Power battery SOC intelligent prediction method and system based on engineering machinery working condition

Abstract

The invention belongs to the technical field of battery parameter prediction, and provides a power battery SOC intelligent prediction method and system based on engineering machinery working conditions, and the method comprises the steps: obtaining a load spectrum of the engineering machinery working conditions; calculating the characteristic current of the engineering machinery power battery under the working condition of the engineering machinery according to the obtained load spectrum, and obtaining an open-circuit voltage-state-of-charge data set by adopting test equipment; optimizing hyper-parameters of the long short-term memory network by adopting a black eagle hybrid intelligent algorithm, and constructing an intelligent prediction model of the state of charge of the power battery; and predicting the state of charge of the power battery based on the power battery open-circuit voltage value obtained in real time and the constructed intelligent prediction model of the state of charge of the power battery, and completing intelligent prediction of the state of charge of the power battery based on the working condition of the engineering machinery.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery parameter prediction, and specifically relates to a power battery SOC intelligent prediction method and system based on engineering machinery working conditions. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] In the field of engineering machinery, the state of charge (SOC) of the power battery has a crucial impact on the performance of the equipment. The main source of SOC power for new energy engineering machinery is the battery pack. Accurate SOC estimation can ensure that new energy engineering machinery reasonably distributes electrical energy during operation and avoid work interruptions due to insufficient power. When the SOC estimation is accurate, the engineering machinery can intelligently adjust the power output according to the remaining power and the needs of the work task, making the operation smoother and more efficient.

[0004] Compared with the SOC state estimation of automobiles, the SOC state estimation error of engineering machinery in specific working conditions is relatively large. Although traditional SOC estimation methods such as the ampere-hour integration method are simple and direct, they have the problem of cumulative error. During the long-term charging and discharging process, due to factors such as the error in current measurement, the SOC calculated by the ampere-hour integration method will gradually deviate from the true value; however, the limitations of existing methods have prompted researchers to continuously explore more accurate and reliable SOC estimation methods for engineering machinery. Summary of the invention

[0005] To solve the above problems, the present invention proposes a method and system for intelligent prediction of power battery SOC based on engineering machinery working conditions, and adopts the Black Hawk hybrid intelligent algorithm to optimize the hyperparameters of the long short-term memory network (LSTM) to obtain an intelligent prediction model of the power battery SOC. The open circuit voltage (OCV)-state of charge (OCV-SOC) of the power battery is obtained according to the load spectrum of the engineering machinery working conditions collected in real time to complete the intelligent prediction of the power battery SOC, thereby improving the accuracy and precision of the power battery state of charge prediction results.

[0006] According to some embodiments, a first solution of the present invention provides a power battery SOC intelligent prediction method based on the working conditions of engineering machinery, which adopts the following technical solutions:

[0007] A power battery SOC intelligent prediction method based on engineering machinery working conditions, comprising:

[0008] Obtain the load spectrum of the construction machinery working condition;

[0009] The characteristic current of the power battery of the construction machinery under the working condition of the construction machinery is calculated according to the obtained load spectrum, and the open circuit voltage-state of charge data set is obtained by using the test equipment;

[0010] The Black Hawk hybrid intelligent algorithm is used to optimize the hyperparameters of the long short-term memory network and build an intelligent prediction model for the state of charge of power batteries;

[0011] Predict the state of charge of the power battery based on the real-time acquired open-circuit voltage value of the power battery and the constructed power battery state of charge intelligent prediction model, and complete the intelligent prediction of the power battery state of charge based on the working conditions of the engineering machinery;

[0012] Among them, the Black Hawk hybrid intelligent algorithm adopted is to distinguish individual behaviors by calculating the hunger degree of the vulture algorithm, combining the aggregation stage in the vulture algorithm with the grasping behavior in the Black Hawk algorithm to find the local optimal solution, and combining the tail-flicking behavior in the artificial fish school algorithm with the alarm behavior in the Black Hawk algorithm to jump out of the local optimal solution to find the global optimal solution, that is, to obtain the Black Hawk hybrid intelligent algorithm.

[0013] As a further technical limitation, the black hawk algorithm includes at least black hawk search, black hawk capture, black hawk grab and black hawk alarm;

[0014] The Black Hawk search formula is:

[0015]

[0016] in, Indicates the position of the first random black hawk when the cycle number is t+1; represents a random position in the search space when the number of cycles is t; α represents a quantitative parameter, and r 1 Represents a random number; t 1 Represents a random number formed by tent mapping; It represents the best position when the number of cycles is t; represents the position of a random black hawk when the number of cycles is t; represents the position of the second random black hawk when the number of cycles is t+1; ub and lb represent the upper and lower bounds of the search space respectively; X t+1 (i, j) indicates the position of the black hawk when the number of cycles is t+1; Minimizefitness() indicates finding the minimum value; X t (i, j) represents the position of the black hawk when the number of cycles is t;

[0017] The Black Hawk capture formula is:

[0018]

[0019] Among them, D1 represents the first position adjustment factor, and D 2 represents the second position adjustment factor, and s 0 is a column vector of dimension d; X * It indicates the position of Black Hawk after the first adjustment;

[0020] The Black Hawk grab formula is:

[0021]

[0022] Among them, X t+1 (i, j) represents the position of the black hawk when the cycle number is t+1; represents the best position when the number of cycles is t; X t (i, j) represents the position of the black hawk when the number of cycles is t; r 3 is a random vector of dimension d; i represents the number of rows in the matrix vector; j represents the number of columns in the matrix vector;

[0023] The Black Hawk alarm formula is:

[0024]

[0025] in, represents the position of the random black hawk in the matrix P when the number of cycles is t+1;

[0026] P(j)=λ j / j! e -λ (j=12···n), λ=-10d sni +15

[0027] d sni =(d si -d smin ) / (d smax -d smin ), β represents a quantitative parameter, and λ n =(λ-5) / 10; d si express The distance from the center of the search space in the i-th dimension, d smin and d smax They are d si The minimum and maximum values ​​of Indicates that X t (i,j) according to The position matrix after rearrangement in order of proximity.

[0028] Furthermore, the hunger degree of the vulture algorithm is in, represents the hunger degree of the individual in row i when the number of cycles is t; when the obtained Hunger represents a random number between [0,1]; h t represents a random number between [-1,1]; t represents the number of loops; T represents the total number of loops; m t is a random number in [-2,2], and w is a constant.

[0029] Furthermore, when Or when the number of local optimization cycles does not exceed the local optimization cycle threshold, the individual enters the local optimization stage; specifically: according to the Black Hawk algorithm, the best position when the number of cycles of the Black Hawk algorithm is t+1 is obtained Based on the Black Hawk grab formula, the position X of the Black Hawk when the cycle number is t+1 is obtained t+1 (i,j), the individual enters the vulture gathering stage, that is

[0030] ;

[0031] in, represents the position of the random vulture in the matrix A when the number of cycles is t+1; represents the first best position when the number of cycles is t; represents the second best position when the number of cycles is t; the position of the random vulture in the matrix A when the number of cycles is t+1 is obtained Substitute into the Black Hawk grab formula to get the first fusion formula

[0032]

[0033] The position of the black hawk when the number of cycles of the combination of the black hawk algorithm's grabbing behavior and the vulture algorithm's gathering behavior is t+1, and its optimal position is By comparing the optimal position of the Black Hawk algorithm grasping behavior when the number of cycles is t+1 The optimal position of the black hawk when the number of cycles is t+1 combined with the black hawk algorithm grabbing behavior and the vulture algorithm gathering behavior Get the best position of the black hawk when the individual enters the local optimization stage

[0034] Add 1 to the number of local optimization cycles, recalculate the hunger degree of the vulture algorithm, and judge again whether the individual enters the local optimization stage. Repeat the local optimization cycle until the individual cannot enter the local optimization stage. At this time, the best position of the black hawk in the local optimization stage is obtained, and the local optimization of the individual is completed.

[0035] when When the number of local optimization cycles exceeds the local optimization cycle number threshold, the individual enters the jump-out stage, which includes alarm behavior and tail-flicking behavior. The alarm behavior or tail-flicking behavior is selected according to the parameters; specifically:

[0036] when When , the tail-flicking behavior of the artificial fish algorithm is adopted, that is:

[0037]

[0038] Among them, Rand() refers to a random function that obeys uniform distribution, and the (0,1) in brackets refers to the value range; t represents the current number of iterations; T represents the maximum number of iterations; XF max Indicates the state of the optimal artificial fish; XF i Indicates artificial fish AF i Status; XF i|next Indicates the artificial fish AF after the position is updated i Status; YF max Indicates the maximum value of nutrient concentration in the artificial fish swarm algorithm; nf f represents the number of artificial fish in the field of view of the artificial fish after the position is updated; step represents the step length, that is, the maximum distance the artificial fish moves each time; δ represents the crowding factor used to control the density of the artificial fish school;

[0039] when When , the alarm behavior of the Black Hawk algorithm is combined with the offset distance in the tail-flicking behavior of the artificial fish algorithm, and the second fusion formula is obtained: ;

[0040] in, Indicates the position of the black hawk when the number of cycles of combining the alarm behavior of the black hawk algorithm with the offset distance in the tail-flicking behavior of the artificial fish algorithm is t+1;

[0041] Execute the jump-out phase, set the number of local optimization cycles to zero and add 1 to the jump-out phase execution number until the jump-out phase execution number exceeds the jump-out phase execution number threshold, thereby obtaining the Black Hawk hybrid intelligent algorithm; otherwise, recalculate the hunger degree of the vulture algorithm and judge again whether the individual enters the local optimization phase or the jump-out phase.

