Hybrid excavator power system output control method and device and hybrid excavator

Abstract

The invention discloses a hybrid excavator power system output control method and device and a hybrid excavator, the hybrid excavator comprises an engine, an ISG motor and a hydraulic pump which are coaxially connected, the ISG motor is electrically connected with a power battery, a vehicle control unit is electrically connected with the engine, the ISG motor, the hydraulic pump and the power battery, and the method comprises the steps that operation parameters of the hybrid excavator are collected; the collected operation parameters are preprocessed and input into a torque prediction model based on an RBF neural network, and a torque prediction value of the hybrid excavator power system is obtained; according to the idle speed button triggering time, the battery SOC and the torque predicted value, working condition recognition is conducted, and the optimal torque working point of the engine is obtained; and performing output control on the power system of the hybrid excavator based on the optimal torque working point of the engine. The power output of the engine and the motor of the hybrid excavator can be effectively coordinated, the energy utilization efficiency and the working efficiency are improved, the fuel consumption is reduced, and the use cost is reduced.

Description

Technical Field

[0001] This invention belongs to the field of hybrid excavator technology, specifically relating to a hybrid excavator power system output control method, device, and hybrid excavator. Background Technology

[0002] Because large excavators have high power requirements, varied working conditions, and often operate continuously, the transition to pure electric power is difficult, which promotes the transformation of large excavators into hybrid excavators.

[0003] The engine control system and methods of hybrid excavators are a key focus and challenge in current technological transformation, directly impacting the vehicle's energy consumption and work efficiency. Currently, most hybrid engine control technologies employ series, parallel, and hybrid architectures. In a series architecture, the engine only drives the generator, resulting in significant energy conversion losses and high energy consumption at high speeds. In a parallel architecture, both the engine and drive motor can drive the load independently or jointly, combining power through a mechanical coupling device. The engine can directly drive the load, achieving high energy utilization and strong power superposition. In a hybrid architecture, the engine can both generate electricity and directly drive the load, often employing a planetary gear power distribution device with intelligent switching modes, providing efficient coverage across all working conditions and achieving optimal fuel consumption and power balance. Due to the compact space distribution and complex operating conditions of large excavators, a parallel hybrid architecture is the most suitable choice. During operation, a significant drop in input speed or insufficient input torque of the hydraulic pump can severely affect the vehicle's operation and increase fuel consumption.

[0004] Hybrid excavators are characterized by large and frequent load changes. How to coordinate the power output of the engine and motor of a hybrid excavator and improve energy utilization efficiency is an urgent problem to be solved. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method, device, and hybrid excavator for controlling the power system output of a hybrid excavator. This method effectively coordinates the power output of the engine and motor of the hybrid excavator, improving energy utilization efficiency and work efficiency, saving fuel consumption, and reducing operating costs.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] Firstly, a method for output control of a hybrid excavator's power system is provided. The hybrid excavator includes an engine, an ISG motor, and a hydraulic pump coaxially connected. The ISG motor is electrically connected to a power battery. A vehicle controller is electrically connected to the engine, ISG motor, hydraulic pump, and power battery. The method is executed by the vehicle controller and includes: acquiring operating parameters of the hybrid excavator; preprocessing the acquired operating parameters and inputting them into a torque prediction model based on an RBF neural network to obtain a torque prediction value for the hybrid excavator's power system; identifying the operating condition based on the idle speed button trigger time, battery SOC, and the torque prediction value to obtain the engine's optimal torque operating point; and performing output control of the hybrid excavator's power system based on the engine's optimal torque operating point.

[0008] Furthermore, the operating parameters of the hybrid excavator include: the output current of the hydraulic pump in the current stage and the previous stage; the opening degree of the boom raise handle, the opening degree of the boom retract handle, the opening degree of the boom raise handle, the opening degree of the boom retract handle, the opening degree of the bucket retract handle, the opening degree of the bucket tilt handle, the requested torque of the hydraulic system, the requested speed of the hydraulic system, the output speed of the power system, and the output torque of the power system.

[0009] Furthermore, the torque prediction model based on the RBF neural network is a three-layer feedforward network, including: an input layer, which receives the input vector, and the number of nodes in the input layer is equal to the dimension of the input vector; a hidden layer, which transforms the input vector received by the input layer through radial basis functions, thereby transforming the input layer data from one space to another; and an output layer, which performs linear weighted summation on the data in the hidden layer space to output the torque prediction value of the hybrid excavator power system.

