an efficient and orderly charging strategy for electric agricultural implements. While there has been research focused solely on the orderly charging of electric vehicles, current solutions for these still lack comprehensive considerations. For example, charging duration is not considered as a dynamic factor; typical photovoltaic power is derived solely through statistical methods, resulting in poor real-time predictions; electricity prices are not considered; battery charging power characteristics and lifecycle cost characteristics are not considered. Furthermore, there are significant differences between the orderly charging of electric vehicles and electric agricultural implements, making it difficult to directly reference the charging strategy for electric vehicles. For example, the battery capacity and charge / discharge power of different agricultural implements vary significantly, leading to significant differences in charging frequency. Furthermore, the charging time of electric agricultural implements is more difficult to predict than that of electric vehicles. Therefore, the above issues in the existing technology remain unresolved.

[0028] In order to solve the above technical problems, the present invention provides an embodiment of an orderly charging method.

[0029] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0030] Figure 1 is a flow chart of an orderly charging method provided according to an embodiment of the present invention. Figure 1 As shown, the method includes the following steps:

[0031] Step S102, determining a theoretical charging time and a predicted charging time of the target device based on the type of the target device;

[0032] Step S104, obtaining the charging power of the battery of the target device as it changes with the charging time and state of charge;

[0033] Step S106, determining an initial charging strategy based on the charging power and the predicted charging time;

[0034] Step S108 , within a preset number of iterations, with the goal of minimizing the sum of the operating cost and the charging dissatisfaction level, based on a preset constraint condition group and in accordance with the adaptive inertia weight, iteratively optimize the initial charging strategy to obtain a target charging strategy.

[0035] Through the above steps, the method of comprehensively considering the consumption of new energy, user charging behavior, peak and valley electricity price cost, battery life loss and charging efficiency characteristics when formulating the charging strategy of electric agricultural implements is adopted. According to the type of target device, the theoretical charging time and predicted charging time of the target device are determined, wherein the theoretical charging time is the shortest time that the target device can theoretically complete charging, and the predicted charging time is the predicted time that the target device is allowed to charge. The charging power used when charging the battery of the target device is determined by the charging time and charge state of the battery. Afterwards, the initial charging strategy is generated based on the determined dynamically changing charging power and the predicted charging time, wherein the initial charging strategy is a charging power composition scheme, including the charging power value corresponding to each moment within the predicted charging time, that is, the charging power value at each time. At this moment, the battery should be charged at what charging power. On the basis of the initial charging strategy, the embodiment of the present invention optimizes the initial charging strategy according to the adaptive inertia weight through iterative calculation, subject to the satisfaction of a preset constraint group, with the goal of minimizing the sum of the operating cost and the charging dissatisfaction corresponding to the charging strategy, so as to obtain a target charging strategy with better charging effect, thereby achieving the purpose of orderly charging of the target device based on the target charging strategy, and determining a charging solution that is in line with the maximization of charging satisfaction and charging benefits, on a long time scale, and in a reasonable and orderly manner, thereby achieving the technical effect of optimizing the charging management of electric agricultural implements to support efficient and reliable agricultural production, thereby solving the technical problems of disordered charging behavior, poor charging effect, and increased battery life loss due to charging behavior when charging electric agricultural implements.

[0036] Wherein, the above-mentioned target equipment is an electric agricultural implement.

[0037] As an optional embodiment, based on the type of the target device, the predicted charging time of the target device is determined, including: based on the type of the target device, obtaining actual charging data within a third predetermined time range before the charging date, wherein the actual charging data includes the actual charging time and the actual charging frequency; determining the actual charging probability distribution corresponding to the actual charging data; performing Gaussian distribution fitting on the actual charging probability distribution to obtain a fitting result; and generating a predicted charging time based on the fitting result.

[0038] Because farmers have their own charging behaviors during the actual production process, for example, they start charging in the evening and stop charging the next morning, etc., that is, the actual allowed charging time may be longer than the theoretical fastest charging time. Therefore, this embodiment predicts the actual allowed charging time of the target device. Different device types may have different corresponding allowed charging times. Therefore, in this embodiment, the actual charging data within the third predetermined time range before the charging day is first obtained according to the type of target device, and the actual charging probability distribution of the target device of this type is determined as the data basis. The actual charging probability distribution is then fitted with a Gaussian distribution, and finally a predicted charging time is randomly generated based on the fitting result. Among them, the actual charging time is the time from connecting to the charging pile to disconnecting from the charging pile.

[0039] It should be noted that when determining the actual charging probability distribution corresponding to the actual charging data, the Monte Carlo simulation algorithm can be used. The Monte Carlo simulation algorithm is a statistical modeling method based on probability theory. It selects a probability density model through the statistical frequency of historical data and uses parameter fitting to obtain the final probability density function, which can then simulate the charging time of farmers.

[0040] The third predetermined time range may be set according to the application scenario or prediction accuracy requirements, for example, it may be the target device within the past 30 days, and so on.

[0041] In addition, the Gaussian distribution used in the embodiment of the present invention has a fitting function formula as follows:

[0042]

[0043] Where x is the charging duration, μ and λ are the inverse Gaussian distribution parameters.

[0044] At the same time, the embodiment of the present invention adopts the incremental heuristic method to calculate the values ​​of parameters μ and δ respectively, setting the initial values ​​of μ and δ to 0.1, the numerical range to [0.1, 100], and the parameter growth interval to 0.1. The specific process is to keep δ unchanged each time and increase the parameter u by 0.1; then calculate the function value F under the parameters μ and δ; then, after u increases to 100, the parameter δ+0.01 is added, and u is calculated again from 0.01; until the parameters μ and δ are both 100. The values ​​of μ and δ with the smallest mean square error between the F sequence and the actual data distribution sequence are selected, and this value is the final Gaussian distribution parameter. At this point, the parameters μ and δ have been determined, and the final charging time probability distribution function is obtained. According to the F(x, μ, δ) distribution, the predicted charging time of the target device can be randomly generated.

