Lifting machine hydraulic power demand prediction method and system fused with working condition recognition

Abstract

This invention provides a method and system for predicting hydraulic power demand of a lifting platform by integrating working condition identification, relating to the field of artificial intelligence technology. First, it acquires basic operating information of the lifting platform's hydraulic system, basic driving information of the AI ​​prediction model, and basic identification information of the working condition identification module, constructing a dynamically coupled and correlated structure. Based on this dynamic coupling and correlated structure, it synchronously collects real-time information to generate a working condition-power mutual drive iterative signal. This working condition-power mutual drive iterative signal is input into a dynamic rule evolution unit to form a working condition-power demand prediction rule. Using this working condition-power demand prediction rule as logic, it calculates the continuously input data, outputs the hydraulic power demand prediction result, and transmits it to the control unit, achieving accurate prediction and dynamic adjustment, improving operational stability and energy utilization efficiency.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and more specifically, to a method and system for predicting the hydraulic power demand of a lifting machine by integrating working condition recognition. Background Technology

[0002] During the operation of a lift, the hydraulic system provides crucial power support. Accurately predicting hydraulic power demand is essential for ensuring stable operation, improving energy efficiency, and extending equipment lifespan. Traditional methods for predicting lift hydraulic power demand often rely solely on historical operating data of the hydraulic system, using simple statistical models or empirical formulas. However, these methods have significant limitations.

[0003] On the one hand, it ignores the differences in hydraulic power requirements of lifts under different working conditions. Lifts face various working conditions in practice, such as different load weights, lifting heights, and operating speeds. These changes significantly affect the power requirements of the hydraulic system, and traditional methods cannot dynamically adjust according to these changes. On the other hand, traditional methods lack a precise perception and feedback mechanism for the real-time operating status of the system, making it difficult to capture subtle changes in the hydraulic system in real time. This leads to a large deviation between predicted results and actual needs, failing to provide a reliable basis for precise control of the lift and making it difficult to meet the requirements of modern industrial production for efficient and stable operation of lifts. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for predicting the hydraulic power demand of a lifting platform by integrating working condition identification, the method comprising: The system acquires basic operating information of the lift hydraulic system, basic driving information of the pre-trained AI prediction model, and basic identification information of the lift working condition identification module. Based on the basic operating information, basic driving information, and basic identification information, a dynamic coupling association structure is constructed. This dynamic coupling association structure defines bidirectional data call and response rules between hydraulic system operating parameters, AI prediction model driving parameters, and working condition identification parameters. Based on the aforementioned dynamic coupling and correlation structure, real-time operating information of the lifting hydraulic system, real-time drive status information of the AI ​​prediction model, and real-time working condition feature information of the working condition identification module are collected synchronously to generate a working condition-power mutual drive iterative signal. The working condition-power mutual drive iterative signal includes the feedback drive signal of the hydraulic system to the AI ​​prediction model, the prediction signal of the AI ​​prediction model to the hydraulic power demand, and the correlation drive signal of the working condition features to the power demand. The working condition-power mutual drive iterative signal is input into the dynamic rule evolution unit. Combined with the operating characteristics of the lifting hydraulic system, the prediction characteristics of the AI ​​prediction model, and the dynamic adaptation characteristics of working condition identification, the working condition-power demand prediction rule is evolved. The working condition-power demand prediction rule includes the adjustment logic of the mutual drive signal and the adaptation standard of power demand prediction. Using the aforementioned working condition-power demand prediction rule as processing logic, the system calculates the continuously input real-time operating information of the hydraulic system, the predicted state data of the AI ​​prediction model, and the real-time characteristic data of the working condition, and outputs the hydraulic power demand prediction result. The hydraulic power demand prediction result is transmitted to the lifting machine hydraulic system control unit through a dynamic coupling association structure.

[0005] In another aspect, embodiments of the present invention also provide a lifting hydraulic power demand prediction system that integrates working condition identification, including a processor and a machine-readable storage medium connected to the processor. The machine-readable storage medium is used to store programs, instructions or code, and the processor is used to run the programs, instructions or code in the machine-readable storage medium to implement the above-described method.

[0006] Based on the above, this embodiment of the invention constructs a dynamic coupling and correlation structure to integrate the basic operating information of the lift's hydraulic system, the basic driving information of the AI ​​prediction model, and the basic identification information of the working condition identification module. It defines bidirectional data call and response rules between various parameters. The working condition-power mutual drive iterative signal generated based on this dynamic coupling and correlation structure covers multiple aspects of information, including hydraulic system feedback, model prediction, and working condition correlation, and can reflect the changes in the lift's power demand under different working conditions. The dynamic rule evolution unit, combining system operation, model prediction, and working condition identification characteristics, evolves into working condition-power demand prediction rules that can effectively adjust the mutual drive signal and clarify the power demand prediction adaptation standard. The final output hydraulic power demand prediction result is transmitted to the control unit via the dynamic coupling and correlation structure, realizing accurate prediction and dynamic adjustment of the lift's hydraulic power demand, improving the lift's operational stability and energy utilization efficiency. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of the execution flow of the lifting hydraulic power demand prediction method based on integrated working condition identification provided in an embodiment of the present invention.

