Power control method and system suitable for hybrid excavator

Abstract

The invention discloses a power control method and system suitable for a hybrid excavator, and relates to the technical field of engineering machinery. The method comprises the following steps: collecting and preprocessing time sequence data; inputting the preprocessed time sequence data into the trained LSTM neural network model, and predicting to obtain a rotary motor power predicted value at a future moment; the battery power is predicted according to the predicted value of the power of the rotary motor, the final required power of the ISG is calculated according to the battery power, and the reduction of the target rotating speed of the electric rotary mechanism is calculated according to whether the predicted power of the battery exceeds a power limit value or not; and according to the final required power of the ISG and the target rotating speed decrement of the electric slewing mechanism, converting into a corresponding ISG motor and electric slewing mechanism control signal. According to the method, the power demand is predicted in advance through the LSTM network, the charging and discharging power and the SOC of the battery are actively stabilized in an ideal interval, the power fluctuation is effectively inhibited, and the service life of the battery is prolonged.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy engineering machinery, in particular to a power control method and system suitable for a hybrid excavator. BACKGROUND

[0002] Under the background of energy saving and emission reduction, the engineering machinery industry is accelerating the green transformation. New energy technology is widely used in engineering machinery products due to its low use cost, energy saving and emission reduction, and strong power. In the process of electrification, engineering machinery such as excavators and cranes with rotating platforms are replacing traditional hydraulic rotation with electric rotation mechanisms. The electric rotation mechanism is composed of a rotating motor and a speed reducer, which drives the platform to rotate and can recover braking energy through electric braking, thereby improving the overall efficiency of the machine.

[0003] However, the electric rotation mechanism requires high bus power during start-stop, and the frequent fluctuation of its power amplitude will have many negative effects on the power battery, such as accelerating battery aging, increasing heat generation, shortening service life, and reducing state estimation accuracy. The sharp fluctuation of charge and discharge power will also accelerate the cycle life of the battery. The power battery is expensive, and its early replacement may result in a total use cost of the hybrid excavator that is not lower than that of the traditional fuel engine. Therefore, how to smoothly control the charge and discharge power and SOC (State of Charge) of the battery is of great economic significance for reducing the use cost of the hybrid excavator and prolonging the battery life.

[0004] The prior art (such as CN202510768734.2, CN202510617471.5, etc.) mainly uses the ISG (Integrated Starter Generator) motor to "peak shaving and valley filling", which assists the engine to output power during heavy load and generates electricity to charge the battery during light load, so as to optimize the engine operating point or meet the power demand of the whole vehicle. However, these schemes generally ignore the negative impact of charge and discharge power or SOC fluctuation on battery life.

[0005] Therefore, there is an urgent need for a power control scheme that can effectively suppress the fluctuation of battery power and SOC to prolong the battery life. SUMMARY

[0006] The present application aims to overcome the shortcomings of the prior art and provide a power control method and system suitable for a hybrid excavator, which can effectively predict and smooth the charge and discharge power of the power battery in the hybrid excavator and stabilize its SOC in the ideal interval, thereby prolonging the battery life and reducing the overall use cost of the machine.

[0007] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0008] In a first aspect, the present application provides a power control method suitable for a hybrid excavator, the method comprising:

[0009] collecting and preprocessing time series data;

[0010] inputting the preprocessed time series data into a trained LSTM neural network model to predict a swing motor power prediction value at a future time; the input of the LSTM neural network model is multivariate time series data within a past time window, and the output is a swing motor power prediction value at a specified future time;

[0011] predicting battery power according to the swing motor power prediction value, calculating ISG final demand power according to the battery power, and calculating an electrical swing mechanism target speed decrement according to whether the battery prediction power exceeds a power limit value;

[0012] converting the ISG final demand power and the electrical swing mechanism target speed decrement into corresponding ISG motor and electrical swing mechanism control signals.

[0013] Further, according to the swing motor power prediction value, the battery power is predicted, the ISG final demand power is calculated according to the battery power, and the electrical swing mechanism target speed decrement is calculated according to whether the battery prediction power exceeds the power limit value, comprising:

[0014] calculating a first battery prediction power according to the swing motor power prediction value;

[0015] obtaining a power limit value of the ISG motor;

[0016] calculating an ISG first target power increment and an ISG second target power increment according to the first battery prediction power;

[0017] calculating the ISG final demand power according to the power limit value of the ISG motor, the ISG first target power increment and the ISG second target power increment;

[0018] updating a second battery prediction power according to the ISG final demand power;

[0019] calculating a swing motor target speed decrement according to the second battery prediction power.

[0020] Further, the time series data includes time series data of handle opening degree, swing motor speed, swing motor torque, swing motor bus voltage and swing motor bus current;

[0021] The collecting and preprocessing time series data comprises:

[0022] The original signals of the handle opening degree, the rotating motor speed, the rotating motor torque, the rotating motor bus voltage and the rotating motor bus current are collected by sensors;

[0023] The original signals are subjected to timestamp alignment, filtering and normalization processing.