[0045] As an optional embodiment, obtaining the charging power of the battery of the target device that changes with the charging time and charge state includes: obtaining the current charge state of the battery; when the current charge state is less than the preset charge state, determining the charging power to be the first power that changes with the charging time in the constant current charging mode; when the current charge state is greater than or equal to the preset charge state, determining the charging power to be the second power that changes with the charging time in the constant voltage charging mode.

[0046] On the one hand, since different types of electric agricultural implements have large differences in battery capacity, rated charging power, and charging efficiency, the charging time of different devices varies greatly, and it is necessary to combine the charging power characteristics of the device battery when determining the charging strategy. On the other hand, the electric agricultural implements of the embodiments of the present invention are mainly based on lithium batteries, wherein the charging power of the battery is actually not constant, and is mainly divided into two stages: constant current and constant voltage. Therefore, the embodiments of the present invention divide the charging power used when charging the target device according to the state of charge of the battery, and divide it into constant current charging mode and constant voltage charging mode accordingly. By obtaining the current state of charge of the battery of the target device in real time during charging, and adopting the charging power in constant current charging mode or constant voltage charging mode according to the current state of charge of the battery, a more efficient charging effect with less battery loss can be achieved. For example, at the beginning of charging, the battery's state of charge is low, and the battery's equivalent internal resistance is small and relatively stable. If constant voltage is used, a large charging current will easily be generated. At this time, the constant current charging mode should be used; when the state of charge reaches the preset state of charge, the battery's equivalent internal resistance increases rapidly. If the constant current mode is continued to be used, the applied voltage requirement increases, which is difficult to meet and causes irreversible damage to the battery. At this time, the constant voltage mode should be used.

[0047] Wherein, corresponding to different charging modes, the first power or the second power can be determined by the following method:

[0048]

[0049] Where, P REV_max is the rated maximum charging power of the battery; when the state of charge is in the range [0, SOC th ], it is in constant current mode and is in the range [SOC th , 1], it is the constant voltage mode; ε(t) is the dynamic charging parameter in the constant voltage stage, and in the embodiment of the present invention, ε(t)=ln0.85; T th is the time corresponding to the state of charge threshold SOCth. In the embodiment of the present invention, the value of SOCth is 0.8.

[0050] As an optional embodiment, the preset constraint condition group includes at least one of the following: charging power and state of charge constraint conditions, charging time constraint conditions, and battery charging and discharging life cost constraint conditions.

[0051] To ensure that the target charging strategy is efficient and feasible, this embodiment comprehensively considers aspects such as new energy consumption, user charging behavior, peak and valley electricity price costs, battery life loss, and charging efficiency characteristics to determine the above-mentioned preset constraint condition group.

[0052] The charging power and state of charge constraints are as follows:

[0053]

[0054] Among them, SOCi is the current state of charge of the battery, SOCend is the state of charge of the battery after charging, SOC0 is the state of charge of the battery during initial charging, PREV is the charging power, and PREV_MAX is the rated maximum charging power of the battery.

[0055] The charging time constraints are as follows:

[0056] 0≤T L_i ≤T R_i

[0057] The above constraints represent the actual charging time T of device i L_i The predicted charging time T based on Monte Carlo simulation should be R_i Complete the charging process.

[0058] The life cycle cost constraints of battery charging and discharging are as follows:

[0059]

[0060] Where, L deep (t) is the depth of charge and discharge; n loss N is the equivalent charge and discharge cycle number of the battery; c It is the designed cycle life of the battery tested by the standard, in times; N loss It is the total number of charging cycles that the battery undergoes when the battery capacity decays by 20%, that is, the actual battery cycle life, in times.

[0061] As an optional embodiment, within a preset number of iterations, with the goal of minimizing the sum of the operating cost and the charging dissatisfaction, the initial charging strategy is iteratively optimized based on a preset constraint condition group and in accordance with an adaptive inertia weight to obtain a target charging strategy, including: determining the operating cost and charging dissatisfaction corresponding to the initial charging strategy; screening the initial charging strategy based on the operating cost and charging dissatisfaction corresponding to the initial charging strategy to obtain a first candidate charging strategy; under the condition of satisfying the preset constraint condition group, iteratively optimizing the first candidate charging strategy in accordance with a preset update method and adaptive inertia weight within a preset number of iterations with the goal of minimizing the sum of the operating cost and charging dissatisfaction to obtain a second candidate charging strategy; determining an optimization evaluation factor for the second candidate charging strategy; and determining the target charging strategy based on the second candidate charging strategy and the optimization evaluation factor.

[0062] In this embodiment, each charging strategy is essentially a charging power composition scheme, including the corresponding charging power value at each moment within the predicted charging duration, that is, indicating the charging power at which the battery should be charged at each moment. Therefore, after determining the charging mode and charging power based on the battery's state of charge and the predicted charging duration, a certain number of charging schemes, namely initial charging strategies, can be randomly generated. The operating costs and charging dissatisfaction levels of these charging schemes are then calculated to determine the optimal and worst-case solutions. Based on the distances between the operating costs and charging dissatisfaction levels corresponding to each charging scheme and the optimal and worst-case solutions, namely the positive and negative ideal distances, these schemes are then screened based on these positive and negative ideal distances to obtain a first set of candidate charging strategies.