[0042] As a further technical limitation, the engineering machinery is new energy engineering machinery, including at least electric bulldozers, electric loaders, electric excavators and electric graders; the load spectrum of the new energy engineering machinery under specific working conditions is collected, and a typical load spectrum is constructed by big data analysis. The constructed typical load spectrum is calculated through a simulation model to obtain the characteristic working current of the battery under the target working condition, that is, the current variation law of the power battery under different load spectra is obtained, and an open circuit voltage-state of charge test of the power battery is performed based on the obtained current variation law to obtain an open circuit voltage-state of charge data set of the power battery.

[0043] As a further technical limitation, the Black Hawk hybrid intelligent algorithm is used to optimize the number of hidden layer neurons and the initial learning rate of the long short-term memory network. The root mean square error between the predicted value and the actual value of the power battery state of charge is used as the fitness value to obtain the optimal combination of the long short-term memory network hyperparameters. The optimization of the long short-term memory network hyperparameters is completed to obtain the intelligent prediction model of the power battery state of charge.

[0044] According to some embodiments, the second solution of the present invention provides a power battery SOC intelligent prediction system based on the working conditions of engineering machinery, which adopts the following technical solutions:

[0045] A power battery SOC intelligent prediction system based on engineering machinery working conditions, comprising:

[0046] An acquisition module, which is configured to acquire a load spectrum of a working condition of the engineering machinery;

[0047] A calculation module, which is configured to calculate the characteristic current of the power battery of the construction machinery under the working condition of the construction machinery according to the acquired load spectrum, and obtain an open circuit voltage-state of charge data set using a test device;

[0048] A building module configured to optimize the hyperparameters of a long short-term memory network using a Black Hawk hybrid intelligent algorithm to build an intelligent prediction model for a power battery state of charge;

[0049] A prediction module, which is configured to predict the state of charge of the power battery based on the real-time acquired open-circuit voltage value of the power battery and the constructed power battery state of charge intelligent prediction model, and complete the intelligent prediction of the power battery state of charge based on the working condition of the engineering machinery;

[0050] Among them, the Black Hawk hybrid intelligent algorithm adopted is to distinguish individual behaviors by calculating the hunger degree of the vulture algorithm, combining the aggregation stage in the vulture algorithm with the grasping behavior in the Black Hawk algorithm to find the local optimal solution, and combining the tail-flicking behavior in the artificial fish school algorithm with the alarm behavior in the Black Hawk algorithm to jump out of the local optimal solution to find the global optimal solution, that is, to obtain the Black Hawk hybrid intelligent algorithm.

[0051] According to some embodiments, a third solution of the present invention provides a computer-readable storage medium, which adopts the following technical solution:

[0052] A computer-readable storage medium stores a program thereon, which, when executed by a processor, implements the steps of the power battery SOC intelligent prediction method based on engineering machinery working conditions as described in the first embodiment of the present invention.

[0053] According to some embodiments, a fourth solution of the present invention provides an electronic device, which adopts the following technical solution:

[0054] An electronic device comprises a memory, a processor and a program stored in the memory and running on the processor, wherein when the processor executes the program, the steps in the method for intelligently predicting the SOC of a power battery based on the working condition of an engineering machinery as described in the first embodiment of the present invention are implemented.

[0055] According to some embodiments, a fifth solution of the present invention provides a computer program product, which adopts the following technical solution:

[0056] A computer program product includes software code, wherein the program in the software code executes the steps in the method for intelligently predicting the SOC of a power battery based on the working condition of an engineering machinery as described in the first embodiment of the present invention.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] The present invention distinguishes individual behaviors by calculating the hunger degree of the vulture algorithm, combines the grabbing behavior in the black hawk algorithm with the gathering behavior in the vulture algorithm, and combines the alarm behavior in the black hawk algorithm with the tail-flicking behavior in the artificial fish school algorithm to obtain a black hawk hybrid intelligent algorithm;

[0059] The present invention collects the load spectrum of engineering machinery working conditions, calculates the characteristic working current of the battery under the target working conditions through big data analysis simulation, obtains the current change law of the power battery, and based on this law, performs OCV-SOC experiments on the power battery and obtains the OCV-SOC data set.

[0060] Based on the OCV-SOC data set, the present invention adopts the Black Hawk hybrid intelligent algorithm to optimize the hyperparameters of the LSTM neural network, obtains the power battery SOC intelligent prediction model, collects the OCV value of the power battery in real time, obtains the SOC value at this time through the model, completes the power battery SOC intelligent prediction based on the engineering machinery working condition, and improves the accuracy and precision of the power battery charge state prediction result. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The drawings in the specification that constitute a part of this embodiment are used to provide a further understanding of this embodiment. The schematic embodiments of this embodiment and their descriptions are used to explain this embodiment and do not constitute improper limitations on this embodiment.

