MHC intelligent hybrid power system and mobile port cranes using the system

Abstract

This invention discloses an MHC intelligent hybrid power system and a mobile port crane using the system. The hybrid power system includes an internal combustion engine generator set, a lithium battery energy storage device, a drive unit, an information acquisition unit, and an energy management controller. The lithium battery energy storage device and the drive unit are connected to the DC bus through a bidirectional DC / DC converter and a four-quadrant inverter, respectively. The energy management controller adopts a management architecture that combines outer-loop rolling prediction and inner-loop sliding mode control. When the MHC is working, the system predicts the load power demand in the future time domain based on the current load information and selects the corresponding working mode. This allows for the rational allocation of power between the internal combustion engine and the lithium battery pack during heavy load phases, ensuring that the internal combustion engine is always in the optimal fuel economy zone, reducing fuel consumption and emissions, and improving energy efficiency. Alternatively, during load reduction or braking phases, the lithium battery pack can recover and store the generated energy, reducing battery capacity and cost.

Description

Technical Field

[0001] This invention belongs to the field of port crane technology, specifically relating to the MHC intelligent hybrid power system and a mobile port crane using the system. Background Technology

[0002] With increasing global emphasis on energy conservation, emission reduction, and green port construction, traditional mobile port cranes (MHCs) powered solely by diesel engines are facing the dual pressures of increasingly stringent environmental regulations and rising operating costs. On the one hand, port operations are typically located near urban areas, and exhaust emissions (NOx, PM) and noise pollution directly impact the quality of life for surrounding residents. On the other hand, fluctuating crude oil prices and increasingly stringent IMO and EU Stage V emission standards are causing a continuous decline in the economic viability of traditional diesel-powered systems. Therefore, finding a power solution that balances low emissions, low energy consumption, and low operating costs while maintaining the high mobility and lifting capacity of MHCs has become an industry consensus.

[0003] In recent years, hybrid power technology for construction machinery has experienced rapid development from proof-of-concept to industrial demonstration. The "oil-electric" parallel / parallel hybrid architecture, initially applied in the automotive field, has been gradually adapted to port machinery, mining machinery, and rail locomotives. However, this hybrid architecture is typically a mild hybrid system consisting of a diesel engine and a high-energy-density lithium battery pack, mostly geared towards stationary port cranes (STS, RMG) or lightly loaded mobile equipment. Directly adapting it to MHC presents certain bottlenecks, such as: ① The instantaneous power of MHC lifting equipment during lifting is too high (when lifting 40-120 t heavy boxes, the instantaneous power can reach 2-3 times the rated power), thus requiring a significant over-allocation of lithium battery packs, resulting in high battery costs, large size, and difficult placement; ② MHC needs to balance relocation and continuous operation. Limited by the layout of port charging stations and battery energy density, during long-term heavy-load operations, the rapid depletion of battery SOC can force frequent start-stop of the diesel engine, reducing its lifespan; ③ MHC relies on resistor consumption or mechanical braking when lowering heavy objects or decelerating, resulting in significant waste of potential / kinetic energy. Summary of the Invention

[0004] To overcome the aforementioned problems in the prior art, this invention provides an MHC intelligent hybrid power system, including an internal combustion engine generator set, a lithium battery energy storage device, a drive unit, an information acquisition unit, and an energy management controller. The lithium battery energy storage device and the drive unit are connected to the DC bus via a bidirectional DC / DC converter and a four-quadrant inverter, respectively. The energy management controller adopts a management architecture that combines outer-loop rolling prediction and inner-loop sliding mode control. During MHC operation, the hybrid power system predicts the future load power demand in the time domain based on current load information and selects the corresponding operating mode. This allows for the rational allocation of power between the internal combustion engine and the lithium battery pack during heavy load phases, ensuring the internal combustion engine always operates in the optimal fuel economy zone, reducing fuel consumption and emissions, and improving energy efficiency. Alternatively, during load reduction or braking phases, the lithium battery pack recovers and stores the generated energy for subsequent operation, which helps reduce battery capacity and cost. Correspondingly, this application also provides a mobile port crane employing an MHC intelligent hybrid power system.