[0024] Further, the training method of the LSTM neural network model comprises:

[0025] A training data set is constructed, the input sample of the data set is the time series data in a past time window, and the output sample is the actual rotating motor power value at a future specified time;

[0026] The mean square error is used as a loss function, the Adam is selected to optimize the parameter matrix and bias of each layer network, and the constructed LSTM neural network model is trained.

[0027] Further, the LSTM neural network model comprises an input layer, a first LSTM layer, a second LSTM layer and an output layer connected in sequence;

[0028] The input layer is used for receiving input data with a shape of (Ts, F), wherein Ts is a time step, and F is a feature dimension;

[0029] The first LSTM layer is used for extracting time sequence features of the input data, and the number of memory cells is 64;

[0030] The second LSTM layer is used for further analyzing the time sequence features and outputting a feature vector of a final time step, and the number of memory cells is 32;

[0031] The output layer is a fully connected layer, which is used for mapping the feature vector to a rotating motor power prediction value.

[0032] Further, the first predicted power of the battery is calculated according to the rotating motor power prediction value, comprising:

[0033] Wherein, the first predicted power of the battery, the predicted rotating motor power, the current ISG motor power.

[0034] Further, the power limit value of the ISG motor is obtained, comprising:

[0035] The ideal charging power limit value , the ideal discharging power limit value , the maximum charging power limit value , and the maximum discharging power limit value The charging power is negative and the discharging power is positive.

[0036] The maximum available power of the ISG motor is calculated according to the hydraulic system power and the engine power limit , the minimum available power . The maximum available power is positive, and the positive power indicates driving; the minimum available power is negative, and the negative power indicates power generation.

[0037] The minimum available power of the ISG motor when working normally is , and the maximum available power is ;

[0038] wherein, is the minimum power of the ISG motor (negative when generating power), is the maximum power of the ISG motor (positive when driving), and the two values depend on the external characteristics of the motor.

[0039] is the current hydraulic system power, which is calculated by measuring the pressure and flow of the hydraulic pump, and is positive power.

[0040] is the current engine power, and the power range is , which is related to the external characteristics of the engine.

[0041] Further, the first target power increment of the ISG is calculated, including:

[0042] According to the first predicted power of the battery and the relationship between the ideal charging power limit and the ideal discharging power limit of the battery, the first power increment is calculated, and the relationship is:

[0043] When is between the ideal charging power limit and the ideal discharging power limit, that is , the ISG motor target power increment , and the ISG motor maintains the current power output;

[0044] When exceeds the ideal discharging power limit, that is , the ISG motor target power increment , is negative, and the ISG motor reduces the current power output;

[0045] When is lower than the ideal charging power limit, that is , the ISG motor target power increment , ISG motor increases the current power output if the value is positive.

[0046] Further, the ISG second target power increment is calculated, including:

[0047] According to the difference between the current SOC of the battery and the target SOC, a preset mapping table is queried to obtain the second power increment .

[0048] The target Soc takes 60%, The value is obtained by table lookup as shown in Table 1, unit: Kw:

[0049] Table 1 ISG second target power increment and second power increment mapping table

[0050]

[0051] Further, according to the power limit value of the ISG motor, the ISG first target power increment and the ISG second target power increment, the ISG final demand power is calculated, including:

[0052] According to the first target power increment of the ISG motor And the second target power increment , the ISG final demand power Is calculated:

[0053] ;

[0054] According to the minimum available power of the ISG motor , the maximum available power Limit the value of the final demand power The size of the value is limited to , ].

[0055] If the ISG final demand power Is greater than the maximum available power of the ISG motor , it is determined that the value of the ISG final demand power The maximum available power of the ISG motor ;

[0056] If the ISG final demand power Is less than the minimum available power of the ISG motor , it is determined that the value of the ISG final demand power The minimum available power of the ISG motor .

[0057] Further, updating the second predicted power of the battery according to the final demand power of the ISG, comprising:

[0058] updating the second predicted power of the battery according to the final demand power of the ISG :

[0059] wherein the final demand power is , the predicted swing motor power.

[0060] Further, calculating the target speed reduction of the electric swing motor according to the second predicted power of the battery, comprising:

[0061] when the second predicted power is greater than the limit discharge power of the battery, the current electric swing power is correspondingly reduced, and the target speed reduction of the electric swing motor is ;

[0062] when the second predicted power is less than the limit charge power of the battery, the current electric swing power is correspondingly reduced, and the target speed reduction of the electric swing motor is ;

[0063] when the second predicted power is greater than the limit charge power of the battery and less than the limit discharge power of the battery, the target speed reduction of the electric swing motor is 0;

[0064] further calculating the speed value that the current electric swing needs to reduce as

[0065] wherein T is the current torque of the electric swing motor.

[0066] Further, converting the control signals of the ISG motor and the electric swing mechanism according to the final demand power of the ISG and the target speed reduction of the electric swing mechanism, comprising:

[0067] calculating the target speed of the electric swing mechanism according to the target speed reduction of the electric swing mechanism, and generating a control signal according to the target speed of the electric swing mechanism to drive the electric swing mechanism to output a corresponding speed;

[0068] calculating the control torque of the ISG according to the final demand power of the ISG divided by the current speed of the ISG motor, and generating a control signal according to the control torque of the ISG to drive the ISG motor to output a corresponding control torque.