[0062] Figure 1 This is a flow chart of a power battery SOC intelligent prediction method based on engineering machinery working conditions in Embodiment 1 of the present invention;

[0063] Figure 2 This is a working principle diagram of the power battery SOC intelligent prediction method based on the working condition of engineering machinery in the first embodiment of the present invention;

[0064] Figure 3 It is a structural block diagram of a power battery SOC intelligent prediction system based on engineering machinery working conditions in the second embodiment of the present invention. DETAILED DESCRIPTION

[0065] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0066] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0067] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0068] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0069] Embodiment 1

[0070] Embodiment 1 of the present invention introduces a method for intelligently predicting the SOC of a power battery based on the working conditions of an engineering machinery.

[0071] like Figure 1 A power battery SOC intelligent prediction method based on the working condition of engineering machinery is shown, comprising:

[0072] Obtain the load spectrum of the construction machinery working condition;

[0073] The characteristic current of the power battery of the construction machinery under the working condition of the construction machinery is calculated according to the acquired load spectrum, and the open circuit voltage-state of charge data set is obtained by using the test equipment;

[0074] The Black Hawk hybrid intelligent algorithm is used to optimize the hyperparameters of the long short-term memory network and build an intelligent prediction model for the state of charge of power batteries;

[0075] Predict the state of charge of the power battery based on the real-time acquired open-circuit voltage value of the power battery and the constructed power battery state of charge intelligent prediction model, and complete the intelligent prediction of the power battery state of charge based on the working conditions of the engineering machinery;

[0076] Among them, the Black Hawk hybrid intelligent algorithm adopted is to distinguish individual behaviors by calculating the hunger degree of the vulture algorithm, combining the aggregation stage in the vulture algorithm with the grasping behavior in the Black Hawk algorithm to find the local optimal solution, and combining the tail-flicking behavior in the artificial fish school algorithm with the alarm behavior in the Black Hawk algorithm to jump out of the local optimal solution to find the global optimal solution, that is, to obtain the Black Hawk hybrid intelligent algorithm.

[0077] The engineering machinery in this embodiment is new energy engineering machinery, including electric bulldozers, electric loaders, electric excavators, electric graders, etc.; load spectra are collected for specific working conditions of new energy engineering machinery, and typical load spectra are constructed by big data analysis. The typical load spectra are calculated by simulation models to obtain the characteristic working current of the battery under the target working conditions, thereby obtaining the current variation law of the power battery under different load spectra, and then based on this law, the power battery is subjected to an OCV-SOC test using experimental equipment to obtain an OCV-SOC data set of the power battery.

[0078] This embodiment uses the Black Hawk hybrid intelligent algorithm to optimize the hyperparameters of the LSTM neural network. Based on the power battery OCV-SOC data set, the number of neurons in the two hidden layers and the initial learning rate of the cells are optimized. Then, the root mean square error of the SOC predicted and measured by the selected battery data is used as the fitness value to obtain the hyperparameter combination that makes the LSTM neural network model estimate the best, thereby obtaining the SOC intelligent prediction model of the power battery; the specific working principle is as follows Figure 2 shown.

[0079] In this embodiment, the Black Hawk algorithm process mainly includes searching, capturing, grabbing, and alarming; specifically:

[0080] The Black Hawk search formula is:

[0081]

[0082] in, Indicates the position of the first random black hawk when the cycle number is t+1; represents a random position in the search space when the number of cycles is t; α represents a quantitative parameter, and r 1 Represents a random number; t 1Represents a random number formed by tent mapping; represents the best position when the number of cycles is t; represents the position of a random black hawk when the number of cycles is t; represents the position of the second random black hawk when the number of cycles is t+1; ub and lb represent the upper and lower bounds of the search space respectively; X t+1 (i, j) represents the position of the black hawk when the cycle number is t+1;

[0083] Minimizefitness() means to find the minimum value; X t (i,j) represents the position of the black hawk when the number of cycles is t.

[0084] The Black Hawk capture formula is:

[0085]

[0086] Among them, D 1 represents the first position adjustment factor, and D 2 represents the second position adjustment factor, and s 0 is a column vector of dimension d, whose elements are between (0.5,1); X * Indicates the position of Black Hawk after the first adjustment; s 0 is a column vector of dimension d.

[0087] The Black Hawk grab formula is:

[0088]

[0089] Among them, X t+1 (i, j) represents the position of the black hawk when the cycle number is t+1; represents the best position when the number of cycles is t; X t (i, j) represents the position of the black hawk when the number of cycles is t; r 3 is a random vector of dimension d, whose elements in each dimension follow a normal distribution; i represents the number of rows in the matrix vector; j represents the number of columns in the matrix vector.

[0090] The Black Hawk alarm formula is:

[0091]

[0092] in, represents the position of the random black hawk in the matrix P when the number of cycles is t+1;

[0093] P(j)=λ j / j! e -λ (j=12···n), λ=-10dsni +15

[0094] d sni =(d si -d smin ) / (d smax -d smin ), β represents a quantitative parameter, and λ n =(λ-5) / 10, d si yes The distance from the center of the search space in the i-th dimension, d smin and d smax They are d si The minimum and maximum values ​​of Is X t According to The position matrix after rearrangement in order of proximity.