[0010] Furthermore, the input vector is:

[0011] ;

[0012] in, ~ These are respectively the boom raising handle opening, boom retracting handle opening, boom raising handle opening, boom retracting handle opening, bucket retracting handle opening, bucket tilting handle opening, hydraulic system requested torque, hydraulic system requested speed, load fluctuation torque compensation value, and power system output torque.

[0013] Furthermore, the input to the torque prediction model based on the RBF neural network also includes the power system compensation torque value, which is determined according to the output current of the hydraulic pump in the current stage and the previous stage, and in accordance with the set compensation rules.

[0014] Furthermore, the set compensation rules include: calculating the difference between the output current of the hydraulic pump in the current stage and the previous stage, and obtaining a correction coefficient based on the rate of change of the output current; when the rate of change of the output current is greater than a set positive threshold or less than a set negative threshold, the correction coefficient is less than 1; when the rate of change of the output current is between the set negative threshold and the positive threshold, the correction coefficient is 1; when the difference between the output current of the hydraulic pump in the current stage and the output current of the hydraulic pump in the previous stage is greater than the product of the set positive threshold and the correction coefficient, the load fluctuation torque compensation value is positive; when the difference between the output current of the hydraulic pump in the current stage and the output current of the hydraulic pump in the previous stage is less than the product of the set negative threshold and the correction coefficient, the load fluctuation torque compensation value is negative; when the output current of the hydraulic pump in the current stage and the output current of the hydraulic pump in the previous stage are between the set negative threshold and the positive threshold, the load fluctuation torque compensation value is 0.

[0015] Furthermore, based on the idle button trigger time, battery SOC, and the predicted torque value, operating condition identification is performed to obtain the engine's optimal torque operating point, including: an idle button trigger time less than 0 seconds indicates the operating state; an idle button trigger time greater than 1 second indicates the idling state; a battery SOC of 30-40% indicates a low SOC range; a battery SOC of 40-80% indicates a medium SOC range; and a battery SOC of 80-90% indicates a high SOC range; a predicted torque value less than 900 Nm indicates a low load; a predicted torque value greater than 900 Nm and less than 1600 Nm indicates a medium load; and a predicted torque value greater than 1600 Nm indicates a high load. In the idling condition, under low load: in the high SOC range and the medium SOC range, the engine's optimal torque operating point is... Under light load conditions: In the high SOC range, the engine's optimal torque operating point is zero torque; in the medium SOC range, the engine's optimal torque operating point is low torque; in the low SOC range, the engine's optimal torque operating point is medium torque; under medium load conditions: In the high SOC range, the engine's optimal torque operating point is low torque; in the medium and low SOC ranges, the engine's optimal torque operating point is medium torque; under heavy load conditions: In the high SOC range, the engine's optimal torque operating point is medium torque; in the medium and low SOC ranges, the engine's optimal torque operating point is high torque.

[0016] Furthermore, based on the engine's optimal torque operating point, the output control of the hybrid excavator's power system is performed, including: under different operating conditions, the engine operates at the optimal torque operating point, and the output torque of the ISG motor is adaptively adjusted according to the engine torque and the torque prediction value;

[0017] ;

[0018] In the formula, This is the output reference torque of the ISG motor. This is the reference torque for the engine under different operating modes. This is the predicted torque value of the hybrid excavator's power system; under idling conditions, when the power battery's SOC is in the high SOC range, the ISG motor operates independently.

[0019] Secondly, a hybrid excavator power system output control device is provided. The hybrid excavator includes an engine, an ISG motor, and a hydraulic pump coaxially connected. The ISG motor is electrically connected to a power battery. A vehicle controller is electrically connected to the engine, ISG motor, hydraulic pump, and power battery. The device is configured in the vehicle controller and includes: a data acquisition module for acquiring the operating parameters of the hybrid excavator; a torque prediction module for preprocessing the acquired operating parameters and inputting them into a torque prediction model based on an RBF neural network to obtain the torque prediction value of the hybrid excavator power system; a working condition identification module for identifying the working condition based on the idle speed button trigger time, battery SOC, and the torque prediction value to obtain the optimal torque operating point of the engine; and an output control module for output control of the hybrid excavator power system based on the optimal torque operating point of the engine.

[0020] Thirdly, a hybrid excavator is provided, wherein the hybrid excavator is equipped with the hybrid excavator power system output control device described in the second aspect.