[0069] Further, calculating the target speed of the electric swing mechanism according to the target speed reduction of the electric swing mechanism, comprising:

[0070] calculating the target speed of the electric swing mechanism according to the target speed reduction of the electric swing mechanism Computing a target rotation speed of an electric slewing mechanism .

[0071] .

[0072] wherein is an initial rotation speed request of the slewing motor calculated according to the handle opening degree.

[0073] In a second aspect, the present application provides a power control system suitable for a hybrid excavator, comprising:

[0074] a data perception layer configured to collect and pre-process time series data of the handle opening degree, the slewing motor speed, the slewing motor torque, the slewing motor bus voltage and the slewing motor bus current;

[0075] an LSTM network prediction layer connected to the data perception layer, configured to input the pre-processed time series data into a trained LSTM neural network model to predict a slewing motor power prediction value at a future time point;

[0076] a power control layer connected to the LSTM network prediction layer, configured to predict a battery power according to the slewing motor power prediction value, to calculate a final demand power of an ISG according to the battery power, and to calculate a target slewing mechanism rotation speed decrement according to whether the battery prediction power exceeds a power limit value;

[0077] an execution layer connected to the power control layer, configured to convert the final demand power of the ISG and the target slewing mechanism rotation speed decrement into corresponding ISG motor and electric slewing mechanism control signals.

[0078] Further, the data perception layer comprises:

[0079] a sensor unit configured to collect original signals of the handle opening degree, the slewing motor speed, the slewing motor torque, the slewing motor bus voltage and the slewing motor bus current;

[0080] a pre-processing unit configured to perform timestamp alignment, filtering and normalization processing on the original signals.

[0081] Further, the LSTM network prediction layer comprises a trained LSTM neural network model, the input of which is multivariate time series data within a past time window, and the output of which is a slewing motor power prediction value at a future specified time point.

[0082] Further, the LSTM neural network model comprises an input layer, a first LSTM layer, a second LSTM layer and an output layer connected in sequence.

[0083] The input layer is used to receive input data in the form of (Ts, F), where Ts is the time step and F is the feature dimension;

[0084] The first LSTM layer is used to extract the time sequence features of the input data, and the number of memory cells is 64;

[0085] The second LSTM layer is used to further analyze the time sequence features and output the feature vector of the final time step, and the number of memory cells is 32;

[0086] The output layer is a fully connected layer, which is used to map the feature vector to the power prediction value of the rotary motor.

[0087] Compared with the prior art, the present application has the following beneficial effects:

[0088] 1. Prospective control: through the precise prediction of the power of the rotary motor by the LSTM network, the power demand of the battery is predicted in advance, realizing the transition from "passive response" to "active adjustment".

[0089] 2. Double regulation mechanism: by calculating the first power increment, the battery power is stabilized in the ideal interval, and by calculating the second power increment, the battery SOC is adjusted to the target value, which effectively smooths the power and SOC fluctuations.

[0090] 3. Safety protection: by calculating the speed decrement of the electric rotary, when the battery power is about to exceed the limit safety range, the power of the electric rotary mechanism is controlled, providing the last safety barrier for the system.

[0091] 4. Prolonging the service life of the battery: through the above-mentioned comprehensive control strategy, the current stress and cycle times of the battery charging and discharging are significantly reduced, thereby effectively prolonging the service life of the battery and reducing the total cost of ownership (TCO) of the user. BRIEF DESCRIPTION OF DRAWINGS

[0092] Figure 1 is the power architecture schematic diagram of the hybrid excavator in the embodiment of the present application.

[0093] Figure 2 is the power control system schematic diagram of the hybrid excavator provided by the embodiment of the present application.

[0094] Figure 3 is the architecture schematic diagram of the LSTM neural network model in the embodiment of the present application.

[0095] Figure 4 is the internal structure schematic diagram of the LSTM memory cell in the embodiment of the present application. DETAILED DESCRIPTION

[0096] The application will be further described below with reference to the drawings. The following examples are only used to more clearly illustrate the technical solutions of the application, and cannot be used to limit the protection scope of the application.

[0097] Example 1

[0098] This embodiment describes in detail a power control method suitable for a hybrid excavator, the flow of the method is as shown in Figure 2 , applied to the hybrid excavator power architecture as shown in Figure 1 , specifically comprising the following steps:

[0099] S101: Collect and pre-process time series data.

[0100] Through the sensors installed on the whole vehicle, the original signals of the handle opening (unit), the swing motor speed (unit rpm), the swing motor torque (unit N•m), the swing motor bus voltage (unit V), and the swing motor bus current (unit A) are collected.

[0101] The original signals are pre-processed, including:

[0102] a) Time stamp alignment: Ensure that all sensor data are synchronized in time series.

[0103] b) Filtering: A first-order low-pass filter is used to filter each data sequence to remove high-frequency noise.

[0104] c) Normalization: The Min-Max normalization algorithm is used to map each feature value to the interval [0, 1].