[0005] For the system, the technical solution of this application is as follows:

[0006] The MHC intelligent hybrid power system includes an internal combustion engine generator set, a lithium battery energy storage device, a drive unit, an information acquisition unit, and an energy management controller. The internal combustion engine generator set is connected to the DC bus via an AC / DC rectifier. The lithium battery energy storage device is connected to the DC bus via a bidirectional DC / DC converter. The drive unit is connected to the DC bus via a four-quadrant inverter. The energy management controller integrates an outer-loop prediction model and an inner-loop control model. The internal combustion engine generator set includes an internal combustion engine and a generator. The battery energy storage device includes a lithium battery pack and a battery management module. During MHC operation, the information acquisition unit collects MHC load information in real time and sends it to the energy management controller. The outer-loop prediction model is based on a rolling time-domain prediction algorithm, which uses the received load data as parameters input to a lightweight temporal convolutional network. In the network, the load power demand of the MHC in the future time domain is predicted. When the load power demand reaches or exceeds a preset high power threshold, or the boom torque demand reaches or exceeds a preset high torque threshold, the hybrid output mode is selected. The multi-objective value function is constructed and solved with the internal combustion engine fuel consumption and lithium battery pack SOC as optimization objectives and emission constraints and noise constraints as constraints, to obtain the optimal power allocation sequence of the internal combustion engine and lithium battery pack in the future time domain. When the load power demand is negative, the energy recovery mode is selected. The inner loop control model is implemented based on the super-spiral sliding mode control algorithm, which sends the corresponding instructions to the internal combustion engine generator set and the lithium battery energy storage device, so that in the hybrid output mode, the internal combustion engine and lithium battery pack output according to the optimal power allocation sequence, and in the energy recovery mode, the lithium battery pack is charged to recover energy.

[0007] Furthermore, the construction process of the inner loop control model is as follows: First, design the sliding surface: , In the formula, For power tracking error, The rate of change of power tracking error. It is a constant. For the desired power value, Given the actual power value; then design the control law: , , ; ; For function Integral in the time domain; This is the gain coefficient.

[0008] Furthermore, the lightweight temporal convolutional network is trained using historical data: First, the historical load data is divided into a training set and a test set; then, the data in the training set is input into the lightweight temporal convolutional network for iterative training, allowing it to continuously learn the load power requirements under different state combinations until a predetermined number of training iterations or the loss function converges; finally, the data in the test set is input into the trained lightweight temporal convolutional network to evaluate its performance, obtain the optimal weights, and complete the training.

[0009] Furthermore, the specific formula for the multi-objective value function is: min J = In the formula, These are the weighting coefficients; Braking fuel consumption rate; This refers to the real-time power of the internal combustion engine. For bench emissions data; For dynamic shadow prices; This is the battery aging equivalent coefficient; , This represents the upper limit of the SOC (State of Charge) of a lithium battery pack. , For regression coefficients, The value is the rotational speed.

[0010] Furthermore, the high power threshold is (0.9-0.95) × the rated power of the internal combustion engine, and the high torque threshold is (0.7-0.75) × the rated allowable overturning torque.

[0011] Furthermore, when the load power demand of the MHC is predicted to be negative in the future time domain, and the recoverable energy is greater than the current absorbable energy of the lithium battery pack, the output power of the internal combustion engine is reduced, and the SOC upper limit constraint of the lithium battery pack is relaxed. This can prevent overcharging from triggering resistor braking and wasting energy, thereby improving energy recovery efficiency.

[0012] Furthermore, the information acquisition unit includes a pin-type load cell for acquiring the load, an incremental rotary encoder for acquiring the lifting speed, a dual-axis tilt sensor for acquiring the boom tilt angle, and an ultrasonic anemometer for acquiring the wind speed. The battery energy storage device includes at least one battery cabinet, and a lithium battery pack and a battery management module housed within the battery cabinet; the battery management module monitors the status of the lithium battery pack and is connected to the energy management controller via a signal. The battery cabinet design improves the sealing reliability of the lithium battery pack and the reliability of its electrical interfaces, thereby ensuring the reliability of the battery energy storage device of this invention in extreme industrial environments such as high humidity and high salt spray conditions in ports.