[0021] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: This invention obtains the torque prediction value of the hybrid excavator's power system by inputting the collected operating parameters into a torque prediction model based on an RBF neural network; it identifies the operating condition based on the idle speed button trigger time, battery SOC, and the torque prediction value to obtain the engine's optimal torque operating point; and it controls the output of the hybrid excavator's power system based on the engine's optimal torque operating point, effectively coordinating the power output of the engine and motor of the hybrid excavator, improving energy utilization efficiency and work efficiency, saving fuel consumption, and reducing operating costs. Attached Figure Description

[0022] Figure 1 This is a schematic block diagram of a hybrid excavator power system provided in an embodiment of the present invention;

[0023] Figure 2 This is a principle block diagram of a hybrid excavator power system output control method provided in an embodiment of the present invention;

[0024] Figure 3 This is a comparison chart of torque output predictions for the hybrid excavator power system in an embodiment of the present invention. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0026] Example 1

[0027] like Figures 1-3 As shown, a method for output control of a hybrid excavator's power system is disclosed. The hybrid excavator includes an engine, an ISG motor, and a hydraulic pump coaxially connected. The ISG motor is electrically connected to a power battery. A vehicle controller is electrically connected to the engine, ISG motor, hydraulic pump, and power battery. The method is executed by the vehicle controller and includes: acquiring operating parameters of the hybrid excavator; preprocessing the acquired operating parameters and inputting them into a torque prediction model based on an RBF neural network to obtain a torque prediction value for the hybrid excavator's power system; identifying the operating condition based on the idle speed button trigger time, battery SOC, and the torque prediction value to obtain the engine's optimal torque operating point; and performing output control of the hybrid excavator's power system based on the engine's optimal torque operating point.

[0028] The output configuration of the hybrid excavator's power system is as follows: Figure 1 As shown, it includes an engine and an ISG motor, which are coaxially connected. The ISG motor is connected to an external power battery, which can charge and discharge the power battery in different working modes. The motor output is connected to a hydraulic pump to directly drive the hydraulic system of the whole vehicle. The hydraulic pump controls the travel hydraulic motor, boom cylinder, bucket cylinder and arm cylinder through a hydraulic circuit to realize the excavator's operation and travel.

[0029] First, the operating parameters of the hybrid excavator during actual operation are collected. The main data include: the output current of the hydraulic pump in the current stage, the output current of the hydraulic pump in the previous stage, the opening degree of the boom raising handle, the opening degree of the boom retracting handle, the opening degree of the boom raising handle, the opening degree of the boom retracting handle, the opening degree of the bucket retracting handle, the opening degree of the bucket tilting handle, the requested torque of the hydraulic system, the requested speed of the hydraulic system, the output speed of the power system, and the output torque of the power system.

[0030] The preprocessing of the collected operating parameters includes first normalizing the collected dataset, then using it as input to train a torque prediction model based on an RBF neural network, and finally inverse normalizing the data to obtain the predicted torque value of the power system. The output power of the engine and ISG motor is allocated based on the engine's optimal torque operating point under different operating conditions. The preprocessing of the collected operating parameters also includes calculating torque compensation values ​​when there are large load fluctuations, and using these load fluctuation torque compensation values ​​as input to the RBF neural network-based torque prediction model.

[0031] S1. Load change, power system output torque compensation calculation.

[0032] The load fluctuation torque compensation value is determined by monitoring the difference between the output current of the hydraulic pump in the current stage and the output current of the hydraulic pump in the previous stage. Based on the rate of change of the output current, a correction coefficient is obtained. When the rate of change is greater than a set positive threshold or less than a set negative threshold, the rate of change is fast, and the correction coefficient should be less than 1, thus narrowing the threshold range and intervening earlier. When the rate of change is between the set negative and positive thresholds, the correction coefficient is 1, meaning intervention is performed according to the original threshold range. The load fluctuation torque compensation value should be positive when the difference between the output current of the hydraulic pump in the current stage and the output current of the hydraulic pump in the previous stage is greater than the product of the set positive threshold and the correction coefficient; it should be negative when the difference is less than the product of the set negative threshold and the correction coefficient; and it is zero when the difference is between the set negative and positive thresholds. In this embodiment, the set positive threshold can be 10, and the set negative threshold can be -10.

[0033] S2. Collect operating parameters during actual operation and calculate the predicted output torque value of the power system.