[0105] S102: Predict the swing motor power.

[0106] The pre-processed time series data is input into a pre-trained LSTM neural network model to predict the swing motor power at a future time (e.g. 1 second later).

[0107] The structure of the LSTM neural network model is as shown in Figure 3 , including an input layer, a first LSTM layer (64 units), a second LSTM layer (32 units), and an output layer (linear activation function). The training of the model uses historical data as samples, and the mean square error (MSE) as the loss function, to learn the mapping relationship from the input sequence to the future power value. The internal calculation process of the LSTM memory cell is as shown in Figure 4 .

[0108] S103: Estimate the first predicted power of the battery.

[0109] The future swing motor power predicted by the LSTM According to the principle of equal power on the bus, the first predicted power of the battery can be calculated as

[0110] wherein is the current ISG motor power, which can be calculated by multiplying the bus voltage and current.

[0111] S104: Calculate the power increment of the ISG motor.

[0112] This step includes the following sub-steps:

[0113] S1041: Calculate the power limit value.

[0114] 1. According to the battery SOC and temperature, calculate the ideal charging power limit value of the battery , the ideal discharging power limit value , the maximum charging power limit value , the maximum discharging power limit value . The charging power is negative, and the discharging power is positive.

[0115] The calculation of the above values should refer to the battery charge and discharge power MAP. For a certain battery at SOC 60% and temperature 25℃, the calculated ideal charging power limit value is -50Kw, the ideal discharging power is 50Kw, the maximum charging power limit value is -100Kw, and the maximum discharging power limit value is 100Kw.

[0116] 2. Calculate the maximum available power , the minimum available power of the ISG motor according to the hydraulic system power and engine power limit value. The maximum available power is positive, and the power is positive, indicating driving; the minimum available power is negative, and the power is negative, indicating power generation.

[0117] In the power series of the engine, ISG motor and hydraulic system, the power balance relationship is

[0118] ;

[0119] is the current hydraulic system power, which can be calculated by measuring the pressure and flow of the hydraulic pump, and is positive power.

[0120] is the current engine power, and its power range is , which is related to the engine external characteristic.

[0121] Therefore, the minimum available power of the ISG motor when working normally is , and the maximum available power is ;

[0122] Pmin is the minimum power of ISG motor (negative value when generating electricity), Pmax is the maximum power of ISG motor (positive value when driving), both values depend on the external characteristics of the motor.

[0123] S105: Calculate the additional power increment of ISG motor.

[0124] According to the difference between the first predicted power of battery and the ideal charge-discharge power, adjust the first target power increment of ISG motor , by adjusting the power of ISG motor to limit the future power of battery within the ideal charge-discharge power range as much as possible.

[0125] When between the ideal charge power limit and the ideal discharge power limit, that is, , the first target power increment of ISG motor , ISG motor maintains the current power output.

[0126] When exceeds the ideal discharge power limit, that is, , the first target power increment of ISG motor , is negative, ISG motor reduces the current power output.

[0127] When is lower than the ideal charge power limit, that is, , the first target power increment of ISG motor , is positive, ISG motor increases the current power output.

[0128] According to the difference between the current Soc of battery and the target Soc, adjust the second target power increment of ISG motor , by adjusting the power of motor to control the SOC of battery around the target SOC value.

[0129] The basis for calculating the SOC increment is that when Soc is higher than target Soc, at this time is positive, ISG motor increases the current power output, less electricity generation or more power consumption; when Soc is lower than target Soc, at this time is negative, ISG motor reduces the current power output, more electricity generation or less power consumption.

[0130] S106: Calculate and limit the final demand power of ISG.

[0131] According to the first target power increment of ISG motor and the second target power increment of ISG motor , calculate the final demand power of ISG . And according to the minimum available power of ISG motor , the maximum available power , limit the value.

[0132] Demand power ;

[0133] And limit the size of the value to ].

[0134] S107: Determine and calculate the speed reduction.

[0135] This step is a safety protection step, including:

[0136] According to the final demand power of ISG, update the second predicted power of battery .

[0137]

[0138] When the rotary motor is driving and braking, there will be peak discharge and charge power, which may cause the second predicted power of the battery to exceed the limit of the battery's charge and discharge power. At this time, the speed of the rotary motor should be limited accordingly, thereby reducing the electric rotary power.

[0139] When greater than the battery limit discharge power Or less than the battery limit charge power At this time, the current electric rotary power should be reduced accordingly, and the reduction value is , Or , and then calculate the speed value that needs to be reduced for the current electric rotary as .

[0140] According to the target speed reduction of the electric rotary mechanism Calculate the target speed of the electric rotary mechanism .

[0141] .

[0142] In the formula The initial speed request of the rotary motor is calculated according to the handle opening, which is usually in a linear proportional relationship with the opening.

[0143] S108: Generate and send control signals.

[0144] According to the final demand power of the ISG motor obtained in step S106 and the target speed decrement obtained in step S107, a corresponding control signal (such as a CAN message) is generated and sent to the ISG motor controller and the electric slewing motor controller respectively for execution, so as to achieve smooth control of the battery power and SOC.