[0095] The hunger degree of the vulture algorithm in this embodiment is

[0096]

[0097] in, represents the hunger degree of the individual in row i when the number of cycles is t; when the obtained Hunger represents a random number between [0,1]; h t represents a random number between [-1,1]; t represents the number of loops; T represents the total number of loops; m t is a random number in [-2,2], w is a constant 2.5, and the smaller w is, the greater the probability that the later algorithm will enter the development stage.

[0098] when Or when the number of local optimization cycles NF does not exceed the local optimization cycle number threshold Tf (in this embodiment, Tf=10), the individual enters the local optimization stage; specifically:

[0099] (1) According to the Black Hawk algorithm, the optimal position when the number of cycles of the Black Hawk algorithm is t+1 is obtained

[0100]

[0101] (2) Based on the Black Hawk grab formula, the Black Hawk’s position X is obtained when the number of cycles is t+1 t+1 (i,j), the individual enters the vulture gathering stage, that is

[0102] ;

[0103] in, represents the position of the random vulture in the matrix A when the number of cycles is t+1; represents the first best position when the number of cycles is t; represents the second best position when the number of cycles is t; the position of the random vulture in the matrix A when the number of cycles is t+1 is obtained Substitute into the Black Hawk grab formula to get the first fusion formula in, It indicates the position of the black hawk when the number of cycles of the combination of the black hawk algorithm grabbing behavior and the vulture algorithm gathering behavior is t+1. The optimal position is

[0104] (3) By comparing the optimal position of the Black Hawk algorithm grasping behavior when the number of cycles is t+1 The optimal position of the black hawk when the number of cycles is t+1 combined with the black hawk algorithm grabbing behavior and the vulture algorithm gathering behavior Get the best position of the black hawk when the individual enters the local optimization stage

[0105] (4) Increase the number of local optimization cycles NF by 1, that is, NF = NF + 1, and recalculate the hunger degree of the vulture algorithm According to the recalculated vulture algorithm hunger Or the number of local optimization cycles NF determines whether the individual enters the local optimization stage or jumps out of the stage; when the individual enters the local optimization stage, the above process is re-executed until the judgment condition for the individual to enter the local optimization stage is not met (the judgment condition for the individual to enter the local optimization stage is Or NF≤Tf), the optimal position of the black hawk in the local optimization stage is obtained; at this time, the local optimization of the individual is completed and the jump-out stage is entered.

[0106] when When the number of local optimization cycles exceeds the local optimization cycle number threshold (the local optimization cycle number threshold Tf in this embodiment is set to 10), the individual enters the jump-out stage, which includes alarm behavior and tail-flicking behavior, and the alarm behavior or tail-flicking behavior is selected according to the parameters; specifically:

[0107] (1) When When , the tail-flicking behavior of the artificial fish algorithm is adopted, that is:

[0108]

[0109] Among them, Rand() refers to a random function that obeys uniform distribution, and the (0,1) in brackets refers to the value range; t represents the current number of iterations; T represents the maximum number of iterations; XF max Indicates the state of the optimal artificial fish; XFi Indicates artificial fish AF i Status; XF i|next Indicates the artificial fish AF after the position is updated i Status; YF max Indicates the maximum value of nutrient concentration in the artificial fish swarm algorithm; nf f represents the number of artificial fish in the field of view of the artificial fish after the position is updated; step represents the step length, that is, the maximum distance the artificial fish moves each time; δ represents the crowding factor used to control the density of the artificial fish school;

[0110] when When , the alarm behavior of the Black Hawk algorithm is combined with the offset distance in the tail-flicking behavior of the artificial fish algorithm, and the second fusion formula is obtained: ;

[0111] in, Indicates the position of the black hawk when the number of cycles of combining the alarm behavior of the black hawk algorithm with the offset distance in the tail-flicking behavior of the artificial fish algorithm is t+1;

[0112] (2) Execute the jump-out phase (i.e., the tail-flicking behavior of the artificial fish algorithm or the combination of the alarm behavior of the Black Hawk algorithm and the offset distance in the tail-flicking behavior of the artificial fish algorithm), set the number of local optimization cycles NF to zero and increase the number of jump-out phase executions NZ by 1, that is, Until the number of executions NZ of the jump-out stage exceeds the threshold number of executions Tt of the jump-out stage (in this embodiment, Tt=20), the Black Hawk hybrid intelligent algorithm is obtained; otherwise, the hunger degree of the vulture algorithm is recalculated, and the hunger degree S of the vulture algorithm obtained by the recalculation is obtained. i t Or the number of local optimization cycles NF determines whether the individual enters the local optimization stage or jumps out of the stage.