[0013] Furthermore, in the hybrid output mode, when the outer loop prediction model predicts that the MHC lifting speed is positive, the load reaches a set threshold, and the load power demand changes too rapidly (exceeding the set threshold) in the future time domain, the inner loop control model directly sends commands to the internal combustion engine generator set and the lithium battery energy storage device. This adjusts the internal combustion engine power to the optimal BSFC point while simultaneously discharging the lithium battery pack. This effectively suppresses the enriched fuel injection and smoke phenomena of the internal combustion engine during transient loading.

[0014] Furthermore, the MHC intelligent hybrid power system of the present invention also has two operating modes: an internal combustion engine-dominated mode and a lithium battery-dominated mode. When the load power demand of the MHC in the future time domain is predicted to be lower than the rated power of the internal combustion engine, the internal combustion engine-dominated mode is selected, allowing the internal combustion engine to operate in its high-efficiency range, while the lithium battery pack uses the surplus power of the internal combustion engine for charging. When the load power demand is less than the sustainable output power of the lithium battery pack, the lithium battery-dominated mode is selected, allowing the lithium battery pack to discharge and output electrical energy, while the internal combustion engine operates at the lowest idle speed.

[0015] Compared with existing technologies, the MHC intelligent hybrid power system of this application includes an internal combustion engine generator set, a lithium battery energy storage device, a drive unit, an information acquisition unit, and an energy management controller. The lithium battery energy storage device and the drive unit are connected to the DC bus via a bidirectional DC / DC converter and a four-quadrant inverter, respectively. The energy management controller adopts a management architecture that combines outer-loop rolling prediction and inner-loop sliding mode control. When the MHC is operating, the hybrid power system adopts a dual-source power split topology of internal combustion engine + lithium battery pack. First, the information acquisition unit collects the current load information of the MHC. Then, based on the load information, it predicts the future load power demand of the MHC in the time domain and selects the corresponding operating mode, including internal combustion engine-dominated mode, hybrid electric output mode, lithium battery-dominated mode, and energy recovery mode. Simultaneously, in the hybrid electric output mode… By using a rolling time-domain prediction algorithm to solve the multi-objective value function, the optimal power allocation sequence of the internal combustion engine and lithium battery pack in the future time domain is obtained. This sequence is then tracked and executed by the inner-loop sliding mode control module. This allows for the rational allocation of power between the internal combustion engine and lithium battery pack during heavy-load phases (when the load power demand reaches or exceeds a preset high-power threshold, or the boom torque demand reaches or exceeds a preset high-torque threshold, indicating a heavy-load phase). This ensures that the internal combustion engine remains in its optimal fuel economy zone, reducing fuel consumption and emissions while improving energy efficiency. Furthermore, in energy recovery mode, the lithium battery pack recovers energy during charging. This allows the potential and kinetic energy generated during load reduction or braking phases (i.e., negative power demand states) to be recovered and stored by the lithium battery pack for subsequent operation. In this case, the lithium battery pack can be designed based on average power, which helps reduce battery capacity and cost.

[0016] For cranes, the technical solution of this application is as follows:

[0017] The mobile port crane adopts the aforementioned MHC intelligent hybrid power system of this application.

[0018] Compared with existing technologies, the mobile port crane of this application adopts a deep hybrid power system of "lithium battery pack-internal combustion engine". It can recover excess energy under load reduction or braking conditions, and can also achieve parallel power supply of internal combustion engine and lithium battery pack under high load and long cycle operation conditions. Through specific energy management technology, the output power of internal combustion engine and lithium battery pack is reasonably distributed, resulting in low fuel consumption and emissions of the whole machine. Attached Figure Description

[0019] Figure 1 This is a block diagram of the MHC intelligent hybrid power system in the embodiments of this application;

[0020] Figure 2 This is a flowchart of the outer ring prediction model in the embodiments of this application;

[0021] Figure 3 These are the four working modes in the embodiments of this application; Figure 3 In this context, 'a' represents the diesel engine-dominated mode. Figure 3 In this context, 'b' represents the lithium battery-dominated mode. Figure 3 In this context, 'c' represents the hybrid electric output mode. Figure 3 In this context, d represents the energy recovery mode. Detailed Implementation

[0022] The present application will be further described below with reference to the accompanying drawings and embodiments, but this should not be construed as limiting the present application. Contents not described in detail in the following embodiments are all common knowledge in the art.