[0145] Embodiment 2

[0146] The embodiment provides a power control system suitable for a hybrid excavator, which is used to execute the method in embodiment 1.

[0147] The power control system of the hybrid excavator is suitable for a power architecture of a hybrid excavator, as shown in the figure. Figure 1 Based on the power architecture, the power control system mainly includes the following contents:

[0148] 1. Data sensing layer

[0149] Through the sensors installed on the whole vehicle, the handle opening degree (unit), the slewing motor speed (unit rpm), the slewing motor torque (unit N•M), the slewing motor bus voltage (unit V) and the slewing motor bus current (A) are collected, and the data is filtered and normalized.

[0150] 2. LSTM network prediction layer

[0151] The data processed by the data sensing layer is provided to the trained LSTM network, and the LSTM network performs electric slewing power prediction according to the time series data and outputs the power prediction value.

[0152] 3. Power control layer

[0153] Through the predicted electric slewing power, the battery power is further predicted, in order to realize the control of the battery charging and discharging power and the SOC in the ideal interval, the ISG motor power increment is calculated, and the electric slewing mechanism target speed decrement is calculated according to whether the battery predicted power exceeds the allowed limit power.

[0154] 4. Execution layer

[0155] According to the first and second ISG motor target power increments and the electric slewing mechanism target speed decrement output by the power control layer, the corresponding ISG motor and electric slewing mechanism control signals are converted.

[0156] The data sensing layer includes:

[0157] The sensor unit is used to collect the original signals of the handle opening degree, the slewing motor speed, the slewing motor torque, the slewing motor bus voltage and the slewing motor bus current;

[0158] A preprocessing unit is configured to perform timestamp alignment, filtering and normalization on the original signal.

[0159] The LSTM network prediction layer comprises a trained LSTM neural network model, an input of which is multivariate time series data in a past time window, and an output of which is a power prediction value of the rotary machine at a specified time in the future.

[0160] The LSTM neural network model comprises an input layer, a first LSTM layer, a second LSTM layer and an output layer connected in sequence.

[0161] The input layer is configured to receive input data in a shape of (Ts, F), where Ts is a time step, and F is a feature dimension.

[0162] The first LSTM layer is configured to extract time sequence features of the input data, and has 64 memory cells.

[0163] The second LSTM layer is configured to further analyze the time sequence features and output a feature vector of a final time step, and has 32 memory cells.

[0164] The output layer is a fully connected layer, configured to map the feature vector to a power prediction value of the rotary machine.

[0165] Embodiment 3

[0166] The embodiment provides a power control method suitable for a hybrid excavator.

[0167] The power control system of the hybrid excavator is designed for a power architecture of a hybrid excavator, as shown in the figure. Figure 1 Based on the power architecture, the designed power control system mainly includes the following contents:

[0168] The data perception layer: through the sensors installed on the whole vehicle, the handle opening degree (unit), the rotary motor speed (unit rpm), the rotary motor torque (unit N•M), the rotary motor bus voltage (unit V) and the rotary motor bus current (A) are collected, and the data are filtered and normalized.

[0169] The LSTM network prediction layer: the data processed by the data perception layer are provided to the trained LSTM network, the LSTM network performs power prediction of the electric rotary according to the time sequence data, and outputs the power prediction value.

[0170] The power control layer: through the predicted electric rotary power, the battery power is predicted, in order to realize the control of the battery charging and discharging power and the SOC in the ideal interval, the ISG motor power increment is calculated, and the electric rotary mechanism target speed decrement is calculated according to whether the battery predicted power exceeds the allowed limit power.

[0171] Execution layer: Based on the final power demand of the ISG output by the power control layer and the reduction of the target speed of the electric rotary mechanism, it converts the power demand of the ISG into corresponding control signals for the ISG motor and electric rotary mechanism.

[0172] Based on the above power control system, the present invention designs a power control method suitable for hybrid excavators, which mainly includes the following steps:

[0173] S1: Sensor data acquisition and preprocessing.

[0174] 1. Data Collection

[0175] Using sensors installed on the vehicle, five physical features were collected: handle opening (%), rotary motor speed (rpm), rotary motor torque (N•M), rotary motor bus voltage (V), and rotary motor bus current (A). The time series of each feature was collected as a dataset.

[0176] Calculate the electric rotation power for each row of data in the dataset, and calculate the electric power of the electric rotation mechanism at the current moment based on the motor bus voltage and current.

[0177] 2. Pretreatment

[0178] a) Align timestamps to ensure that the data collected by the sensors is consistent over time.

[0179] b) Eigenvalue filtering: First-order low-pass filtering is performed on the time series data of each sensor to remove high-frequency noise.

[0180] c) Eigenvalue normalization: In order to eliminate the influence of different units and dimensions of eigenvalues, the collected sensor data is normalized by Min-Max, and the range of each eigenvalue is limited to [0 1].

[0181] S2: Construction of LSTM network training samples

[0182] Predict the power value for a future specified time period based on sensor data from the past.

[0183] 1. Input sample definition:

[0184] The time window width is =60, corresponding to a 6-second time window, sampling frequency. It is 10Hz.