[0034] (1) Collect data on the operation of the hybrid excavator's power system, process the data, and normalize the sample data based on the boom lifting handle opening, boom retraction handle opening, boom lifting handle opening, boom retraction handle opening, bucket retraction handle opening, bucket tilting handle opening, hydraulic system requested torque, hydraulic system requested speed, power system output torque, and load fluctuation torque compensation value determined by the output current of the hydraulic pump in the current stage and the output current of the hydraulic pump in the previous stage. Then, create a working condition sample input data model.

[0035] (2) The RBF neural network is a feedforward network consisting of three layers. The first layer is the input layer, and the number of nodes is equal to the dimension of the input. The second layer is the hidden layer, and the number of neurons depends on the needs of the problem. The transformation function of this layer is the radial basis function. The third layer is the output layer, which completes the output by weighted summation of the hidden layer data.

[0036] The input layer nodes are: boom raising handle opening x1, boom retracting handle opening x2, boom raising handle opening x3, boom retracting handle opening x4, bucket retracting handle opening x5, bucket tilting handle opening x6, hydraulic system requested torque x7, hydraulic system requested speed x8, load fluctuation torque compensation value x9, and power system output torque x1. 10 Combine them into a vector with 10 nodes: As the input vector, its dimension is the number of sample data.

[0037] The hidden layer nodes consist of nodes directly connected to the input nodes, and the radial basis kernel function is a Gaussian radial basis function.

[0038] ;

[0039] In the formula, σ is the width of the Gaussian kernel.

[0040] The torque prediction model based on the RBF neural network requires a two-step mapping from input to output. First, the input layer samples are transformed using radial basis functions, thus converting the input layer data from one space to another. The output of the i-th hidden unit is:

[0041] ;

[0042] In the formula, These are the radial basis functions of the hidden layer, specifically the Gaussian radial basis functions. It is the input vector; It is the center point of the i-th neuron. It is the width of the Gaussian kernel. It is the Euclidean norm, and the sample points are... To the center point The distance.

[0043] Then, the data in the hidden layer space are linearly weighted and summed to obtain the output:

[0044] ;

[0045] In the formula, n is the number of hidden nodes. This represents the weights between the hidden layer and the output layer.

[0046] The output layer contains predicted values ​​of the power system output torque under different operating characteristics.

[0047] S3, Engine torque condition identification.

[0048] A hybrid system operating mode classification model is established based on the idle button trigger time, battery SOC, and hydraulic system torque request status as inputs. An idle button trigger time less than 0 seconds indicates the operating state, while an idle button trigger time greater than 1 second indicates the idling state. Battery SOC between 30-40% is considered low SOC, between 40-80% is medium SOC, and between 80-90% is high SOC. A predicted powertrain output torque less than 900 Nm indicates low load, greater than 900 Nm and less than 1600 Nm indicates medium load, and greater than 1600 Nm indicates high load. The optimal torque operating point allocation for the engine under different operating conditions is shown in Table 1.

[0049] Table 1: Optimal Torque Operating Point of Engine

[0050]

[0051] S4, torque output of the power system.

[0052] Based on the engine's universal characteristic curve, the optimal torque operating range of the engine under different speed conditions can be obtained. Through this optimal torque operating range, the best torque points for low, medium, and high torque of the engine are determined to be 300 Nm, 1100 Nm, and 1500 Nm, respectively. The ISG motor's output torque is adaptively adjusted based on the engine torque and the predicted output torque values ​​of the powertrain.

[0053] ;

[0054] In the formula, Provides the reference torque for the ISG motor output. This is the reference torque for the engine under different operating modes. This is the predicted output torque value of the power system.

[0055] When the hybrid excavator is idling, and the power battery SOC is at a high level (high SOC range), the ISG motor works independently during idling.

[0056] Large mining excavators operate under complex conditions and consume a lot of energy. They need to work under varying loads, sometimes performing heavy-duty excavation, and sometimes making long-distance relocations. Under these different operating conditions, the hybrid excavator's power system must consider both fuel economy and excavation efficiency. In this invention, the engine and electric motor work together. The engine provides stable and continuous power, while the ISG electric motor dynamically compensates for the power loss, assisting the engine in completing its tasks. This keeps the engine operating within a relatively efficient range. Simultaneously, the rapid response characteristics of the electric motor reduce the impact of sudden load torque changes, speed fluctuations, and battery SOC on work efficiency, thereby reducing fuel consumption and improving overall efficiency.