[0023] To address the issues of peak power surges, powertrain over-sizing, low energy recovery efficiency, and wasted braking energy in existing MHC hybrid power systems, this invention provides an intelligent MHC hybrid power system. This system employs a dual-source collaborative power split topology of an internal combustion engine (specifically a diesel engine in this embodiment) and a lithium battery pack. Utilizing a rolling time-domain prediction algorithm, it predicts load, lifting speed, boom tilt angle, and wind speed at the start of each lifting operation. This allows for the rational allocation of power between the diesel engine and the lithium battery pack during the predicted heavy-load phase, ensuring the diesel engine operates in its high-efficiency range and maintaining the lithium battery pack's SOC within a healthy range of 30-70%. This enables continuous operation for over 12 hours without recharging, thus resolving range anxiety and refueling concerns during continuous heavy-load operations. Simultaneously, during the predicted load lowering and braking phases, the system adjusts the upper limit of the lithium battery pack's SOC as needed. Combined with a bidirectional DC / DC converter and a four-quadrant inverter, a 100% energy recovery path is achieved, recharging the lithium battery pack with the potential and kinetic energy generated during load lowering and braking deceleration. The specific structure is as follows.

[0024] Example:

[0025] The mobile port crane in this embodiment adopts the MHC intelligent hybrid power system.

[0026] See Figure 1The MHC intelligent hybrid power system includes a diesel generator set, a lithium battery energy storage device, a drive unit, an information acquisition unit, and an energy management controller. The information acquisition unit collects load information from the MHC. The diesel generator set is connected to the DC bus via an AC / DC rectifier. The lithium battery energy storage device is connected to the DC bus via a bidirectional DC / DC converter. The drive unit is connected to the DC bus via a four-quadrant inverter. The diesel generator set includes a diesel engine and a generator. The battery energy storage device includes a lithium battery pack and a battery management module. The energy management controller integrates an outer-loop prediction model and an inner-loop control model. The outer-loop prediction model is based on a rolling time-domain prediction algorithm, operating with a 1-second cycle. It is used to predict the future load power demand of the MHC in the future time domain based on the collected load information, select the corresponding operating mode, and solve for the optimal power allocation sequence between the diesel engine and the lithium battery pack in the future time domain. The inner-loop control model is based on a super-spiral sliding mode control algorithm, operating with a 50ms cycle. It is used to send corresponding commands to the diesel generator set and the lithium battery energy storage device.

[0027] In this embodiment, the battery energy storage device includes two battery cabinets; each battery cabinet houses a lithium battery pack and a battery management module (BMS); the battery management module monitors the status of the lithium battery pack and is connected to the energy management controller via a CAN bus; the four-quadrant inverter is connected to the energy management controller using the Profibus protocol; the diesel generator set, AC / DC rectifier, and bidirectional DC / DC converter are respectively connected to the energy management controller via a CAN bus; the energy management controller uses a Siemens S7-1200 PLC.

[0028] In this embodiment, the information acquisition unit includes a pin-type load cell for acquiring the load weight, an incremental rotary encoder for acquiring the lifting speed, a dual-axis tilt sensor for acquiring the boom tilt angle, and an ultrasonic anemometer for acquiring the wind speed. The load weight represents the mass of the load, the lifting speed represents the vertical motion state of the load, the boom tilt angle represents the spatial geometric position of the load, and the wind speed quantifies random environmental disturbances (actual boom torque = load torque + wind torque). Using these four types of information as the main inputs to the load information can constitute a multi-dimensional, dynamic-static combined complete working condition perception system, thereby ensuring the accuracy of subsequent prediction results.

[0029] The energy management and control method for an MHC intelligent hybrid power system includes the following steps.

[0030] Step 1: When the MHC is working, the information acquisition unit collects the load information of the MHC in real time, including the lifting weight, lifting speed, boom tilt angle and wind speed, and sends it to the energy management controller.

[0031] Step 2: The outer loop prediction model inputs the received load data as parameters into the lightweight temporal convolutional network to predict the operating state of the MHC in the future time domain (e.g., 10s), and selects the corresponding operating mode based on the load power required for the operating state, including diesel engine-dominated mode, hybrid electric output mode, lithium battery-dominated mode, and energy recovery mode.