[0063] After obtaining the first candidate charging strategy, a preset update method (for example, a population update formula) and an adaptive inertia weight can be used to update multiple charging schemes that serve as the first candidate charging strategy. At this time, one iteration is completed, and then the operating cost and charging dissatisfaction of the first candidate charging strategy are calculated again. Similar to the above method for determining the first candidate charging strategy, the optimal solution of the second calculation is obtained. If the optimal solution is better than the first, the optimal solution of the first iteration is replaced. If no better solution is obtained, the original solution remains unchanged, and the above iterative steps are repeated until the preset number of iterations is reached, then the calculation is stopped, and the final optimization solution, i.e., the second candidate charging strategy, and the operating cost and charging dissatisfaction corresponding to its optimization solution are output.

[0064] The preset update method may be as follows:

[0065]

[0066] Where, X iis the charging power composition scheme, s is the number of iterations of the algorithm; w is the adaptive inertia weight; R1 and R2 are random coefficients with a value range of [0,1]; X Pbest and X Gbest are the individual optimal solution and the global optimal solution respectively; C1 and C2 are acceleration factors.

[0067] Then, the optimization evaluation factor is determined based on the operating cost and charging dissatisfaction corresponding to the second candidate charging strategy. Since the objective function in this embodiment includes two sub-goals, namely the target cost and the charging dissatisfaction, and the two sub-goals may have their own optimal charging solutions, and this embodiment requires the sum of the two to be minimized as the final objective function, therefore, the optimization evaluation factor is introduced in this embodiment. The optimization evaluation factor is used to characterize the degree to which the two sub-goals of target cost and charging dissatisfaction are close to the optimal solution at the same time. The smaller the value, the closer it is, that is, the better the charging solution.

[0068] Among them, the method for determining the optimization evaluation factor is as follows:

[0069]

[0070] Where C j To optimize the evaluation factors, are the positive and negative ideal distances respectively.

[0071] in, The calculation method is as follows:

[0072]

[0073] Where f′ cost_j (s) represents the normalized operating cost of charging scheme j at s iterations, f sati_j (s) represents the normalized charging dissatisfaction degree of charging scheme j at the sth iteration.

[0074] It should be noted that in this embodiment, to improve the optimization efficiency of the initial charging strategy, the TOPSIS ranking method can be used for iterative optimization. TOPSIS is a typical multi-objective decision-making method that is simple, flexible, and efficient. It can effectively eliminate dimension and intuitively select the optimal charging solution based on its relative distance from the "ideal solution."

[0075] As an optional embodiment, the method for determining the operating cost and the degree of charging dissatisfaction includes: predicting the photovoltaic power generation power that changes with time on the charging day based on the historical sunlight data within a first predetermined time range before the charging day; predicting the basic load that changes with time on the charging day based on the historical load curve within a second predetermined time range before the charging day; determining the theoretical charging time of the target device based on the type of the target device; determining the equivalent cycle life of the battery in this charging based on the charge and discharge depth of the battery; determining the battery life loss cost based on the equivalent cycle life; and determining the operating cost and charging dissatisfaction of the initial charging strategy based on the photovoltaic power generation power, basic load, theoretical charging time, predicted charging time, charging power and battery life loss cost.

[0076] For photovoltaic power generation, this embodiment can use a BP neural network simulation method. The neural network model is trained based on historical light data within a predetermined time range before the charging date as sample data. The trained model is then used to predict the photovoltaic power generation on the charging date. For example, the 24-hour light data for the current day can be predicted based on the historical light data from the last 30 days, with a data collection period of 1 hour. The 30 days of historical light data are used as sample data, with a data volume of 24×30. Each time the neural network model is trained using this sample data, the input and output are both 24×1 historical light data. The output is corrected using the actual light data of the following day, and feedback is used to correct the neural network model. When the error is less than the target accuracy, the simulation is terminated to obtain the final prediction model. The prediction model is then used to obtain the light prediction curve for the next day, i.e., the photovoltaic power generation over time during the charging day in this embodiment.

[0077] Regarding the base load, different farmers may have different electricity consumption habits and a certain pattern of electricity consumption in the short term. Therefore, this embodiment adopts a method of approximating the daily load average to determine the base load during the charging day based on the corresponding average value of each moment in the historical load curve. For example, the historical load curve of the last 7 days can be selected, and the average load value can be obtained at each hour to approximate the load at the same moment of the next day. The calculation formula is as follows:

[0078]

[0079] Where, t is a certain moment in 24 hours; M is the number of recent historical load days, which is 7 in this embodiment; m is an integer variable in the range of [1, M]; P load (t) is the uncontrollable typical load at time t; P load_M (t,m) is the load power at the mth day and the tth moment in the history.

[0080] Regarding the battery life loss cost, this embodiment first determines the battery cycle charge life loss relationship between the battery charge and discharge depth and the discharge interval, and the formula is as follows:

[0081]

[0082] Where N loss The total number of charging cycles the battery undergoes when the battery capacity decays by 20%, that is, the actual battery cycle life, in times; N c is the designed cycle life of the battery tested by the standard, in times; β1 and β2 are the test data fitting parameters; L deep 、 are respectively the depth of discharge and the standard depth of discharge, where L deep It is the difference between the state of charge before and after charging. The value is 0.8; SOC ref It is the standard value of state of charge, which is 0.8.

[0083] The discharge depth L deep The calculation formula is as follows:

[0084] L deep =SOC end -SOC0

[0085] The above formula converts the actual number of battery cycles into the standard number of cycles. Therefore, the depth of charge and discharge of each charge can be converted into the number of cycle life according to the battery cycle charge life loss relationship formula. The formula is as follows:

[0086]

[0087] Where n loss The number of cycles equivalent to this charge; SOC dis_ref , SOC ref All of them are parameters less than 1. Therefore, from the discharge depth L deep From the calculation formula, we can see that the higher the SOC at the start of discharge and the greater the depth of discharge, the greater the battery life loss.

[0088] The formula for calculating the battery life loss cost is as follows:

[0089]

[0090] Where S bat Battery investment and construction costs; C bat The equivalent loss cost of the device under current charging.