[0185] Feature Dimension =5, corresponding to the 5 physical quantities collected above.

[0186] The dimension of each input sample is ( , ).

[0187] The time window sliding step is = 10, assuming the total amount of data collected is seconds, then the total amount of samples N is

[0188] ;

[0189] 2. Output sample definition:

[0190] In order to realize the prediction of future power values, the data set of the electrical slewing power of the future P steps is taken as the output sample based on the time stamp of the sample. Due to the working characteristics of the excavator slewing, the electrical slewing power changes greatly in an instant, so the prediction time should not be too long, and it is recommended to be 1s.

[0191] S3: LSTM network design and training

[0192] The LSTM network structure is shown in the figure.

[0193] 1. Input layer (Input Layer)

[0194] The function of this input layer is to receive the sequence data of the above-mentioned five physical quantities collected by the original sensor, without any calculation. The shape of this input layer is (N, T, F), and the dimension of each time input is F.

[0195] 2. First layer LSTM (LSTM Layer 1)

[0196] The main function of this layer LSTM is to extract the short-term and long-term dependence between each physical quantity and the fault, and extract the time sequence features. The number of LSTM memory cells is designed to be 64.

[0197] The internal structure of each LSTM memory cell is shown in Figure 4 . For each time step, the hidden state and memory cell are calculated step by step, and the internal calculation formula is as follows:

[0198] Forget gate:

[0199] Input gate:

[0200] Candidate memory cell:

[0201] Update memory cell:

[0202] Output gate:

[0203] ​​Update the hidden state:

[0204] where is the sigmoid activation function;

[0205] is the tanh activation function;

[0206] are trainable parameter matrices and biases;

[0207] element-wise multiplication;

[0208] This layer keeps the output of each time step for the lower layer, and the final output at each time is , which has a shape of (1, 32). .

[0209] 3. Second LSTM layer (LSTM Layer 2)

[0210] This layer of LSTM receives the output of the first layer and further analyzes and decides the time series features. The number of memory cells in the LSTM layer is designed to be 32.

[0211] The internal memory cell calculation formula of this layer of LSTM is the same as that of the first layer of LSTM.

[0212] This layer only keeps the output of the last time step for the lower layer, and the final output at each time is , which has a shape of (1, 32).

[0213] 4. Output layer

[0214] The role of this layer is to map the features extracted by the LSTM layer to the power space of the rotary motor. The number of units is designed to be 1, representing the predicted power value of the rotary motor at the future 1s time point. The calculation formula of this layer is:

[0215] .

[0216] where is the linear activation function, which is suitable for continuous values such as power values.

[0217] are the trainable parameter matrices and biases of this layer.

[0218] 5. Loss function

[0219] During the sample training phase, the training target is to determine the optimization direction of the network model through the loss function. The mean square error function is used as the loss function, which is beneficial to handle large errors and capture peak power, etc., and is defined as follows:

[0220] ;

[0221] wherein, is the number of samples; is the output for training; is the output predicted by the network;

[0222] In the training process, the parameter matrix and bias of each layer of the network are optimized by iteration of samples, such as selecting Adam, so that tends to 0.

[0223] S5: First predicted power of battery calculation

[0224] Future power of rotary motor predicted by LSTM According to the principle of equal power on the bus, the first predicted power of the battery can be calculated as

[0225] ;

[0226] wherein is the current ISG motor power, which can be calculated by multiplying the bus voltage and current of the motor.

[0227] S6: Power limit value calculation

[0228] According to the battery SOC and temperature, the ideal charging power limit value , ideal discharging power limit value , maximum charging power limit value , maximum discharging power limit value of the battery are calculated. The charging power is negative, and the discharging power is positive.

[0229] The calculation of the above values should refer to the charge and discharge power MAP of the battery. When the SOC of a certain battery is 60% and the temperature is 25℃, the ideal charging power limit value calculated is-50Kw, the ideal discharging power is 50Kw, the maximum charging power limit value is-100Kw, and the maximum discharging power limit value is 100Kw.

[0230] According to the hydraulic system power and engine power limit value, the maximum available power , minimum available power of the ISG motor are calculated. The maximum available power is positive, and the power is positive, indicating driving; the minimum available power is negative, and the power is negative, indicating power generation.

[0231] In the power series of the engine, ISG motor and hydraulic system, the power balance relationship is

[0232] ;

[0233] The current hydraulic power, which can be calculated by measuring the pressure and flow of the hydraulic pump, is the positive power.

[0234] The current engine power, which is in the range of , is related to the engine external characteristics.

[0235] The minimum available power of the ISG motor when it is working normally is thus calculated as , and the maximum available power is ;

[0236] The minimum power of the ISG motor (negative value when generating electricity) is , and the maximum power of the ISG motor (positive value when driving) is related to the external characteristics of the motor.

[0237] S7: ISG first target power increment calculation

[0238] The first target power increment of the ISG motor is adjusted according to the difference between the first predicted power of the battery and the ideal charge-discharge power , so as to limit the future power of the battery to the ideal charge-discharge power range as much as possible by adjusting the power of the ISG motor.