[0032] When the load power demand is lower than the rated power of the diesel engine, the diesel engine-dominated mode is selected; when the load power demand is less than the sustainable output power of the lithium battery pack, the lithium battery-dominated mode is selected; when the load power demand is negative, the energy recovery mode is selected; when the load power demand reaches or exceeds the preset high power threshold (0.95 × rated diesel engine power), or the boom torque demand reaches or exceeds the preset high torque threshold (0.7 × rated allowable overturning torque), the hybrid electric output mode (i.e., the dual-source parallel mode) is selected. With diesel engine fuel consumption and lithium battery pack SOC as optimization objectives, and emission constraints and noise constraints as constraints, a multi-objective value function is constructed and solved to obtain the optimal power allocation sequence of the diesel engine and lithium battery pack in the future time domain.

[0033] Since the output capacity of a single power source is limited, when the MHC enters high-power operation, the total output power will increase after the two power sources are connected in parallel, thereby effectively sharing the load of the diesel engine and avoiding problems such as overload, overheating, and shutdown due to long-term operation close to full load. When the MHC enters high-torque operation (at which time the boom torque margin is small and the safety risk is high), the dual-source parallel connection can provide a more stable and continuous power output, preventing the boom movement from going out of control due to power fluctuations, and thus avoiding the safety hazard of equipment overturning.

[0034] Load power typically includes lifting power (the net power required to vertically raise / lower the cargo), wind load power (the power required to overcome the additional torque exerted by wind pressure on the spreader, cargo, boom, etc.), and luffing / boom pitching power (the power required to overcome gravitational torque plus friction when changing the boom tilt angle). Lifting power P = m * g * V; where m is the load (kg), and g = 9.81 m / s². 2 V is the lifting speed (m / s); wind pressure q = 0.5*ρ*v²*C; ρ is the air density, v is the wind speed (m / s), and C is the overall drag coefficient, typically 1.0 to 1.3 for containers / bulk cargo; in this embodiment, an incremental rotary encoder mounted on the top pulley is used to measure the lifting speed, and the wind load additional power is included in the lifting power. Luffing / boom pitching power P = (W L +0.5W B )*L*cos(α)*ω / η;W L For lifting heavy loads, W BLet L be the boom's self-weight, α be the boom length, ω be the boom tilt angle, ω be the boom's angular velocity, and η be the overall efficiency of the transmission system. Boom torque = W L *L*cos(α).

[0035] In this embodiment, the lightweight temporal convolutional network is trained using historical data: First, the historical load data (lifting weight, hoisting speed, boom tilt angle, and wind speed) is divided into a training set and a test set; then, the data in the training set is input into the lightweight temporal convolutional network for iterative training, allowing it to continuously learn the load power demand under different state combinations (when the power demand is negative, it indicates that there is recoverable energy), until a predetermined number of training iterations is reached (or the loss function converges); then, the data in the test set is input into the trained lightweight temporal convolutional network to evaluate its performance, obtain the optimal weights, and the training is complete.

[0036] In this embodiment, the specific formula for the multi-objective value function is: min J = In the formula, These are the weighting coefficients; Braking fuel consumption rate, in g / kWh, derived from the diesel engine MAP chart; This refers to the real-time power of the diesel engine. The test emissions data were derived from the diesel engine MAP (obtained from test bench tests before the diesel engine leaves the factory); The dynamic shadow price (adjusted according to port air quality, which is 1 in this embodiment); This is the battery aging equivalent coefficient (obtained by the battery manufacturer through accelerated aging tests). , This represents the upper limit of the SOC (State of Charge) of a lithium battery pack. The empirical formula for sound pressure level dB(A) is as follows: , For rotational speed, The regression coefficients were calibrated through measurements taken between 22:00 and 06:00 at night.

[0037] In this embodiment, the construction process of the inner loop control model is as follows: First, design the sliding surface: , In the formula, For power tracking error, The rate of change of power tracking error. It is a constant. , For the desired power value, Given the actual power value; then, design the control law: , , ; ; For function Integral in the time domain; This is the gain coefficient.

[0038] Step 3: The inner-loop control model sends the corresponding commands to the diesel generator set and lithium battery energy storage device, causing them to operate according to the corresponding working modes. See the table below.