[0113] This embodiment adopts the Black Hawk hybrid intelligent algorithm to optimize the hyperparameters of the LSTM neural network. The specific process is: based on the power battery OCV-SOC data set, the number of neurons in the two hidden layers and the initial learning rate of the cells are optimized, and then the root mean square error of the SOC predicted and measured by the selected battery data is used as the fitness value, and the hyperparameter combination that makes the LSTM neural network model estimate the best is obtained to obtain the SOC intelligent prediction model of the power battery; the OCV value of the new energy engineering machinery during the operation process is collected in real time, and the SOC value at this time is obtained based on the obtained power battery SOC intelligent prediction model, thereby completing the power battery SOC intelligent prediction based on the engineering machinery working condition.

[0114] This embodiment distinguishes individual behaviors by calculating the hunger degree of the vulture algorithm, combines the grasping behavior in the black hawk algorithm with the aggregation behavior in the vulture algorithm to improve the local optimization ability, combines the alarm behavior in the black hawk algorithm with the tail-flicking behavior in the artificial fish school algorithm, and selects the tail-flicking behavior or the combined behavior to improve the global search ability; the fusion algorithm can better balance the global exploration and local development capabilities of the algorithm in the search process, dynamically adjust the search focus according to the characteristics of the problem, more effectively find the global optimal solution, accelerate the convergence process of the algorithm, and improve the accuracy of the solution.

[0115] This embodiment is based on the obtained Black Hawk hybrid intelligent algorithm and the OCV-SOC data set to optimize the hyperparameters of the LSTM neural network to obtain a power battery SOC intelligent prediction model, collect the OCV value of the power battery in real time, and obtain the SOC value at this time through the model, thereby obtaining a power battery SOC intelligent prediction method based on the working conditions of engineering machinery; this intelligent prediction method is used to accurately predict the SOC value of new energy engineering machinery, thereby ensuring the smooth progress of work as well as work quality and efficiency.

[0116] Embodiment 2

[0117] The second embodiment of the present invention introduces a power battery SOC intelligent prediction system based on the working conditions of engineering machinery.

[0118] like Figure 3 The power battery SOC intelligent prediction system based on the working condition of the engineering machinery shown includes:

[0119] An acquisition module, which is configured to acquire a load spectrum of a working condition of the engineering machinery;

[0120] A calculation module, which is configured to calculate the characteristic current of the power battery of the construction machinery under the working condition of the construction machinery according to the acquired load spectrum, and obtain an open circuit voltage-state of charge data set using a test device;

[0121] A building module configured to optimize the hyperparameters of a long short-term memory network using a Black Hawk hybrid intelligent algorithm to build an intelligent prediction model for a power battery state of charge;

[0122] A prediction module, which is configured to predict the state of charge of the power battery based on the real-time acquired open-circuit voltage value of the power battery and the constructed power battery state of charge intelligent prediction model, and complete the intelligent prediction of the power battery state of charge based on the working condition of the engineering machinery;

[0123] Among them, the Black Hawk hybrid intelligent algorithm adopted is to distinguish individual behaviors by calculating the hunger degree of the vulture algorithm, combining the aggregation stage in the vulture algorithm with the grasping behavior in the Black Hawk algorithm to find the local optimal solution, and combining the tail-flicking behavior in the artificial fish school algorithm with the alarm behavior in the Black Hawk algorithm to jump out of the local optimal solution to find the global optimal solution, that is, to obtain the Black Hawk hybrid intelligent algorithm.

[0124] The detailed steps are the same as the power battery SOC intelligent prediction method based on engineering machinery working conditions provided in Example 1, and will not be repeated here.

[0125] Embodiment 3

[0126] Embodiment 3 of the present invention provides a computer-readable storage medium.

[0127] A computer-readable storage medium stores a program thereon, which, when executed by a processor, implements the steps of the power battery SOC intelligent prediction method based on engineering machinery working conditions as described in the first embodiment of the present invention.

[0128] The detailed steps are the same as the power battery SOC intelligent prediction method based on engineering machinery working conditions provided in Example 1, and will not be repeated here.

[0129] Embodiment 4

[0130] A fourth embodiment of the present invention provides an electronic device.

[0131] An electronic device comprises a memory, a processor and a program stored in the memory and running on the processor, wherein when the processor executes the program, the steps in the method for intelligent prediction of SOC of a power battery based on an engineering machinery working condition as described in the first embodiment of the present invention are implemented.

[0132] The detailed steps are the same as the power battery SOC intelligent prediction method based on engineering machinery working conditions provided in Example 1, and will not be repeated here.

[0133] Embodiment 5

[0134] Embodiment 5 of the present invention provides a computer program product.

[0135] A computer program product includes software code, wherein the program in the software code executes the steps in the method for intelligent prediction of SOC of a power battery based on an engineering machinery working condition as described in the first embodiment of the present invention.

[0136] The detailed steps are the same as the power battery SOC intelligent prediction method based on engineering machinery working conditions provided in Example 1, and will not be repeated here.

[0137] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0138] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0139] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0140] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0141] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0142] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0143] The above description is only a preferred embodiment of the present embodiment and is not intended to limit the present embodiment. For those skilled in the art, the present embodiment may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present embodiment shall be included in the protection scope of the present embodiment.