[0008] Figure 2 This is a schematic diagram of the hardware architecture of the lifting hydraulic power demand prediction system with integrated working condition identification provided in an embodiment of the present invention. Detailed Implementation

[0009] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1This is a flowchart illustrating a method for predicting hydraulic power demand of a lift based on integrated working condition identification, provided in one embodiment of the present invention. The following is a detailed description of this method for predicting hydraulic power demand of a lift based on integrated working condition identification.

[0010] Step S110: Obtain the basic operating information of the lifting hydraulic system, the basic driving information of the pre-trained AI prediction model, and the basic identification information of the lifting condition identification module. Based on the basic operating information, basic driving information, and basic identification information, construct a dynamic coupling association structure. This dynamic coupling association structure defines the bidirectional data call and response rules between the hydraulic system operating parameters, the AI ​​prediction model driving parameters, and the condition identification parameters.

[0011] This embodiment uses a two-post lift in an auto repair shop as an application scenario, where the lift is used for lifting and repairing cars. When constructing the dynamically coupled and interconnected structure, it is first necessary to collect various basic data of the hydraulic system during the initial stage of normal operation. This data includes basic information on hydraulic oil storage, hydraulic pump operation, hydraulic valve control, and the movement of hydraulic actuators. Simultaneously, basic driving information of a pre-trained AI prediction model is acquired. This AI prediction model is used to predict the hydraulic power requirements of the lift, and its basic driving information covers input parameter types, output signal formats, etc. Basic identification information of the lift's operating condition identification module also needs to be collected. This module is used to determine the current operating condition of the lift, such as unloaded, lightly loaded, or heavily loaded lifting. By integrating the above three types of basic information, a dynamic coupling and correlation structure is constructed. This dynamic coupling and correlation structure clarifies how the hydraulic system operating parameters, AI prediction model driving parameters, and working condition identification parameters conduct bidirectional data calls and responses. For example, when the working condition identification module identifies a change in working condition, how does it pass the relevant parameters to the AI ​​prediction model, and how does the AI ​​prediction model adjust the driving parameters based on these parameters, thereby affecting the hydraulic system operating parameters?

[0012] Step S111: Collect basic operating information of the lifting platform's hydraulic system. This basic operating information includes hydraulic oil storage status information, hydraulic pump operating basic information, hydraulic valve control basic information, and hydraulic actuator operation basic information. The hydraulic oil storage status information includes oil quantity, oil temperature, and oil pressure. The hydraulic pump operating basic information includes rated flow rate, rated pressure, and operating mode. The hydraulic valve control basic information includes valve type, control signal range, and response time. The hydraulic actuator operation basic information includes cylinder bore, rod diameter, and stroke.

[0013] In the hydraulic system of a two-post lift, hydraulic oil storage status information is collected by sensors installed on the hydraulic oil tank. Oil quantity reflects the total amount of hydraulic oil in the tank; oil temperature affects the viscosity and flowability of the hydraulic oil; and oil pressure is the basic pressure for normal system operation. The hydraulic pump, as the power source, has a rated flow rate that determines the volume of hydraulic oil delivered per unit time, and a rated pressure that is the maximum working pressure it can provide. Operating modes include manual and automatic control; this information can be obtained from the hydraulic pump's technical documentation and the controller. Regarding basic hydraulic valve control information, the lift's hydraulic system includes various types of valves, such as directional valves and relief valves. Each valve has its corresponding control signal range, such as the value range of voltage or current signals, and response time—the time from receiving the control signal to the valve completing its action. The hydraulic actuator is the lifting hydraulic cylinder; its cylinder bore and rod diameter determine the output force, and its stroke limits the lifting height. These parameters are obtained through actual measurements of the hydraulic cylinder.

[0014] Step S112: Extract the basic driving information of the pre-trained AI prediction model. The basic driving information includes the input parameter type, output signal format, basic prediction calculation logic, and basic characteristics of the driving response of the AI ​​prediction model.

[0015] In predicting the hydraulic power demand of a lift, the pre-trained AI prediction model requires input parameters that are relevant to the lift's operating status, such as hydraulic oil temperature, system pressure, pump speed, load weight, lifting speed, and ambient temperature. The output signal format must meet the receiving requirements of the hydraulic system control unit, typically containing a standardized data structure including predicted pressure, flow rate, and power parameters. The basic logic for prediction calculation is trained based on historical operating data. By analyzing the power consumption patterns under different operating conditions, a mapping relationship between input and output is established, which may include feature preprocessing, model inference, and result post-processing. The fundamental characteristics of the drive response include the model's sensitivity to input changes and its output response speed. For example, the time it takes for the model to output a new prediction when load parameters change directly affects the real-time control of the hydraulic system.

[0016] Step S113: Collect basic identification information from the lifting machine operating condition identification module. The basic identification information includes operating condition classification standard information, basic information for operating condition feature extraction, basic information for the association between operating condition and power demand, and basic information for dynamic switching identification of operating conditions. The operating condition classification standard information includes a preset list of operating condition categories; the basic information for operating condition feature extraction includes a list of feature sensors to be extracted; the basic information for the association between operating condition and power demand includes typical power values ​​corresponding to various operating conditions in historical data; and the basic information for dynamic switching identification of operating conditions includes threshold conditions for determining operating condition switching.