[0039] Work mode lithium battery pack diesel engine Load status Diesel engine dominant Charge Work rise Lithium batteries dominate Discharge standby None / Light Hybrid output Discharge Work rise Energy recovery Charge Work decline

[0040] Specifically, in diesel-dominated mode, the diesel engine operates in its high-efficiency range, and the lithium-ion battery pack utilizes the diesel engine's surplus power for charging (see...). Figure 3 a). In lithium battery-dominated mode, the diesel engine operates at minimum idle speed, and the lithium battery pack discharges to output electrical energy (see...). Figure 3 b). In hybrid output mode, the diesel engine operates in its high-efficiency range, and the lithium battery pack discharges to output electrical energy (see...). Figure 3 c; This is typically used under heavy load, long-cycle operation, i.e., the heavy load phase); simultaneously, the inner-loop control model sends real-time power commands from the optimal power allocation sequence to the diesel generator set and lithium battery energy storage device, causing the diesel engine and lithium battery pack to execute the corresponding power output. In energy recovery mode, the lithium battery pack is charged to recover energy (see...). Figure 3 d; adopted during load reduction or braking phase), and when the predicted recoverable energy is greater than the current absorbable energy of the lithium battery pack, the output power of the diesel engine is reduced and the SOC upper limit constraint of the lithium battery pack is relaxed to prevent overcharging from triggering resistor braking and wasting energy, thereby improving energy recovery efficiency.

[0041] Furthermore, in the hybrid power output mode, when the outer loop prediction model predicts an impending lifting peak, the inner loop control model directly sends commands to the diesel generator set and lithium battery energy storage device. This pre-adjusts the diesel engine power to the optimal BSFC point (e.g., 3 seconds in advance) and simultaneously discharges the lithium battery pack (e.g., when a container is lifted from the bottom of a ship's hold; the real-time power requirement will surge the instant the heavy object leaves the ground, while the diesel engine is already in its most efficient operating state, thus suppressing the enriched fuel injection and smoke phenomena during transient loading). When the predicted lifting speed of the MHC is positive, the load reaches the set threshold, and the rate of change of load power demand exceeds the set threshold in the future time domain, it indicates that a lifting peak is imminent.

[0042] To verify that the hybrid power system of this invention can effectively reduce fuel consumption and improve energy recovery efficiency, the inventors also built a hybrid power experimental verification platform for MHC lifting equipment. Centered on an energy management controller (EMS), the platform includes functions such as dynamic power allocation (based on the coordinated implementation of outer loop rolling prediction and inner loop sliding mode control), adaptive switching of operating modes, and fault warning. It is equipped with an information acquisition unit, a battery management unit (BMS), a diesel generator set, an AC / DC rectifier, a bidirectional DC / DC converter, and a four-quadrant inverter. The energy management controller is connected to the actuator (drive unit) of the lifting equipment using the Profibus protocol. Operators can configure parameters, set safety thresholds, and monitor the operating status online through a human-machine interface. Experiments showed that the energy management controller dynamically allocates the output power of the diesel engine and lithium battery pack based on operating condition prediction and real-time load identification, ensuring that the diesel engine is always in the optimal fuel economy zone, resulting in a fuel saving rate of 20-30%.

[0043] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art can still modify the technical solutions described in the foregoing examples or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, or improvements made to the disclosed technical features in the above general description and / or specific embodiments (including examples) based on the disclosure of this application, without departing from the constituent elements of the invention, should be included within the protection scope of the present invention.

Claims

1. The MHC intelligent hybrid power system, characterized in that: It includes an internal combustion engine generator set, a lithium battery energy storage device, a drive unit, an information acquisition unit, and an energy management controller; The internal combustion engine generator set is connected to the DC bus via an AC / DC rectifier; the lithium battery energy storage device is connected to the DC bus via a bidirectional DC / DC converter; the drive unit is connected to the DC bus via a four-quadrant inverter; the energy management controller integrates an outer loop prediction model and an inner loop control model; the internal combustion engine generator set includes an internal combustion engine and a generator; the battery energy storage device includes a lithium battery pack and a battery management module. During MHC operation, the information acquisition unit collects MHC load information in real time and sends it to the energy management controller. The outer loop prediction model is based on a rolling time-domain prediction algorithm, which uses the received load data as parameters to input into a lightweight temporal convolutional network to predict the load power demand of the MHC in the future time domain. When the load power demand reaches or exceeds a preset high-power threshold, or the boom torque demand reaches or exceeds a preset high-torque threshold, the hybrid output mode is selected. A multi-objective value function is constructed and solved using internal combustion engine fuel consumption and lithium battery pack SOC as optimization objectives, and emission and noise constraints as constraints, to obtain the optimal power allocation sequence for the internal combustion engine and lithium battery pack in the future time domain. When the load power demand is negative, the energy recovery mode is selected. The inner loop control model is based on a super-spiral sliding mode control algorithm, which sends corresponding commands to the internal combustion engine generator set and the lithium battery energy storage device. In the hybrid output mode, the internal combustion engine and lithium battery pack output according to the optimal power allocation sequence; in the energy recovery mode, the lithium battery pack is charged for energy recovery.