[0239] When is between the ideal charge power limit and the ideal discharge power limit, i.e. , the first ISG motor target power increment , and the ISG motor maintains the current power output.

[0240] When exceeds the ideal discharge power limit, i.e. , the first ISG motor target power increment , is negative, and the ISG motor reduces the current power output.

[0241] When is lower than the ideal charge power limit, i.e. , the first ISG motor target power increment , is positive, and the ISG motor increases the current power output.

[0242] S8: ISG second target power increment calculation

[0243] The second target power increment of the ISG motor is adjusted according to the difference between the current Soc of the battery and the target Soc , so as to control the battery SOC around the target SOC value by adjusting the power of the motor.

[0244] The basis for calculating the SOC increment is that when the Soc is higher than the target Soc, the value of is positive, the ISG motor increases the current power output, and less power is generated or more power is consumed; when the Soc is lower than the target Soc, the value of is negative, the ISG motor reduces the current power output, and more power is generated or less power is consumed.

[0245] The target Soc can be 60%, The value can be obtained by referring to Table 1, and the unit is Kw:

[0246] Table 1 Second ISG target power increment and second power increment mapping table

[0247]

[0248] S9: ISG final demand power calculation

[0249] According to the first target power increment of the ISG motor and the second target power increment , the ISG final demand power is calculated. And according to the minimum available power of the ISG motor , the maximum available power is limited.

[0250] The demand power ;

[0251] And the size of the value is limited to ].

[0252] S10: Second battery prediction power calculation

[0253] According to the final demand power of the ISG, the second prediction power of the battery is updated .

[0254]

[0255] S11: Rotary motor target speed decrement calculation

[0256] When the rotary motor is driving and braking, there will be peak discharge and charge power, which may cause the second prediction power of the battery to exceed the limit of the battery charge and discharge power, at which time the speed of the rotary motor should be limited accordingly, thereby reducing the electric rotary power.

[0257] When greater than the battery limit discharge power or less than the battery limit charge power At this time, the current electric slewing power should be reduced accordingly, and the reduction value is , or , and then the current electric slewing speed value to be reduced is calculated .

[0258] S12: final demand speed calculation of slewing motor

[0259] According to the target slewing speed reduction of the electric slewing mechanism Calculate the target slewing speed of the electric slewing mechanism .

[0260] .

[0261] In the formula is the initial slewing motor speed request calculated according to the handle opening, which is usually in linear proportional relationship with the opening.

[0262] The power of the slewing motor is predicted through the LSTM network, and then the future power of the power battery is predicted. The power of the ISG motor is adjusted to control the power battery power and SOC in the ideal interval, reduce the battery charging and discharging, and is conducive to prolonging the battery life.

[0263] The parameters in the LSTM network designed in this paper are only one embodiment, and changing other layers such as the number of units, the number of layers, etc. also belongs to the scope of the present technology.

[0264] The power regulation method of the ISG in this paper is only one embodiment, and changing its data or using other formulas also belongs to the scope of the present technology.

[0265] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can 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 code.

[0266] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks. Figure 1 one or more flow or blocks.

[0267] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks. Figure 1 one or more flow or blocks.

[0268] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks. Figure 1 one or more flow or blocks.

[0269] The above only is the preferred embodiment of the present application, it should be pointed out that, for those skilled in the technical field, without departing from the technical principles of the present application, can also make a number of improvements and variations, these improvements and variations should also be considered as the protection scope of the present application.

Claims

1. A power control method suitable for hybrid excavators, characterized in that, The method includes: Collect and preprocess time series data; The preprocessed time series data is input into a trained LSTM neural network model to predict the power of the rotary motor at a future time. The input of the LSTM neural network model is multivariate time series data within a past time window, and the output is the predicted power of the rotary motor at a specified future time. The battery power is predicted based on the predicted power value of the rotary motor, the final required power of the ISG is calculated based on the battery power, and the target speed reduction of the electric rotary mechanism is calculated based on whether the predicted battery power exceeds the power limit. Based on the final power requirement of the ISG and the reduction in the target speed of the electric rotary mechanism, the corresponding control signals for the ISG motor and electric rotary mechanism are converted.

2. The power control method for hybrid excavators according to claim 1, characterized in that, The battery power is predicted based on the predicted power value of the rotary motor. The final power requirement of the ISG is calculated based on the battery power. The target speed reduction of the electric rotary mechanism is calculated based on whether the predicted battery power exceeds the power limit, including: Calculate the first predicted power of the battery based on the predicted power value of the rotary motor; Obtain the power limit of the ISG motor; Based on the first predicted power of the battery, calculate the first target power increment of the ISG and the second target power increment of the ISG. Calculate the final power requirement of the ISG based on the power limit of the ISG motor, the first target power increment of the ISG, and the second target power increment of the ISG. Update the second predicted power of the battery based on the final power demand of the ISG; The reduction in the target speed of the rotary motor is calculated based on the second predicted power of the battery.