[0017] The basic identification information for the lift operating condition identification module includes several preset operating condition categories, such as no-load standby, light-load lifting, heavy-load lifting, static support after lifting, and descent. The basic information for operating condition feature extraction includes a list of sensors used to extract operating condition features, such as weight sensors, speed sensors, pressure sensors, and environmental sensors. These sensors provide raw data for operating condition identification. The basic information for linking operating conditions with power demand analyzes historical data to obtain the typical power value range for each type of operating condition; for example, the power value under light-load lifting conditions is usually lower than that under heavy-load lifting conditions. The basic information for dynamic operating condition switching identification includes threshold conditions used to determine whether an operating condition needs to be switched; for example, when the load change exceeds a certain value, it is determined that the load condition is being switched from one to another.

[0018] Step S114: Divide the basic operating information of the hydraulic system into multiple basic operating units according to functional attributes. Each basic operating unit corresponds to a core functional module of the hydraulic system. Each basic operating unit contains the key basic parameters and operating constraints of the core functional module.

[0019] The basic operating information of a hydraulic system can be divided into multiple basic operating units based on functional attributes. The hydraulic oil storage operating unit corresponds to the hydraulic oil storage module, which includes key basic parameters such as oil quantity, oil temperature, and oil pressure, as well as operating constraints such as the lower limit of oil quantity, upper and lower limits of oil temperature, and normal range of oil pressure. The hydraulic pump operating unit corresponds to the hydraulic pump module, which includes key parameters such as rated flow rate, rated pressure, and operating mode, as well as constraints such as the pump's operating temperature range and continuous operating time limits. The hydraulic valve control operating unit corresponds to the hydraulic valve control module, which includes key parameters such as valve type, control signal range, and response time, as well as constraints such as control signals not exceeding the range and response times not being too long. The hydraulic actuator operating unit corresponds to the hydraulic cylinder module, which includes key parameters such as cylinder bore, rod diameter, and stroke, as well as constraints such as the stroke not exceeding the maximum permissible stroke.

[0020] Step S115: Divide the basic driving information of the AI ​​prediction model into multiple driving basic units according to the function dimension. Each driving basic unit corresponds to a core driving module of the AI ​​prediction model. Each driving basic unit contains the key driving parameters and response constraints of the core driving module.

[0021] The fundamental driving information of AI prediction models is divided into three categories based on their function: input parameter processing, prediction computation, and output signal generation. The input parameter processing unit corresponds to the model's input processing module, including key driving parameters such as the number, data type, and value range of input parameters, as well as response constraints for quickly and accurately receiving and preprocessing input parameters, such as parameter normalization. The prediction computation unit corresponds to the model's computation module, including key parameters such as internal weight coefficients and bias values, as well as response constraints ensuring that computation time does not exceed a set threshold while maintaining accuracy. The output signal generation unit corresponds to the model's output module, including key parameters such as the format and accuracy requirements of the output signal, as well as response constraints ensuring that the output signal conforms to the control unit's receiving standards and has minimal transmission delay.

[0022] Step S116: Divide the basic identification information of the working condition identification module into multiple working condition identification basic units according to the identification dimension. Each working condition identification basic unit corresponds to a core identification module of the working condition identification module. Each working condition identification basic unit contains the key identification parameters and adaptation constraints of the core identification module.

[0023] The basic identification information of the operating condition identification module is divided into four units based on identification dimensions: operating condition classification identification, operating condition feature extraction, operating condition dynamic correlation, and operating condition switching identification. The operating condition classification identification unit corresponds to the operating condition classification module, containing a pre-defined list of operating condition categories and key identification parameters such as feature thresholds for each category, as well as adaptation constraints for accurately matching the current operating condition features with the pre-defined categories. The operating condition feature extraction unit corresponds to the feature extraction module, containing key parameters such as a list of feature sensors to be extracted and feature extraction methods, as well as adaptation constraints for accurate and real-time feature extraction. The operating condition dynamic correlation unit corresponds to the operating condition and dynamic correlation module, containing key parameters such as typical dynamic values ​​corresponding to various operating conditions, as well as adaptation constraints for quickly querying the range of typical dynamic values ​​based on the current operating condition. The operating condition switching identification unit corresponds to the operating condition switching detection module, containing key parameters such as threshold conditions for operating condition switching, as well as adaptation constraints for timely and accurate identification of operating condition switching events.

[0024] Step S117: Establish a preliminary mapping relationship between the operating basic unit, the driving basic unit, and the working condition identification basic unit. The mapping is based on the parameter type adaptability, functional correlation, and constraint compatibility of the operating basic unit, the driving basic unit, and the working condition identification basic unit, forming a preliminary mapping set.

[0025] When establishing initial mapping relationships, parameter type compatibility is considered. For example, structural parameters such as cylinder bore and rod diameter of the hydraulic actuator operating base unit are matched with load-related input parameter types of the AI ​​prediction model input parameter processing drive base unit. Based on functional correlation, the operating mode of the hydraulic pump operating base unit is associated with the operating condition category of the operating condition classification and identification base unit; for example, automatic control modes may correspond to different operating conditions. Based on constraint compatibility, the response time constraints of the hydraulic valve control operating base unit must be compatible with the computation time constraints of the AI ​​prediction model prediction calculation drive base unit. Based on these criteria, related units are paired to form an initial mapping set. For example, the oil temperature parameter of the hydraulic oil storage operating base unit is associated with the ambient temperature sensor feature extraction of the operating condition feature extraction base unit, and simultaneously with the temperature input parameters of the AI ​​prediction model input parameter processing drive base unit.