2. The MHC intelligent hybrid power system according to claim 1, characterized in that, The construction process of the inner loop control model is as follows: First, design the sliding surface: , In the formula, For power tracking error, The rate of change of power tracking error. It is a constant. For the desired power value, Given the actual power value; then, design the control law: , , ; ; For function Integral in the time domain; This is the gain coefficient.

3. The MHC intelligent hybrid power system according to claim 1, characterized in that, The lightweight temporal convolutional network is trained using historical data: First, the historical load data is divided into a training set and a test set; then, the data in the training set is input into the lightweight temporal convolutional network for iterative training, allowing it to continuously learn the load power requirements under different state combinations until a predetermined number of training iterations or the loss function converges; finally, the data in the test set is input into the trained lightweight temporal convolutional network to evaluate its performance, obtain the optimal weights, and complete the training.

4. The MHC intelligent hybrid power system according to claim 3, characterized in that: The specific formula for the multi-objective value function is: min J = In the formula, These are the weighting coefficients; Braking fuel consumption rate; This refers to the real-time power of the internal combustion engine. For bench emissions data; For dynamic shadow prices; This is the battery aging equivalent coefficient; , This represents the upper limit of the SOC (State of Charge) of a lithium battery pack. , For regression coefficients, The value is the rotational speed.

5. The MHC intelligent hybrid power system according to claim 1, characterized in that: The high power threshold is (0.9-0.95) × the rated power of the internal combustion engine, and the high torque threshold is (0.7-0.75) × the rated allowable overturning torque.

6. The MHC intelligent hybrid power system according to claim 1, characterized in that: When the load power demand of the MHC is predicted to be negative in the future time domain, and the recoverable energy is greater than the current absorbable energy of the lithium battery pack, the output power of the internal combustion engine is reduced, and the SOC upper limit constraint of the lithium battery pack is relaxed.

7. The MHC intelligent hybrid power system according to any one of claims 1 to 6, characterized in that: The information acquisition unit includes a pin-type load cell for acquiring the load, an incremental rotary encoder for acquiring the lifting speed, a dual-axis tilt sensor for acquiring the boom tilt angle, and an ultrasonic anemometer for acquiring the wind speed.

8. The MHC intelligent hybrid power system according to claim 7, characterized in that: In the hybrid output mode, when the outer loop prediction model predicts that the lifting speed of the MHC is positive, the load reaches the set threshold, and the load power demand changes faster than the set threshold in the future time domain, the inner loop control model directly sends commands to the internal combustion engine generator set and the lithium battery energy storage device to adjust the internal combustion engine power to the optimal BSFC point and at the same time discharge the lithium battery pack.

9. The MHC intelligent hybrid power system according to claim 1, characterized in that: It also features two operating modes: an internal combustion engine-dominated mode and a lithium battery-dominated mode. When the predicted load power demand of the MHC in the future time domain is lower than the rated power of the internal combustion engine, the internal combustion engine-dominated mode is selected, allowing the internal combustion engine to operate in its high-efficiency range, while the lithium battery pack uses the surplus power of the internal combustion engine to charge. When the load power demand is less than the sustainable output power of the lithium battery pack, the lithium battery-dominated mode is selected, allowing the lithium battery pack to discharge and output electrical energy, while the internal combustion engine operates at the lowest idle speed.

10. A mobile port crane, characterized in that: The MHC intelligent hybrid power system described in claim 1 is adopted.

Images