[0091] In summary, the following methods can be used to determine the operating cost and charging dissatisfaction:

[0092]

[0093]

[0094] Where, F c (t) is the function value optimized for real-time objectives; f cost (t) is the farmer’s operating cost function; f sati (t) is the user dissatisfaction function of farmers; C electric (t) is the time-of-use electricity price; P load (t) is the uncontrollable basic load; P PV (t) is the real-time photovoltaic power generation; T L_i is the actual charging time of the i-th electric agricultural implement; T R_i represents the predicted charging time of the i-th electric agricultural implement; P REV_i (t) is the charging power of the i-th electric agricultural implement at time t; C bat_i (t) is the equivalent charging life loss cost of the i-th electric agricultural implement at time t; SOC end_i (t) is the state of charge of the i-th device at time t; the function std() and the function mean() represent the functions of calculating the mean square error and the mean value respectively; ti is the charge state of the i-th device at time t

[0095] Where, T L_i / T R The smaller the value, the shorter the charging time and the lower the user dissatisfaction; 1-SOC end_i The smaller the value of (t), the closer the battery state of charge is to 1 when charging ends, and the lower the user dissatisfaction is. The smaller the values ​​of function std() and function mean() are, the lower the per-unit value of the peak-to-valley difference is for the user, the higher the power quality is, and the lower the user dissatisfaction is.

[0096] As an optional embodiment, the value of the adaptive inertia weight is determined by the current number of iterations in the iterative optimization process. This embodiment proposes an adaptive inertia weight whose value is determined by the current number of iterations as follows:

[0097]

[0098] Where w max and w m i n are the upper and lower limits of the inertia weight, which can be 0.9 and 0.5 respectively; s is the current number of iterations; γ is the empirical adjustment coefficient, which ranges from [1,30] and is 3 in this embodiment.

[0099] As can be seen from the above formula, the adaptive inertia weight is dynamically adjusted according to the number of iterations. In the initial optimization stage, w maintains a large value to ensure the global optimization ability of the algorithm, that is, it can quickly determine the range of the preferred charging plan. In the later stage of optimization, w is small, maintaining the algorithm's local optimization, that is, it can accurately determine the optimal charging plan.

[0100] Based on the above embodiments and optional embodiments, the present invention also proposes an optional implementation method, which is described below.

[0101] An optional embodiment of the present invention proposes an orderly charging method for mobile agricultural facilities based on power consumption behavior and battery characteristics. Figure 2 This is a flow chart of a method for orderly charging mobile agricultural facilities based on power consumption behavior and battery characteristics according to an optional embodiment of the present invention. Figure 2 As shown, the method includes the following steps: First, a photovoltaic power prediction method based on a neural network, a typical load generation method based on curve superposition, and a Monte Carlo simulation-based probability model for the charging time of electric agricultural implements are established to establish a predictive photovoltaic, load, and charging time generation method; then, a mathematical model is established between battery charging efficiency and state of charge (SOC), and a battery depth of discharge and life cost model is established for the discharge interval; third, a method for optimizing the sequential charging power of electric agricultural implements over a long time scale is established; finally, a multi-objective adaptive particle swarm algorithm based on TOPSIS sorting is used to solve the model, resulting in an optimized solution for sequential charging of agricultural facilities including agricultural photovoltaics. The following describes an optional embodiment of the present invention in detail.

[0102] (1) Agricultural photovoltaic prediction and uncontrollable typical load simulation

[0103] This optional embodiment of the present invention is primarily used in large-scale agricultural planting industrial parks with photovoltaic systems. The optimization targets the charging time and power of various agricultural tools connected to charging piles. This optional embodiment of the present invention needs to consider photovoltaic power and uncontrollable load conditions during the charging cycle, thus requiring photovoltaic power prediction methods and uncontrollable typical load simulation methods.

[0104] The photovoltaic forecast uses a BP neural network simulation method to predict the current day's 24-hour sunlight data based on the last 30 days of historical sunlight data, with a data sampling period of 1 hour. The 30 days of historical sunlight data are used as sample data, with a data volume of 24×30. The neural network model is trained using this sample data each time, with both the input and output being 24×1 of historical sunlight data. The output is corrected using the next day's actual sunlight data, and feedback is used to correct the neural network model. The simulation is terminated when the error falls below the target accuracy, resulting in the final prediction model. The prediction model is then used to generate the next day's sunlight forecast curve.

[0105] The electricity load in agricultural plantations has certain user electricity usage habits and a certain pattern of electricity consumption in the short term. An optional embodiment of the present invention uses an approximate daily load average replacement method. The historical load curve for the past seven days is selected, and the average load value is calculated at each hour to approximate the load at the same time the next day. The calculation formula is as follows:

[0106]

[0107] Where, t is a certain moment in 24 hours; M is the number of recent historical load days, which can be set to 7 in the optional embodiment of the present invention; m is an integer variable in the range of [1, M]; P load (t) is the uncontrollable typical load at time t; P load_M (t,m) is the load power at the mth day and the tth moment in the history.

[0108] (2) Electric agricultural implement charging behavior simulation and charging power model

[0109] 1) Charging time simulation based on Monte Carlo simulation

[0110] In actual production, farmers have their own charging behaviors. For example, they start charging in the evening and stop charging the next morning. The actual allowed charging time may be longer than the theoretical fastest charging time TL.

[0111] Therefore, an optional implementation of the present invention needs to simulate and evaluate the charging time of farmers for each type of equipment. The Monte Carlo simulation algorithm is a statistical modeling method based on probability theory. It selects a probability density model through the statistical frequency of historical data, and uses it to fit the parameters to obtain the final probability density function, which can then simulate and generate the charging time of farmers. Taking a certain agricultural charging device i as an example, an optional implementation of the present invention uses the actual charging time of device i in the past 30 days (that is, the time from connecting to the charging pile to disconnecting from the charging pile) data, and counts the duration and frequency of each charging, and draws a probability partition diagram. According to the probability distribution diagram, an optional implementation of the present invention uses Gaussian distribution for fitting, and its fitting function formula is as follows:

[0112]

[0113] Where x is the charging duration, μ and λ are the inverse Gaussian distribution parameters.