[0026] Step S118: Construct a test dataset containing basic operating state parameters of the hydraulic system, basic drive state parameters of the AI ​​prediction model, and basic state parameters of the working condition identification; input the test dataset into each mapping relationship in the preliminary mapping set, and record the success rate and delay time of data transmission.

[0027] When constructing the test dataset, basic operating parameters of the hydraulic system of the lift under different working conditions are collected, such as different combinations of oil quantity, oil temperature, and oil pressure; basic drive state parameters of the AI ​​prediction model, such as different combinations of input parameter types and output signal formats; and basic state parameters for working condition identification, such as different working condition categories and feature sensor data. These data are then combined to form the test dataset. Each mapping relationship in the initial mapping set is input, and the proportion of successful data transmissions to the total number of tests is recorded as the success rate, and the time from data input to transmission completion is recorded as the delay time.

[0028] Step S119: Based on the success rate and delay time of the data transmission, adjust the data conversion parameters or link paths of the mapping relationships in the initial mapping set, and then repeatedly use the test dataset to test until the data transmission success rate of all mapping relationships exceeds the preset threshold and the delay time is lower than the preset threshold.

[0029] Based on the success rate and latency obtained from the tests, if the success rate of a certain mapping relationship is lower than a preset threshold or the latency is higher than a preset threshold, its data conversion parameters or link paths are adjusted. For example, the conversion formula from hydraulic oil temperature parameters to AI prediction model input parameters is adjusted to improve accuracy, or the link path is optimized to reduce intermediate steps and lower latency. After adjustment, the test is repeated until the success rate and latency of all mapping relationships meet the standards.

[0030] Step S1110: The optimized mapping relationship is structurally integrated according to the operation process of the hydraulic system, the driving process of the AI ​​prediction model, and the identification process of the working condition identification module to form a multi-level mapping structure including input mapping, output mapping, and feedback mapping. A dynamic adjustment method is embedded in the multi-level mapping structure. This dynamic adjustment method can automatically adjust the parameter association weights or mapping logic in the mapping relationship according to the changes in the hydraulic system operation state, the driving state of the AI ​​prediction model, and the working condition identification state. The multi-level mapping structure including the dynamic adjustment method is solidified into a dynamically coupled association structure.

[0031] The optimized mapping relationships are structurally integrated according to the hydraulic system operation flow (from hydraulic oil storage to actuator action), the AI ​​prediction model-driven flow (input processing to output generation), and the working condition recognition module's recognition flow (data acquisition to working condition judgment), forming a multi-level mapping structure. Input mapping inputs hydraulic system and working condition parameters into the AI ​​model, output mapping outputs the model's prediction results to the hydraulic control unit, and feedback mapping provides state feedback for each component. Dynamic adjustment methods are embedded in the multi-level mapping structure. By monitoring system state changes, the parameter association weights or mapping logic of the mapping relationships are automatically adjusted, such as increasing the weight of relevant parameters when working conditions change. Finally, the multi-level mapping structure, including the dynamic adjustment method, is solidified into a dynamically coupled association structure.

[0032] Step S1110-1: Define the triggering factors for changes in the operating status of the hydraulic system. The triggering factors include: load sensor reading changes exceeding a threshold, ambient temperature sensor reading changes exceeding a threshold, system cumulative operating time reaching a maintenance cycle threshold, or fault codes of critical hardware components being triggered.

[0033] Triggers for changes in the operating status of a hydraulic system include: load sensor readings exceeding thresholds, i.e., changes in lifting load exceeding the set range; ambient temperature sensor readings exceeding thresholds, as changes in ambient temperature affect hydraulic oil performance; the system's cumulative operating time reaching the maintenance cycle threshold, indicating that increased operating time requires maintenance; and fault codes of critical hardware components being triggered, such as malfunctions in hydraulic pumps or hydraulic valves.

[0034] Step S1110-2: Analyze the triggering factors for changes in the state driven by the AI ​​prediction model. These factors include various situations that may lead to changes in the state driven by the AI ​​prediction model, such as changes in the distribution of input data, changes in the prediction scenario, drift of model parameters, and fluctuations in operating resources. Analyze the triggering factors for changes in the state of the operating condition identification. These factors include various situations that may lead to changes in the state of the operating condition identification, such as switching of operating condition types, load fluctuations, changes in environmental conditions, and drift of identification parameters.

[0035] Triggers for AI prediction model-driven state changes include changes in input data distribution (i.e., changes in the distribution of input parameters); changes in the prediction scenario, such as changes in the lifting machine's operating environment; model parameter drift (changes in internal model parameters over time); and fluctuations in operating resources, such as changes in CPU utilization and memory usage. Triggers for changes in operating condition recognition state include switching operating condition types, such as from no-load to heavy-load; load fluctuations (changes in load weight); changes in environmental conditions, such as humidity and dust affecting sensor accuracy; and drift in recognition parameters, such as changes in feature extraction parameters and classification thresholds in the operating condition recognition module.

[0036] Step S1110-3: Configure a monitoring index for each triggering factor of the hydraulic system. The monitoring index for load change is the load sensor reading; the monitoring index for environmental condition change is the temperature sensor reading; the monitoring index for accumulated running time is the system timer reading; the monitoring index for hardware wear is the preset vibration or current characteristic value; set corresponding monitoring indexes for various triggering factors that drive state changes by AI prediction model, and set corresponding monitoring indexes for various triggering factors that identify state changes by working condition. Each monitoring index can reflect the degree of change and the scope of influence of the corresponding triggering factor.