Claims

1. A power battery SOC intelligent prediction method based on engineering machinery working conditions, characterized in that: include: Obtain the load spectrum of the construction machinery working condition; The characteristic current of the power battery of the construction machinery under the working condition of the construction machinery is calculated according to the obtained load spectrum, and the open circuit voltage-state of charge data set is obtained by using the test equipment; The Black Hawk hybrid intelligent algorithm is used to optimize the hyperparameters of the long short-term memory network and build an intelligent prediction model for the state of charge of power batteries; Predict the state of charge of the power battery based on the real-time acquired open-circuit voltage value of the power battery and the constructed power battery state of charge intelligent prediction model, and complete the intelligent prediction of the power battery state of charge based on the working conditions of the engineering machinery; Among them, the Black Hawk hybrid intelligent algorithm adopted is to distinguish individual behaviors by calculating the hunger degree of the vulture algorithm, combining the aggregation stage in the vulture algorithm with the grasping behavior in the Black Hawk algorithm to find the local optimal solution, and combining the tail-flicking behavior in the artificial fish school algorithm with the alarm behavior in the Black Hawk algorithm to jump out of the local optimal solution to find the global optimal solution, that is, to obtain the Black Hawk hybrid intelligent algorithm.

2. A power battery SOC intelligent prediction method based on engineering machinery working conditions as described in claim 1, characterized in that: The black hawk algorithm at least includes black hawk search, black hawk capture, black hawk grab and black hawk alarm; The Black Hawk search formula is: in, Indicates the position of the first random black hawk when the cycle number is t+1; represents a random position in the search space when the number of cycles is t; α represents a quantitative parameter, and r1 represents a random number; t1 represents a random number formed by tent mapping; represents the best position when the number of cycles is t; represents the position of a random black hawk when the number of cycles is t; represents the position of the second random black hawk when the number of cycles is t+1; ub and lb represent the upper and lower bounds of the search space respectively; X t+1 (i, j) represents the position of the black hawk when the cycle number is t+1; Minimizefitness() means to find the minimum value; X t (i, j) represents the position of the black hawk when the number of cycles is t; The Black Hawk capture formula is: Where D1 represents the first position adjustment factor, and D2 represents the second position adjustment factor, and s0 is a column vector of dimension d; X * It indicates the position of Black Hawk after the first adjustment; The Black Hawk grab formula is: Among them, X t+1 (i, j) represents the position of the black hawk when the cycle number is t+1; represents the best position when the number of cycles is t; X t (i,j) represents the position of the black hawk when the number of cycles is t; r3 is a random vector with dimension d; i represents the number of rows in the matrix vector; j represents the number of columns in the matrix vector; The Black Hawk alarm formula is: in, represents the position of the random black hawk in the matrix P when the number of cycles is t+1; P(j)=λ j / j!e -λ (j=12···n), λ=-10d sni +15 d sni =(d si -d smin ) / (d smax -d smin ), β represents a quantitative parameter, and λ n =(λ-5) / 10; d si express The distance from the center of the search space in the i-th dimension, d smin and d smax They are d si The minimum and maximum values ​​of Indicates that X t (i,j) according to The position matrix after rearrangement in order of proximity.

3. A power battery SOC intelligent prediction method based on engineering machinery working conditions as described in claim 2, characterized in that: The hunger degree of the vulture algorithm is in, represents the hunger degree of the individual in row i when the number of cycles is t; when the obtained Hunger represents a random number between [0,1]; h t represents a random number between [-1,1]; t represents the number of loops; T represents the total number of loops; m t is a random number in [-2,2], and w is a constant.