[0114] In an optional embodiment of the present invention, an incremental heuristic method is used to calculate the values ​​of parameters μ and δ, respectively. The initial values ​​of μ and δ are set to 0.1, the numerical range is [0.1, 100], and the parameter growth interval is 0.1. The specific process is to first keep δ unchanged each time and increase the parameter u by 0.1; then calculate the function value F under the parameters μ and δ; then, after u increases to 100, the parameter δ is increased by 0.01, and u is calculated again from 0.01 until the parameters μ and δ are both 100. The values ​​of μ and δ with the smallest mean square error between the F sequence and the actual data distribution sequence are selected, and this value is the final Gaussian distribution parameter. At this point, the parameters μ and δ have been determined, and the final charging duration probability distribution function F(x, μ, δ) is obtained. According to the F(x, μ, δ) distribution, the charging duration TR of device i can be randomly generated.

[0115] 2) Two-stage battery charging power calculation model

[0116] The battery capacity, rated charging power, and charging efficiency of electric agricultural implements in the optional embodiments of the present invention vary greatly. Therefore, the charging time of different devices varies greatly and needs to be combined with the charging power characteristics of the device battery. The charging time calculation formula is as follows:

[0117]

[0118] Where TL is the charging time; SOC end is the state of charge after charging; SOC0 is the state of charge at initial charging; Q C is the capacity of the battery; η REV (t) is the charging efficiency of the device at time t; P REV (t) is the charging power at time t;

[0119] The electric agricultural implements of the optional embodiment of the present invention are mainly based on lithium batteries, in which the charging power of the battery is not actually constant, and is mainly divided into two stages: constant current and constant voltage. In the initial charging stage, the state of charge of the lithium battery is low, and the equivalent internal resistance of the battery is small and relatively stable. If constant voltage is used, it is easy to generate a large charging current. At this time, the constant current charging mode should be used; when the state of charge reaches the threshold SOC th After that, the equivalent internal resistance of the battery increases rapidly. If the constant current mode is continued, the applied voltage requirement will increase, which is difficult to meet and will cause irreversible damage to the battery. At this time, the constant voltage mode should be adopted. An optional embodiment of the present invention establishes a real-time charging power model based on a two-stage charging mode. The formula is as follows:

[0120]

[0121] Where, P REV_max is the rated maximum charging power of the battery; when the state of charge is in the range [0, SOC th], it is in constant current mode and is in the range [SOC th , 1], it is the constant voltage mode; ε(t) is the dynamic charging parameter in the constant voltage stage, and in the optional embodiment of the present invention, ε(t)=ln0.85; T th The time corresponding to the state of charge threshold SOCth, where SOC th The value is 0.8.

[0122] 3) Lifecycle cost model for battery charging depth and range

[0123] An optional embodiment of the present invention establishes a battery cycle charge life loss model that takes into account the depth of discharge and the discharge interval. The formula is as follows:

[0124]

[0125] Where N loss The total number of charging cycles the battery undergoes when the battery capacity decays by 20%, that is, the actual battery cycle life, in times; N c is the designed cycle life of the battery tested by the standard, in times, which is taken as 1000 times in the optional embodiment of the present invention; β1 and β2 are the test data fitting parameters, which are taken as 1.98 and 2.79 respectively in the optional embodiment of the present invention; L deep 、 are respectively the depth of discharge and the standard depth of discharge, where L deep It is the difference between the state of charge before and after charging. The value is 0.8; SOC ref It is the standard value of state of charge, which is 0.8.

[0126] The discharge depth L deep The calculation formula is as follows:

[0127] L deep =SOC end -SOC0 (6)

[0128] Formula (7) converts the actual number of battery cycles into the standard number of cycles. Therefore, the depth of charge and discharge of each charge can be converted into the number of cycle life according to formula (5). The formula is as follows:

[0129]

[0130] Where n loss The number of cycles equivalent to this charge; SOC dis_ref , SOC ref All of them are parameters less than 1. Therefore, from formula (6), we can see that when the SOC at the start of discharge is higher and the depth of discharge is greater, the battery life loss is greater.

[0131] The formula for calculating the battery life loss cost is as follows:

[0132]

[0133] Where S bat Battery investment and construction costs; C bat The equivalent loss cost of the device under current charging.

[0134] (3) Long-term charging power optimization model

[0135] An optional embodiment of the present invention establishes a long-scale rolling real-time charging power optimization method for electric agricultural implements. First, based on the Monte Carlo simulation of farmer charging time evaluation method, a predicted charging time is obtained for each device. Then, within this time, a two-stage battery charging power characteristic model is considered, as well as the battery life cost loss characteristics. With the goal of minimizing comprehensive operating costs and user dissatisfaction, the charging power of the device at each moment is reasonably arranged.