[0037] Monitoring indicators are configured for triggering factors in the hydraulic system. For load changes, the monitoring indicator is the load sensor reading, reflecting the degree of load change; for environmental condition changes, the monitoring indicator is the temperature sensor reading, reflecting changes in ambient temperature; for accumulated running time, the monitoring indicator is the system timer reading, recording the running duration; and for hardware wear, the monitoring indicator is preset vibration or current characteristic values, reflecting hardware wear. Monitoring indicators are also set for triggering factors in the AI ​​prediction model, such as changes in the input data distribution (mean, variance, etc.) and scene identification parameters (for changes in the prediction scenario). Finally, monitoring indicators are set for triggering factors in operating condition identification, such as changes in the operating condition category identifier (for switching operating condition types) and the rate of change in load sensor readings (for load fluctuations).

[0038] Step S1110-4: Construct a query table that records the type and level of different triggering factors and their correspondence with the weights associated with the parameters to be adjusted or the mapping logic to be switched.

[0039] Construct a query table to record the type (e.g., hydraulic system, AI model, operating condition identification) and level (e.g., slight, moderate, severe) of triggering factors, as well as the corresponding parameter association weight adjustment values ​​or mapping logic identifiers. For example, when the load on the hydraulic system changes drastically, the corresponding parameter association weight is adjusted so that the weight of the hydraulic pump rated pressure parameter increases and the weight of the hydraulic valve response time parameter decreases; when the distribution of the AI ​​model input data changes significantly, the corresponding mapping logic is switched to a new input data preprocessing mapping logic.

[0040] Step S1110-5: Set up a monitoring indicator acquisition unit to collect various monitoring indicator data from the hydraulic system, AI prediction model and working condition identification module in real time; establish a monitoring data analysis unit to analyze the collected monitoring indicator data in real time, identify the changing status of triggering factors, and determine whether it is necessary to start adjusting the mapping parameters. A monitoring indicator acquisition unit was established. Through a distributed data acquisition architecture, acquisition nodes were installed at key components of the hydraulic system, the computer running the AI ​​model, and the sensors of the working condition identification module to collect various monitoring indicator data in real time. A monitoring data analysis unit was also established to perform real-time analysis of the acquired data, such as calculating the change in load sensor readings and the rate of change, comparing them with trigger factor level thresholds, identifying the change status of trigger factors, and determining whether to initiate mapping parameter adjustments.

[0041] Step S1110-6: When the monitoring data analysis unit identifies that the triggering factor has been activated, it queries the query table according to the type and level of the triggering factor to obtain the corresponding adjustment strategy. The adjustment strategy includes parameter association weight adjustment value or mapping logic identifier. When the monitoring data analysis unit identifies that a triggering factor has been activated, such as a moderate change in the load of the hydraulic system, it queries the lookup table according to the type and level of the triggering factor to obtain an adjustment strategy. For example, the parameter association weight adjustment value is to increase the weight of the rated flow parameter of the hydraulic pump, increase the weight of the cylinder diameter parameter of the hydraulic actuator, or the mapping logic is identified as "mapping logic B".

[0042] Step S1110-7: Based on the determined adjustment strategy, adjust the parameter association weights or mapping logic of the corresponding mapping relationships in the multi-level mapping structure so that the mapping relationships can adapt to the changed hydraulic system operating state, AI prediction model driving state and working condition identification state; in an offline test environment, use historical data or simulated data to verify the data transmission success rate and latency time under the adjusted mapping relationships, and confirm whether they meet the preset verification threshold.

[0043] Based on the adjustment strategy, adjust the parameter association weights or mapping logic of the corresponding mapping relationships in the multi-level mapping structure. For example, adjust the weight of the rated flow parameter of the hydraulic pump when mapping it to the input parameter of the AI ​​model, or switch the mapping logic. In an offline testing environment, use historical data or simulated data to test the adjusted mapping relationship, record the data transmission success rate and latency time, compare them with the preset verification threshold to confirm whether they meet the requirements. If they do not meet the requirements, re-query the adjustment strategy or fine-tune it, and test again until the target is met.

[0044] Step S120: Based on the dynamic coupling and correlation structure, synchronously collect the real-time operation information of the lifting hydraulic system, the real-time drive status information of the AI ​​prediction model, and the real-time working condition feature information of the working condition identification module, and generate a working condition-power mutual drive iterative signal. The working condition-power mutual drive iterative signal includes the feedback drive signal of the hydraulic system to the AI ​​prediction model, the prediction signal of the AI ​​prediction model to the hydraulic power demand, and the correlation drive signal of the working condition features to the power demand.

[0045] During the operation of the two-post lift, the dynamic coupling and correlation structure synchronously collects real-time operating information of the hydraulic system, real-time drive status information of the AI ​​prediction model, and real-time working condition feature information of the working condition identification module at a preset frequency. The real-time operating information of the hydraulic system includes the real-time status of hydraulic oil flow, hydraulic pump operation, hydraulic valve control, and actuator actions; the real-time drive status information of the model includes model input parameters, calculation progress, output signals, and response effects; and the real-time working condition feature information includes working condition type, load, speed, and environmental adaptation characteristics. The above information is time-series labeled and paired, the correlation between features is analyzed, and feedback drive signals, prediction signals, and correlation drive signals are generated. These are integrated to form a working condition-power mutual drive iterative signal, reflecting the dynamic mutual drive relationship among the three.