[0057] This invention addresses the characteristics of hybrid excavators, which experience large and frequent load variations. It employs load torque compensation, power system output torque prediction, and engine operating condition identification. By correcting the load request torque, the invention obtains the power system's advance torque compensation value under different hydraulic pump current variations. Using a torque prediction model based on an RBF neural network, the invention provides predicted power system torque values ​​under different handle openings and load requests. Based on the predicted power system torque values, idle button trigger duration, and battery SOC status, a fuzzy rule algorithm categorizes engine operating conditions into idle, light load, medium load, and heavy load conditions. Combining the optimal engine torque operating range under different conditions, the invention selects the optimal engine torque point as a benchmark, ensuring stable engine operation within the high-efficiency range. The ISG motor utilizes speed control to smooth out peak and valley loads in the power system, improving energy utilization and operating efficiency, thereby saving fuel consumption and reducing operating costs.

[0058] The load fluctuation torque compensation value in this invention can also be calculated using other feedforward control methods. The torque prediction model based on the RBF neural network also employs Markov probability theory algorithms for dynamic prediction and adjustment.

[0059] Example 2

[0060] Based on the hybrid excavator power system output control method described in Embodiment 1, this embodiment provides a hybrid excavator power system output control device. The hybrid excavator includes an engine, an ISG motor, and a hydraulic pump coaxially connected. The ISG motor is electrically connected to a power battery. The vehicle controller is electrically connected to the engine, ISG motor, hydraulic pump, and power battery respectively. The device is configured in the vehicle controller and includes:

[0061] The data acquisition module is used to collect the operating parameters of the hybrid excavator;

[0062] The torque prediction module is used to preprocess the collected operating parameters and input them into the torque prediction model based on the RBF neural network to obtain the torque prediction value of the hybrid excavator power system.

[0063] The operating condition identification module is used to identify the operating condition based on the idle speed button trigger time, battery SOC and the torque prediction value, and obtain the engine's optimal torque operating point.

[0064] The output control module is used to control the output of the hybrid excavator power system based on the engine's optimal torque operating point.

[0065] It also includes a load fluctuation torque compensation module, which is used to calculate the compensation torque value and intervene in advance when the load fluctuation is large.

[0066] Example 3

[0067] Based on the hybrid excavator power system output control device described in Embodiment 2, this embodiment provides a hybrid excavator equipped with the hybrid excavator power system output control device described in Embodiment 2.

[0068] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for output control of a hybrid excavator's power system, characterized in that, The hybrid excavator includes a coaxially connected engine, ISG motor, and hydraulic pump. The ISG motor is electrically connected to the power battery. The vehicle controller is electrically connected to the engine, ISG motor, hydraulic pump, and power battery. The method is executed by the vehicle controller and includes: Collect operating parameters of the hybrid excavator; The collected operating parameters are preprocessed and input into the torque prediction model based on the RBF neural network to obtain the torque prediction value of the hybrid excavator power system. Based on the idle speed button trigger time, battery SOC and the torque prediction value, the operating condition is identified to obtain the engine's optimal torque operating point. Based on the engine's optimal torque operating point, the output control of the hybrid excavator's power system is performed.

2. The hybrid excavator power system output control method according to claim 1, characterized in that, The operating parameters of a hybrid excavator include: The output current of the hydraulic pump in the current stage and the previous stage; The opening of the boom raising handle, boom retracting handle, boom raising handle, boom retracting handle, bucket retracting handle, bucket tilting handle, hydraulic system requested torque, hydraulic system requested speed, power system output speed, and power system output torque.

3. The hybrid excavator power system output control method according to claim 2, characterized in that, The torque prediction model based on the RBF neural network consists of a three-layer feedforward network, including: An input layer is used to receive an input vector, and the number of nodes in the input layer is equal to the dimension of the input vector; The hidden layer transforms the input vector received by the input layer through radial basis functions, thereby transforming the input layer data from one space to another. The output layer is used to perform linear weighted summation of the data in the hidden layer space and output the torque prediction value of the hybrid excavator power system.

4. The hybrid excavator power system output control method according to claim 3, characterized in that, The input vector is: ； in, ~ These are respectively the boom raising handle opening, boom retracting handle opening, boom raising handle opening, boom retracting handle opening, bucket retracting handle opening, bucket tilting handle opening, hydraulic system requested torque, hydraulic system requested speed, load fluctuation torque compensation value, and power system output torque.