[0136] 1) Objective function

[0137] An optional embodiment of the present invention arranges the charging power of each device on a long time scale. The objective function of the optimization model used is an orderly charging optimization model of user dissatisfaction and comprehensive cost, where user dissatisfaction includes three aspects: charging efficiency, the fullest state of charge at the end of charging, and the best power quality (i.e., the minimum peak-to-valley difference in load). The formula is as follows:

[0138] F c (t) = min{f cost (t)+f sati (t)} (9)

[0139] in:

[0140]

[0141]

[0142] Where, F c (t) is the function value optimized for real-time objectives; f cost (t) is the farmer’s operating cost function; f sati (t) is the user dissatisfaction function of farmers; C electric (t) is the time-of-use electricity price; P load (t) is the uncontrollable basic load; P PV (t) is the real-time photovoltaic power generation; T L_i is the actual charging time of the i-th electric agricultural implement; T R_i represents the predicted charging time of the i-th electric agricultural implement; P REV_i(t) is the charging power of the i-th electric agricultural implement at time t; C bat_i (t) is the equivalent charging life loss cost of the i-th electric agricultural implement at time t; SOC end_i (t) is the state of charge of the i-th device at time t; the function std() and the function mean() represent the functions of calculating the mean square error and the mean value respectively; ti is the charge state of the i-th device at time t

[0143] Where, T L_i / T R The smaller the value, the shorter the charging time and the lower the user dissatisfaction; 1-SOC end_i The smaller the value of (t), the closer the battery state of charge is to 1 when charging ends, and the lower the user dissatisfaction is. The smaller the values ​​of function std() and function mean() are, the lower the per-unit value of the peak-to-valley difference is for the user, the higher the power quality is, and the lower the user dissatisfaction is.

[0144] 2) Constraints

[0145] 1. Charging power and capacity constraints for agricultural tools

[0146]

[0147] 2. Charging time limit

[0148] 0≤T L_i ≤T R_i (13)

[0149] The above constraints represent the actual charging time T of device i L_i The predicted charging time T based on Monte Carlo simulation should be R_i Complete the charging process.

[0150] 3. Battery charge and discharge life cost constraints

[0151]

[0152] In the formula, the constraints consider the charge and discharge depth L deep (t), battery equivalent charge and discharge cycle number n loss Constraints.

[0153] (4) Multi-objective particle swarm optimization algorithm based on TOPSIS sorting

[0154] The sub-objectives of the multi-objective optimization model established in an optional embodiment of the present invention are user dissatisfaction and comprehensive energy cost, respectively. These two values ​​have different dimensions and cannot be directly summed. TOPSIS is a typical multi-objective decision-making method that is simple, flexible, and efficient. It can effectively eliminate dimensions and intuitively select the optimal charging solution based on its relative distance from the "ideal solution." Furthermore, an optional embodiment of the present invention first designs an adaptive inertia weight to dynamically adjust local and global traversal capabilities based on the number of iterations.

[0155] 1) Multi-objective particle swarm solution model

[0156] The optimization variable of the optional embodiment of the present invention is the charging time T of the agricultural implement connected to the charging pile. R_i The charging power at each moment within , the sequence formula of the optimized variables is as follows:

[0157] X i ={P REV_1 (t1,t2L,T R_1 ),P REV_2 (t1,t2L,T R_2 ),LP REV_i (t1,t2L,T R_i )} (15)

[0158] Where, P REV_i is the charging power of the i-th electric agricultural implement, T R_i The predicted charging time is: i (i.e., a charging power composition scheme composed of electric agricultural implements), is a particle in the particle swarm algorithm, and the population size in the particle swarm algorithm is M, the number of iterations is s, and the optimization speed of the particle swarm algorithm is V j =(v j1 ,v j2 ,...,v jM ), the iterative calculation formula of example j in the particle swarm algorithm is as follows:

[0159]

[0160] Where s is the number of iterations of the algorithm; w is the inertia weight of the particle swarm algorithm; R1 and R2 are random coefficients with a value range of [0,1]; X Pbest and X Gbest are the individual optimal solution and the global optimal solution respectively; C1 and C2 are acceleration factors, and the value of the optional embodiment of the present invention is 1.49.

[0161] 2) Adaptive inertia weight parameter optimization

[0162] In an optional embodiment of the present invention, the size of the inertia weight w determines the optimization iteration speed of the particle swarm algorithm. In an optional embodiment of the present invention, an adaptive inertia weight is designed to dynamically adjust the inertia weight w according to the number of iterations. In the initial optimization stage, w is kept at a large value to ensure the global optimization ability of the algorithm. In the later stage of optimization, w is small to maintain the local optimization of the algorithm. The formula is as follows:

[0163]

[0164] Where w max and w min are the upper and lower limits of the inertia weight, with values ​​of 0.9 and 0.5 respectively; s is the current number of iterations; γ is the empirical adjustment coefficient, with a range of [1,30] and a value of 3 in an optional embodiment of the present invention.

[0165] 3) Multi-objective optimal solution selection based on TOPSIS

[0166] Since each particle j is composed of a charging scheme, the objective function f of each particle j in s iterations can be calculated according to the scheme. cost_j (s) and f sati_j (s), it is necessary to further select the optimal particle of the current iteration according to the objective function. TOPSIS first needs to standardize the objective function to eliminate the dimensional difference. The formula is as follows:

[0167]

[0168] Where f′ cost_j (s) represents the normalized objective function value of particle j at the sth iteration; max{f cost_j (s)}, min{f cost_j (s)} are the sub-goals f at the end of s iterations. cost The maximum and minimum values ​​that appear; similarly for the sub-goal f sati_j (s) The normalized formula is as follows:

[0169]

[0170] The next step is to calculate the minimum particle distance vector group and the maximum particle distance vector group of TOPSIS. The calculation formula is as follows:

[0171]

[0172] Where, are the positive and negative ideal distances respectively.

[0173] The next step is to find the particle fit based on the positive and negative ideal distances. The formula is as follows:

[0174]

[0175] Where C j is the fit of particle j. The smaller the value, the better the particle.

[0176] The following is an illustration of the practical application of the optional embodiments of the present invention.