[0046] Step S121: Through the hydraulic information acquisition port of the dynamically coupled structure, continuously acquire real-time operating information of the lifting machine hydraulic system at a preset acquisition frequency. The real-time operating information includes the real-time status of hydraulic oil flow, the real-time status of hydraulic pump operation, the real-time status of hydraulic valve control, and the real-time status of hydraulic actuator action. Through the model information acquisition port of the dynamically coupled structure, acquire real-time drive status information of the AI ​​prediction model at a rhythm synchronized with the hydraulic information acquisition frequency. The real-time drive status information includes the real-time values ​​of the model's input parameters, the real-time progress of prediction calculation, the real-time status of output signals, and the real-time effect of drive response. Through the working condition information acquisition port of the dynamically coupled structure, acquire real-time working condition feature information of the working condition identification module at a rhythm synchronized with the hydraulic information acquisition frequency. The real-time working condition feature information includes current working condition type identification information, working condition load feature information, working condition operating speed feature information, and working condition environment adaptation feature information.

[0047] The hydraulic information acquisition port of the dynamically coupled and correlated structure is connected to various sensors of the hydraulic system to collect real-time operating information at a preset frequency. The real-time status of hydraulic oil flow is obtained through flow and pressure sensors, such as real-time flow rate and pressure loss along the flow path; the real-time status of hydraulic pump operation is acquired through speed, pressure, and temperature sensors, including real-time speed, outlet pressure, and pump body temperature; the real-time status of hydraulic valve control is acquired by position sensors and control signal acquisition devices, such as valve core position, opening pressure, and control signal voltage; the real-time status of hydraulic actuator movement is acquired through displacement and pressure sensors, such as hydraulic cylinder extension length and cylinder pressure.

[0048] The model information acquisition port connects to the computer running the AI ​​prediction model to synchronously collect real-time drive status information. Real-time input parameter values ​​include specific values ​​for parameters such as hydraulic oil temperature and hydraulic pump speed; the real-time progress of prediction calculation is obtained through the model's internal calculation step counter; the real-time status of output signals includes predicted power demand values ​​and signal stability indicators; the real-time effect of drive response is represented by the percentage deviation between predicted and actual values.

[0049] The working condition information acquisition port is connected to the working condition identification module's sensors and processing unit to synchronously acquire real-time working condition characteristic information. Current working condition type identification information is output by the working condition identification module, such as the "heavy-load rapid lifting working condition" identifier; working condition load characteristic information is represented by the real-time load value and rate of change of the load sensor; working condition operating speed characteristic information is obtained from the real-time speed and acceleration of the speed sensor; working condition environment adaptation characteristic information includes parameters such as ambient temperature, humidity, and ground inclination.

[0050] Step S122: The collected real-time operation information, real-time drive status information, and real-time working condition characteristic information of the hydraulic system are time-series marked. The set of hydraulic real-time information with time-series marking, the set of model real-time information with time-series marking, and the set of working condition real-time information with time-series marking are paired according to time identifier to form an information pairing set. Each information pairing contains hydraulic real-time information, model real-time information, and working condition real-time information at the same time point.

[0051] The three types of information collected are stamped with precise timestamps based on a unified clock source, accurate to the microsecond level, by a timing stamping unit. The time-stamped information is then aggregated into hydraulic, model, and real-time operating condition information sets, respectively. These sets are then paired according to time identifiers, and the three types of information with the same timestamp are combined into information pairs to form information pair sets, ensuring that the system state is reflected at the same point in time.

[0052] Step S123: For the hydraulic real-time information in each information pair, calculate the difference, moving average, or derivative of its specific parameters at the current time relative to the previous N times, and use the calculation results as hydraulic real-time features; for the model real-time information in each information pair, extract the values ​​of its input parameters, model calculation progress identifier, and output signal, and use the extracted content as model real-time features; for the working condition real-time information in each information pair, extract its current working condition type identifier, load sensor reading, speed sensor reading, and environmental sensor reading, and use the extracted content as working condition real-time features.

[0053] For specific parameters of real-time hydraulic information, such as hydraulic pump outlet pressure, hydraulic cylinder pressure, and hydraulic oil temperature, calculate the difference, moving average, and derivative between the current moment and the previous N moments, as real-time hydraulic characteristics, reflecting the dynamic change characteristics of the parameters.

[0054] For real-time model information, extract the values ​​of input parameters, model computation progress identifiers (such as "feature extraction stage"), and output signal values ​​as real-time features of the model.

[0055] For real-time operating condition information, extract the current operating condition type identifier, load sensor reading, speed sensor reading, and environmental sensor reading to form real-time operating condition characteristics.

[0056] Step S124: Input the hydraulic real-time features, model real-time features and working condition real-time features in the same information pair into the correlation analysis model, and output a result matrix representing the correlation coefficient or correlation rules between the hydraulic real-time features, model real-time features and working condition real-time features, as the three-dimensional feature correlation result.