5. The hybrid excavator power system output control method according to claim 4, characterized in that, The input to the torque prediction model based on the RBF neural network also includes the power system compensation torque value, which is determined according to the output current of the hydraulic pump in the current stage and the previous stage, and in accordance with the set compensation rules.

6. The hybrid excavator power system output control method according to claim 5, characterized in that, The established compensation rules include: Calculate the difference between the output current of the hydraulic pump in the current stage and the previous stage, and obtain the correction coefficient based on the rate of change of the output current; When the rate of change of the output current is greater than the set positive threshold or less than the set negative threshold, the correction coefficient is less than 1. When the rate of change of the output current is between the set negative threshold and the positive threshold, the correction factor is 1; When the difference between the hydraulic pump output current in the current stage and the hydraulic pump output current in the previous stage is greater than the product of the set positive threshold and the correction coefficient, the load fluctuation torque compensation value is positive. When the difference between the hydraulic pump output current in the current stage and the hydraulic pump output current in the previous stage is less than the product of the set negative threshold and the correction coefficient, the load fluctuation torque compensation value is negative. When the output current of the hydraulic pump in the current stage and the output current of the hydraulic pump in the previous stage are between the set negative threshold and the positive threshold, the load fluctuation torque compensation value is 0.

7. The hybrid excavator power system output control method according to claim 3, characterized in that, Based on the idle button trigger time, battery SOC, and the predicted torque value, operating conditions are identified to obtain the engine's optimal torque operating point, including: If the idle button trigger time is less than 0 seconds, the machine is in operation mode; if the idle button trigger time is greater than 1 second, the machine is in idle mode. A battery SOC of 30-40% is considered low SOC, 40-80% is considered medium SOC, and 80-90% is considered high SOC. The predicted torque value is less than 900 Nm, which indicates a low load; the predicted torque value is greater than 900 Nm and less than 1600 Nm, which indicates a medium load; and the predicted torque value is greater than 1600 Nm, which indicates a high load. Under idling conditions and low load: in the high SOC and medium SOC ranges, the engine's optimal torque operating point is zero torque; in the low SOC range, the engine's optimal torque operating point is medium torque. Under light load conditions: in the high SOC range, the engine's optimal torque operating point is zero torque; in the medium SOC range, the engine's optimal torque operating point is low torque; and in the low SOC range, the engine's optimal torque operating point is medium torque. Under medium load conditions: in the high SOC range, the engine's optimal torque operating point is the engine's low torque; in the medium SOC range and the low SOC range, the engine's optimal torque operating point is the engine's medium torque. Under heavy load conditions: in the high SOC range, the engine's optimal torque operating point is the engine's medium torque; in the medium SOC range and the low SOC range, the engine's optimal torque operating point is the engine's high torque.

8. The hybrid excavator power system output control method according to claim 7, characterized in that, Based on the engine's optimal torque operating point, output control is performed on the hybrid excavator's power system, including: Under different operating conditions, the engine operates at the optimal torque operating point, and the output torque of the ISG motor is adaptively adjusted according to the engine torque and the torque prediction value. ； In the formula, This is the output reference torque of the ISG motor. This is the reference torque for the engine under different operating modes. This refers to the predicted torque value of the hybrid excavator's power system. Under idling conditions, when the power battery SOC is in the high SOC range, the ISG motor operates independently.

9. A power system output control device for a hybrid excavator, characterized in that, The hybrid excavator includes a coaxially connected engine, ISG motor, and hydraulic pump. The ISG motor is electrically connected to the power battery. The vehicle controller is electrically connected to the engine, ISG motor, hydraulic pump, and power battery. The device is configured in the vehicle controller and includes: The data acquisition module is used to collect the operating parameters of the hybrid excavator; The torque prediction module is used to preprocess the collected operating parameters and input them into the torque prediction model based on the RBF neural network to obtain the torque prediction value of the hybrid excavator power system. The operating condition identification module is used to identify the operating condition based on the idle speed button trigger time, battery SOC and the torque prediction value, and obtain the engine's optimal torque operating point. The output control module is used to control the output of the hybrid excavator power system based on the engine's optimal torque operating point.

10. A hybrid excavator, characterized in that, The hybrid excavator is equipped with the hybrid excavator power system output control device as described in claim 9.

Images