[0177] The optional implementation method of the present invention uses a certain agricultural industrial park as an example verification object. The electric agricultural implements are mainly mobile agricultural facilities. The equipment types and main parameters are shown in Table 1. There are 4 charging piles in the agricultural park, all of which use 7kW slow charging. The charging pile contains a sensor that can sense the real-time SOC status of the battery when charging. It also has the functions of delay, timing and charging / stopping according to instructions. The charging pile load is powered by a 10kV / 0.4kV box transformer. At the same time, the transformer also supplies power to other basic uncontrollable electrical loads in the agricultural park. The park is also equipped with distributed photovoltaics with an installed capacity of 100kVA. It supplies power to the power grid through an inverter. The operation mode is self-generation and self-use, and the surplus power is connected to the grid.

[0178] Table 1

[0179]

[0180]

[0181] A multi-objective particle swarm algorithm was used with a population size of 50 and an upper limit of 200 iterations. The time-of-use electricity price cost is as follows: 0.261 yuan / kWh during off-peak hours (10:00 PM to 6:00 AM), 0.51 yuan / kWh during normal hours (6:00 AM to 8:00 AM and 3:00 PM to 5:00 PM), and 0.759 yuan / kWh during peak hours (8:00 AM to 3:00 PM and 5:00 PM to 9:00 PM).

[0182] Figure 3 This is a schematic diagram of historical data provided according to an optional embodiment of the present invention. Gaussian fitting parameters are used to simulate the charging time of various agricultural tools. Taking an electric forklift as an example, the probability density distribution of the charging time of its historical data and the Gaussian fitting distribution are shown in FIG. Figure 3 As shown in the figure, the Gaussian fitting parameters μ and δ are 3.4 and 1.3, respectively. Similarly, the fitting parameters μ and δ for the mobile electric irrigation machine are 6.4 and 0.2, the fitting parameters for the electric pickup truck are 2.6 and 0.2, the fitting parameters for the small electric transport vehicle are 4.3 and 0.5, and the fitting parameters for the mobile agricultural vehicle are 6.0 and 1.0.

[0183] At 4:00 PM on a particular day, four charging stations were connected to an electric pickup truck with a state of charge (SOC0) of 0.6, an electric forklift with a SOC0 of 0.8, an electric irrigation machine with a SOC0 of 0.3, and a mobile agricultural vehicle with a SOC0 of 0.4. Using Gaussian distributions for the respective charging times, we calculated charging times of 3 hours, 8 hours, 6 hours, and 5 hours, respectively.

[0184] The charging power of each device within the current charging time is optimized. The objective function is the operating cost and farmer satisfaction. The multi-objective particle swarm algorithm based on TOPSIS sorting is used to solve the problem. Figure 4 3 is a schematic diagram of an optimal particle fit curve provided according to an optional embodiment of the present invention.

[0185] The multi-objective algorithm of the optional implementation mode of the present invention achieved reliable iterative convergence in the 48th time, and the convergence result of the best fit was 0.056. The comprehensive operating cost at this time was 138.592 yuan, compared with the cost of disordered charging of 162.984 yuan, the cost was reduced by 14.964%. The user dissatisfaction was 3.407, which was a decrease of 1.275% compared with the cost of disordered charging of 3.451.

[0186] Figure 5 is a power stack diagram of the optimal ordered charging scheme obtained after convergence of the algorithm provided in an optional embodiment of the present invention, wherein: Figure 5 The load of power facilities is above the horizontal axis, and the photovoltaic power generation power is below the horizontal axis. Figure 5 As can be seen, at 4:00 PM and 5:00 PM, photovoltaic power generation is high, at 53.5 kW and 29 kW, respectively. During these times, each electric agricultural implement is charged at maximum power. During the evening hours from 6:00 PM to 8:00 PM, photovoltaic power decreases, and the state of charge (SOC) of each electric agricultural implement reaches 80%. An optional embodiment of the present invention automatically switches the charging power to a constant voltage charging state, allowing the maximum charging power to decrease in an orderly manner. Simultaneously, while meeting the respective charging times, charging power is rationally allocated to ensure the lowest peak-to-valley difference in charging load and the highest possible state of charge (SOC). From 9:00 PM to 11:00 PM, the only time the electric forklift is charging, and according to the principle of charging as quickly as possible for the user, the electric forklift is fully charged, with no charging load and only basic other electrical loads.

[0187] In summary, the optional embodiments of the present invention have the following advantages:

[0188] 1. A predictive generation method for distributed photovoltaic and typical loads is proposed, along with a Monte Carlo simulation-based charging duration prediction method. This makes the optimization scheme forward-looking and can effectively guide farmers in effective charging. The scheme fully considers the actual charging power characteristics and lifecycle cost characteristics of batteries, and considers the charging power changes in the constant current and constant voltage stages of battery charging, improving the conventional constant power charging mode.

[0189] 2. At the same time, the equivalent life cost is calculated more deeply based on the battery charging depth and state of charge range, so that the orderly charging plan is more in line with the actual situation of the battery;

[0190] 3. Establish a long-term charging power optimization model. This model can adjust the charging power of different devices at different times based on different devices, different initial states of charge, and predicted charging times. This optimization model is more flexible and applicable. It also considers comprehensive operating costs and user satisfaction, covering actual factors such as electricity purchase costs, battery life costs, load peak-valley differences, battery charging efficiency, and battery fullness, taking into account a comprehensive range of factors.

[0191] 4. A particle swarm algorithm combining TOPSIS sorting and inertia weight is used to solve the optimization model. This algorithm can adjust the optimization capability in real time. At the same time, it can effectively eliminate the dimension in each iteration process, making the selection more concise and efficient. Combined with dynamic inertia weight, the calculation efficiency can be further improved.

[0192] According to an embodiment of the present invention, an orderly charging device is also provided. Figure 6 is a structural block diagram of an orderly charging device provided according to an embodiment of the present invention, such as Figure 6 As shown, the device includes: a first determination module 61, an acquisition module 62, a second determination module 63 and an optimization module 64. The device is described below.