[0057] Hydraulic, model, and real-time operating condition features paired with the same information are combined into a comprehensive feature vector, which is then input into the correlation analysis model. After standardizing the features, the model calculates the correlation coefficient between any two features; positive values ​​indicate a positive correlation, negative values ​​indicate a negative correlation, and the larger the absolute value, the stronger the correlation. Simultaneously, correlation rules are mined, such as the fact that when load and pressure exceed thresholds, the output signal often exceeds a specific value. The correlation coefficients and correlation rules are organized into a result matrix, i.e., the three-dimensional feature correlation result.

[0058] Step S1241: Extract key parameter items of hydraulic real-time features from the same information pair. The key parameter items of hydraulic real-time features can directly reflect the core operating state of the hydraulic system at that point in time. The key parameter items of hydraulic real-time features include hydraulic oil flow change parameters, hydraulic pump working state parameters, hydraulic valve control and adjustment parameters, and hydraulic actuator action response parameters. Extract key parameter items of model real-time features from the same information pair. The key parameter items of model real-time features can directly reflect the core driving state of the AI ​​prediction model at that point in time. The key parameter items of model real-time features include input parameter change parameters, calculation progress parameters, output signal strength parameters, and response effect feedback parameters. Extract key parameter items of working condition real-time features from the same information pair. The key parameter items of working condition real-time features can directly reflect the core state of the current working condition. The key parameter items of working condition real-time features include working condition type identification parameters, load intensity parameters, operating speed parameters, and environmental adaptation parameters.

[0059] Key parameters of real-time hydraulic characteristics are extracted, including hydraulic oil flow change parameters such as flow rate change rate and flow velocity non-uniformity; hydraulic pump operating status parameters such as speed fluctuation rate, outlet pressure pulsation amplitude, and pump body temperature change gradient; hydraulic valve control and adjustment parameters such as control signal response delay time, valve core position adjustment accuracy, and valve opening change; and hydraulic actuator action response parameters such as hydraulic cylinder extension and retraction speed change rate, cylinder pressure build-up time, and action synchronization.

[0060] Key parameters of the model's real-time characteristics include: input parameter variation parameters (including the magnitude, frequency, and synergy of each input parameter); computation progress parameters (including computation time per step, frequency of computation step jumps, and percentage of execution time for each module); output signal strength parameters (including the fluctuation range of the output predicted value, signal-to-noise ratio, and trend); and response feedback parameters (including the deviation rate between predicted and actual values, the trend of deviation, and prediction accuracy evaluation indicators).

[0061] Key parameters for real-time operating conditions include: operating condition type identifier (discrete category identifier); load intensity parameters (real-time weight value, weight change rate, and distribution uniformity); operating speed parameters (real-time rise / fall speed, speed change acceleration, and stability index); and environmental adaptability parameters (ambient temperature influence coefficient, humidity influence degree, and ground flatness influence parameter).

[0062] Step S1242: Group the key parameters of the hydraulic real-time features, the key parameters of the model real-time features, and the key parameters of the operating condition real-time features according to their functional categories to form hydraulic parameter group, model parameter group, and operating condition parameter group. Each group in the hydraulic parameter group, model parameter group, and operating condition parameter group contains multiple key parameters that are functionally related.

[0063] Key parameters of real-time hydraulic characteristics are grouped by function: hydraulic pump speed fluctuation rate, outlet pressure pulsation amplitude, etc. are classified into the hydraulic pump power parameter group; flow rate change rate, valve control signal response delay time, etc. are classified into the hydraulic control parameter group; hydraulic cylinder extension and retraction speed change rate, cylinder pressure build-up time, etc. are classified into the hydraulic execution parameter group.

[0064] The key parameters of the model's real-time features are grouped as follows: the magnitude and frequency of input parameter changes are grouped into the model input processing parameter group; the time consumed in calculation steps and the proportion of module execution time are grouped into the model calculation process parameter group; and the amplitude of output signal fluctuations and prediction deviation rate are grouped into the model output feedback parameter group.

[0065] The key parameters of real-time operating conditions are grouped as follows: operating condition type identifier, real-time load weight value, etc. are classified into the operating condition load core parameter group; real-time rise / fall speed, speed stability index, etc. are classified into the operating condition speed parameter group; ambient temperature influence coefficient, humidity influence degree, etc. are classified into the operating condition environment parameter group.

[0066] Step S1243: Perform time-series variation analysis on each key parameter item in the hydraulic parameter group to determine the variation trend and magnitude of each key parameter item in the hydraulic parameter group at the current time point relative to historical adjacent time points, forming a time-series variation description of hydraulic parameters; perform time-series variation analysis on each key parameter item in the model parameter group to determine the variation trend and magnitude of each key parameter item in the model parameter group at the current time point relative to historical adjacent time points, forming a time-series variation description of model parameters; perform time-series variation analysis on each key parameter item in the operating condition parameter group to determine the variation trend and magnitude of each key parameter item in the operating condition parameter group at the current time point relative to historical adjacent time points, forming a time-series variation description of operating condition parameters.

[0067] For each key parameter item in the hydraulic parameter group, such as the hydraulic pump speed fluctuation rate, take the current and the previous N time points, determine the trend of change (rising, falling, stable), calculate the difference between the current value and the average of the previous N values ​​as the change amplitude, and form a time series description of hydraulic parameter changes.

[0068] The time-series variation analysis methods for model parameter groups and operating condition parameter groups are similar, forming time-series variation descriptions of model parameters and operating condition parameters, respectively.