3. The power control method for hybrid excavators according to claim 1, characterized in that, The time series data includes time series data of handle opening degree, rotary motor speed, rotary motor torque, rotary motor bus voltage, and rotary motor bus current; The collection and preprocessing of time series data includes: The raw signals of handle opening degree, rotary motor speed, rotary motor torque, rotary motor bus voltage and rotary motor bus current are collected by sensors. The original signal is then subjected to timestamp alignment, filtering, and normalization.

4. The power control method for hybrid excavators according to claim 1, characterized in that, The training method for the LSTM neural network model includes: Construct a training dataset, wherein the input samples of the dataset are the time series data within a past time window, and the output samples are the actual rotary motor power values ​​at a specified future time. Using mean squared error as the loss function, Adam is selected to optimize the parameter matrix and bias of each layer of the network, and the constructed LSTM neural network model is trained.

5. The power control method for hybrid excavators according to claim 2, characterized in that, The first predicted power of the battery is calculated based on the predicted power value of the rotary motor, including: ； in, The first predicted power of the battery, For the predicted rotary motor power, This represents the current power of the ISG motor.

6. The power control method for hybrid excavators according to claim 5, characterized in that, Obtain the power limits for the ISG motor, including: Based on the battery's SOC and temperature, and referring to the battery's charge / discharge power MAP, the ideal charging power limit of the battery is calculated. Ideal discharge power limit Maximum charging power limit Maximum discharge power limit The charging power is negative, and the discharging power is positive. The maximum available power of the ISG motor is calculated based on the hydraulic system power and engine power limits. Minimum available power The maximum available power is a positive value, indicating that the power is driving; the minimum available power is a negative value, indicating that the power is generating. Calculate the minimum available power of the ISG motor during normal operation. The maximum available power is ; in, This is the minimum power of the ISG motor. These two values ​​represent the maximum power of the ISG motor and depend on the motor's external characteristics. The current power of the hydraulic system is calculated by measuring the pressure and flow rate of the hydraulic pump, and is a positive power. This represents the current engine power, and its power range is... .

7. The power control method for hybrid excavators according to claim 6, characterized in that, Calculating the first target power increment of the ISG includes: Based on the first predicted power of the battery With respect to the ideal charging power limit of the battery and ideal discharge power limit The relationship is used to calculate the first power increment. The relationship is as follows: when When the ideal charging power limit is between the ideal discharging power limit and the ideal charging power limit, i.e. The first ISG motor target power increment The ISG motor maintains its current power output; when When the ideal discharge power limit is exceeded, i.e. The first ISG motor target power increment , When the value is negative, the ISG motor reduces its current power output; when When the charging power is below the ideal charging power limit, i.e. The first ISG motor target power increment , When the value is positive, the ISG motor increases its current power output; Calculating the second target power increment of the ISG includes: Based on the difference between the battery's current SOC and the target SOC, a preset mapping table is consulted to obtain the second power increment. .

8. The power control method for hybrid excavators according to claim 7, characterized in that, Based on the power limit of the ISG motor, the first target power increment of the ISG, and the second target power increment of the ISG, calculate the final power requirement of the ISG, including: Based on the first target power increment of the ISG motor Second target power increment Calculate the final power demand of the ISG : ； Based on the minimum available power of the ISG motor Maximum available power Limit this value: reduce the final power demand. The value is limited to [ , ].

9. The power control method for hybrid excavators according to claim 1, characterized in that, Based on the ISG's final power demand, the second predicted power of the battery is updated, including: The second predicted power of the battery is updated based on the final power demand of the ISG. : ；； in, For the final required power, The predicted power of the rotary motor; Based on the battery's second predicted power, the reduction in the target speed of the rotary motor is calculated, including: When the second predicted power is greater than the battery's maximum discharge power At that time, the current electric rotary power will be reduced accordingly, and the target speed of the rotary motor will decrease by [amount missing]. ; When the second predicted power is less than the battery's maximum charging power At that time, the current electric rotary power will be reduced accordingly, and the target speed of the rotary motor will decrease by [amount missing]. ; When the second predicted power is greater than the battery's maximum charging power and less than the battery's maximum discharging power. At that time, the target speed of the rotary motor decreases. 0; Calculate the required reduction in rotational speed for the current electric rotary valve. ; Where T is the current torque of the rotary motor.

10. A power control system suitable for hybrid excavators, characterized in that, include: The data sensing layer is used to collect and preprocess time-series data of handle opening, rotary motor speed, rotary motor torque, rotary motor bus voltage, and rotary motor bus current. The LSTM network prediction layer, connected to the data perception layer, is used to input the preprocessed time series data into the trained LSTM neural network model to predict the power prediction value of the rotary motor at future times. The power control layer, connected to the LSTM network prediction layer, is used to predict the battery power based on the predicted power value of the rotary motor, calculate the final power requirement of the ISG based on the battery power, and calculate the target speed reduction of the electric rotary mechanism based on whether the predicted battery power exceeds the power limit. The execution layer, connected to the power control layer, is used to convert the final power demand of the ISG and the target speed reduction of the electric rotary mechanism into corresponding control signals for the ISG motor and electric rotary mechanism.

Images