[0193] A first determination module 61 is used to determine the theoretical charging time and predicted charging time of the target device based on the type of the target device; an acquisition module 62 is connected to the above-mentioned first determination module 61, and is used to obtain the charging power of the battery of the target device as the charging time and state of charge change; a second determination module 63 is connected to the above-mentioned acquisition module 62, and is used to determine the initial charging strategy based on the charging power and the predicted charging time; an optimization module 64 is connected to the above-mentioned second determination module 63, and is used to iteratively optimize the initial charging strategy based on the preset constraint condition group and the adaptive inertia weight within a preset number of iterations, with the goal of minimizing the sum of the operating cost and the degree of charging dissatisfaction, to obtain a target charging strategy.

[0194] According to an embodiment of the present invention, a computer-readable storage medium is further provided. The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned orderly charging methods.

[0195] According to an embodiment of the present invention, a computer device is also provided, comprising: a memory and a processor, wherein the memory stores a computer program; and the processor is configured to execute the computer program stored in the memory, wherein when the computer program is executed, the processor executes any one of the above-mentioned orderly charging methods.

[0196] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0197] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0198] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0199] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0200] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0201] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc. Various media that can store program codes.

[0202] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. An orderly charging method, characterized in that: include: Determining a predicted charging time of the target device based on a type of the target device; Obtaining the charging power of the battery of the target device as it changes with charging time and state of charge; determining an initial charging strategy based on the charging power and the predicted charging duration; Within a preset number of iterations, with the goal of minimizing the sum of the operating cost and the degree of charging dissatisfaction, based on a preset set of constraints and in accordance with the adaptive inertia weight, the initial charging strategy is iteratively optimized to obtain a target charging strategy; Among them, determining the predicted charging duration of the target device based on the type of the target device includes: obtaining actual charging data within a third predetermined time range before the charging day based on the type of the target device, wherein the actual charging data includes actual charging duration and actual charging frequency; determining the actual charging probability distribution corresponding to the actual charging data; performing Gaussian distribution fitting on the actual charging probability distribution to obtain a fitting result; and generating the predicted charging duration based on the fitting result.

2. The method according to claim 1, characterized in that The acquiring of the charging power of the battery of the target device as it changes with the charging time and the state of charge includes: Obtaining the current state of charge of the battery; When the current state of charge is less than a preset state of charge, determining that the charging power is a first power that varies with charging time in a constant current charging mode; When the current state of charge is greater than or equal to the preset state of charge, the charging power is determined to be a second power that varies with charging time in a constant voltage charging mode.

3. The method according to claim 1, characterized in that The preset constraint condition group includes at least one of the following: Charging power and state of charge constraints, charging time constraints, and battery charging and discharging life cost constraints.

4. The method according to claim 1, wherein The method further comprises: iteratively optimizing the initial charging strategy within a preset number of iterations, with the goal of minimizing the sum of the operating cost and the degree of charging dissatisfaction, based on a preset constraint condition group and in accordance with the adaptive inertia weight, to obtain a target charging strategy, including: Determining the operating cost and charging dissatisfaction level corresponding to the initial charging strategy; Screening the initial charging strategy based on the operating cost and charging dissatisfaction level corresponding to the initial charging strategy to obtain a first candidate charging strategy; Under the condition that the preset constraint condition group is satisfied, according to the preset update method and the adaptive inertia weight, within the preset number of iterations, with the goal of minimizing the sum of the operating cost and the charging dissatisfaction level, the first candidate charging strategy is iteratively optimized to obtain a second candidate charging strategy; determining an optimization evaluation factor for the second candidate charging strategy; The target charging strategy is determined based on the second candidate charging strategy and the optimization evaluation factor.

5. The method according to claim 1, wherein The method for determining the operating cost and the degree of charging dissatisfaction includes: Based on historical sunlight data within a first predetermined time range before the charging day, predicting the photovoltaic power generation power that varies with time during the charging day; predicting a base load that varies over time on the charging day based on a historical load curve within a second predetermined time range before the charging day; Determining a theoretical charging time of the target device based on the type of the target device; Determining the equivalent cycle life of the battery in this charging process based on the charge and discharge depth of the battery; Determining a battery life loss cost based on the equivalent cycle life number; Based on the photovoltaic power generation power, the base load, the theoretical charging time, the predicted charging time, the charging power and the battery life loss cost, the operating cost and the charging dissatisfaction level of the initial charging strategy are determined respectively.

6. The method according to any one of claims 1 to 5, characterized in that The value of the adaptive inertia weight is determined by the current iteration number in the iterative optimization process.

7. An orderly charging device, characterized in that: The device is used to perform the orderly charging method according to any one of claims 1 to 6, comprising: A first determining module is configured to determine a theoretical charging time and a predicted charging time of the target device based on a type of the target device; An acquisition module, configured to acquire the charging power of the battery of the target device as it changes with charging time and state of charge; a second determining module, configured to determine an initial charging strategy based on the charging power and the predicted charging duration; an optimization module for iteratively optimizing the initial charging strategy within a preset number of iterations, with the goal of minimizing the sum of the operating cost and the degree of charging dissatisfaction, based on a preset constraint condition group and in accordance with an adaptive inertia weight, to obtain a target charging strategy; Among them, the first determination module is also used to obtain actual charging data within a third predetermined time range before the charging date based on the type of the target device, wherein the actual charging data includes actual charging duration and actual charging frequency; determine the actual charging probability distribution corresponding to the actual charging data; perform Gaussian distribution fitting on the actual charging probability distribution to obtain a fitting result; and generate the predicted charging duration based on the fitting result.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the orderly charging method according to any one of claims 1 to 6.

9. A computer device, characterized in that: include: memory and processor, The memory stores a computer program; The processor is configured to execute a computer program stored in the memory, and when the computer program is executed, the processor is enabled to execute the orderly charging method according to any one of claims 1 to 6.

Images