4. A power battery SOC intelligent prediction method based on engineering machinery working conditions as described in claim 3, characterized in that: when Or when the number of local optimization cycles does not exceed the local optimization cycle threshold, the individual enters the local optimization stage; specifically: according to the Black Hawk algorithm, the best position when the number of cycles of the Black Hawk algorithm is t+1 is obtained Based on the Black Hawk grab formula, the Black Hawk's position X is obtained when the cycle number is t+1 t+1 (i,j), the individual enters the vulture gathering stage, that is ； in, represents the position of the random vulture in the matrix A when the number of cycles is t+1; represents the first best position when the number of cycles is t; represents the second best position when the number of cycles is t; the position of the random vulture in the matrix A when the number of cycles is t+1 is obtained Substitute into the Black Hawk grab formula to get the first fusion formula in, It indicates the position of the black hawk when the number of cycles of the combination of the black hawk algorithm grabbing behavior and the vulture algorithm gathering behavior is t+1. The optimal position is By comparing the optimal position of the Black Hawk algorithm grasping behavior when the number of cycles is t+1 The optimal position of the black hawk when the number of cycles is t+1 combined with the black hawk algorithm grabbing behavior and the vulture algorithm gathering behavior Get the best position of the black hawk when the individual enters the local optimization stage Add 1 to the number of local optimization cycles, recalculate the hunger degree of the vulture algorithm, and judge again whether the individual enters the local optimization stage. Repeat the local optimization cycle until the individual cannot enter the local optimization stage. At this time, the best position of the black hawk in the local optimization stage is obtained, and the local optimization of the individual is completed. when When the number of local optimization cycles exceeds the local optimization cycle number threshold, the individual enters the jump-out stage, which includes alarm behavior and tail-flicking behavior. The alarm behavior or tail-flicking behavior is selected according to the parameters; specifically: when When , the tail-flicking behavior of the artificial fish algorithm is adopted, that is: Among them, Rand() refers to a random function that obeys uniform distribution, and the (0,1) in brackets refers to the value range; t represents the current number of iterations; T represents the maximum number of iterations; XF max Indicates the state of the optimal artificial fish; XF i Indicates artificial fish AF i Status; XF i|next Indicates the artificial fish AF after the position is updated i Status; YF max Indicates the maximum value of nutrient concentration in the artificial fish swarm algorithm; nf f represents the number of artificial fish in the field of view of the artificial fish after the position is updated; step represents the step length, that is, the maximum distance the artificial fish moves each time; δ represents the crowding factor used to control the density of the artificial fish school; when When , the alarm behavior of the Black Hawk algorithm is combined with the offset distance in the tail-flicking behavior of the artificial fish algorithm, and the second fusion formula is obtained: in, Indicates the position of the black hawk when the number of cycles of combining the alarm behavior of the black hawk algorithm with the offset distance in the tail-flicking behavior of the artificial fish algorithm is t+1; Execute the jump-out phase, set the number of local optimization cycles to zero and add 1 to the jump-out phase execution number until the jump-out phase execution number exceeds the jump-out phase execution number threshold, thereby obtaining the Black Hawk hybrid intelligent algorithm; otherwise, recalculate the hunger degree of the vulture algorithm and judge again whether the individual enters the local optimization phase or the jump-out phase.

5. The method for intelligent prediction of power battery SOC based on engineering machinery working conditions as claimed in claim 1, characterized in that: The engineering machinery is new energy engineering machinery, including at least an electric bulldozer, an electric loader, an electric excavator and an electric grader; the load spectrum of the specific working condition of the new energy engineering machinery is collected, and a typical load spectrum is constructed by big data analysis. The constructed typical load spectrum is calculated through a simulation model to obtain the characteristic working current of the battery under the target working condition, that is, the current variation law of the power battery under different load spectra is obtained, and an open circuit voltage-state of charge test of the power battery is performed based on the obtained current variation law to obtain an open circuit voltage-state of charge data set of the power battery.

6. A power battery SOC intelligent prediction method based on engineering machinery working conditions as described in claim 1, characterized in that: The Black Hawk hybrid intelligent algorithm is used to optimize the number of hidden layer neurons and the initial learning rate of the long short-term memory network. The root mean square error between the predicted value and the actual value of the power battery state of charge is used as the fitness value to obtain the optimal combination of the long short-term memory network hyperparameters. The optimization of the long short-term memory network hyperparameters is completed, and an intelligent prediction model for the power battery state of charge is obtained.

7. A power battery SOC intelligent prediction system based on engineering machinery working conditions, characterized in that: include: An acquisition module, which is configured to acquire a load spectrum of a working condition of the engineering machinery; A calculation module, which is configured to calculate the characteristic current of the power battery of the construction machinery under the working condition of the construction machinery according to the acquired load spectrum, and obtain an open circuit voltage-state of charge data set using a test device; A building module configured to use a Black Hawk hybrid intelligent algorithm to optimize the hyperparameters of a long short-term memory network and build an intelligent prediction model for a power battery state of charge; A prediction module, which is configured to predict the state of charge of the power battery based on the real-time acquired open-circuit voltage value of the power battery and the constructed power battery state of charge intelligent prediction model, and complete the intelligent prediction of the power battery state of charge based on the working condition of the engineering machinery; Among them, the Black Hawk hybrid intelligent algorithm adopted is to distinguish individual behaviors by calculating the hunger degree of the vulture algorithm, combining the aggregation stage in the vulture algorithm with the grasping behavior in the Black Hawk algorithm to find the local optimal solution, and combining the tail-flicking behavior in the artificial fish school algorithm with the alarm behavior in the Black Hawk algorithm to jump out of the local optimal solution to find the global optimal solution, that is, to obtain the Black Hawk hybrid intelligent algorithm.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the power battery SOC intelligent prediction method based on the working condition of engineering machinery as described in any one of claims 1 to 6 are implemented.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the steps of the power battery SOC intelligent prediction method based on the engineering machinery working condition as described in any one of claims 1-6 are implemented.

10. A computer program product comprising software code, characterized in that The program in the software code executes the steps of the power battery SOC intelligent prediction method based on the working condition of engineering machinery as described in any one of claims 1-6.

Images