[0069] Step S1244: Compare the time-series change descriptions of hydraulic parameters, model parameters, and operating conditions parameters parameter one by one, calculate the correlation coefficient of any two parameters in the hydraulic parameter group, model parameter group, and operating conditions parameter group in time, and mark a pair of parameters with a correlation coefficient greater than a preset positive threshold as having a cooperative change relationship.

[0070] The time-series changes of the three types of parameters are compared parameter by parameter. The correlation coefficient of any two parameters is calculated. Parameter pairs with a correlation coefficient greater than a preset positive threshold are marked as having a cooperative relationship, that is, when one parameter changes, the other parameter tends to change in the same direction.

[0071] Step S1245: For parameter pairs marked as having a cooperative change relationship, calculate the average change of the other parameter corresponding to the unit change of one parameter, and record the average change as a description of the cooperative effect.

[0072] For pairs of parameters that change in synergy, such as the amplitude of hydraulic pump outlet pressure pulsation and the real-time weight value of the load, calculate the average change of the other parameter corresponding to the unit change of one parameter, record it as a description of synergistic influence, and quantify the degree of mutual influence.

[0073] Step S1246: Calculate the correlation coefficient of any two parameters in the hydraulic parameter group, model parameter group, and working condition parameter group over time, and mark a pair of parameters with a correlation coefficient less than a preset negative threshold as having a restrictive relationship.

[0074] Calculate the correlation coefficient between any two parameters. Parameter pairs that are less than a preset negative threshold are marked as having a restrictive relationship, meaning that when one parameter increases, the other parameter tends to decrease.

[0075] Step S1247: Analyze the triggering conditions for the changes of each parameter in the constraint relationship, determine the mutual constraint mechanism between each parameter in the constraint relationship, and form a description of the constraint influence.

[0076] Analyze the triggering conditions for changes in the limiting parameters. For example, an increase in the rate of change of the hydraulic cylinder's extension / retraction speed is triggered by an increase in the rate of change of the control signal; a decrease in the speed stability index is triggered by speed fluctuations exceeding the specified range. Identify the mutual constraint mechanisms, such as how an increase in the rate of change of speed leads to a decrease in speed stability, thus forming a description of the limiting effects.

[0077] Step S1248: Integrate the description of synergistic effects and the description of restrictive effects, and supplement the analysis of the time synchronization of parameter changes, the analysis of functional correlation, and the analysis of working condition-power correlation to form a three-dimensional feature correlation result that comprehensively reflects the relationship between the real-time hydraulic characteristics, the real-time model characteristics, and the real-time working condition characteristics.

[0078] The analysis integrates the descriptions of coordination and constraints, supplements them with time synchronization analysis (the degree of synchronization of parameter changes), functional correlation analysis (the functional relationship between parameters), and operating condition-power correlation analysis (the impact of parameter relationships on power demand), forming a three-dimensional feature correlation result.

[0079] Step S125: When the three-dimensional feature association result indicates that a specific parameter in the hydraulic real-time feature exceeds the preset range, an adjustment instruction is generated. The adjustment instruction includes the name of the AI ​​prediction model driving parameter to be adjusted and the adjustment amount.

[0080] The three-dimensional feature association results show that when a specific parameter in the real-time hydraulic features exceeds the preset range, such as the hydraulic oil temperature change rate exceeding the range, an adjustment instruction is generated based on the correlation between the parameter and the model driving parameters. This instruction includes the name of the model driving parameter to be adjusted and the adjustment amount, in order to compensate for the impact on the prediction.

[0081] Step S126: Based on the three-dimensional feature association results, generate a prediction signal for hydraulic power demand by the AI ​​prediction model. This prediction signal includes preliminary prediction data and adaptation adjustment suggestions for hydraulic power demand under the coupling of real-time features of the model and real-time features of the working condition. Based on the three-dimensional feature association results, generate a correlation driving signal for power demand based on working condition features. This correlation driving signal includes key feature parameters and adjustment direction suggestions in the real-time features of the working condition that affect hydraulic power demand.

[0082] Based on the 3D feature correlation results, the AI ​​prediction model couples the model with real-time operating conditions to calculate preliminary prediction data and adaptation adjustment suggestions for hydraulic power demand, forming a prediction signal. The operating condition feature-correlated drive signal contains key operating condition feature parameters that affect power demand and adjustment direction suggestions, such as suggesting an increase in power demand when the load increases.

[0083] Step S127: Integrate the feedback drive signal, prediction signal and associated drive signal according to time identifier to form a working condition-power mutual drive iteration signal containing three-way drive information. Each working condition-power mutual drive iteration signal corresponds to a collaborative prediction demand at a time point.

[0084] Feedback, prediction, and associated drive signals are integrated according to time identifiers to form a working condition-power mutual drive iterative signal, which contains three-way drive information and corresponds to the collaborative prediction requirements at each time point.

[0085] Step S130: Input the working condition-power mutual drive iterative signal into the dynamic rule evolution unit, and combine the operating characteristics of the lifting hydraulic system, the prediction characteristics of the AI ​​prediction model and the dynamic adaptation characteristics of working condition identification to evolve into a working condition-power demand prediction rule. The working condition-power demand prediction rule includes the adjustment logic of the mutual drive signal and the adaptation standard of power demand prediction.

[0086] The dynamic rule evolution unit receives ordered working condition-power mutual drive iteration signals, analyzes the core driving elements, combines the hydraulic system operating characteristics (response speed, load capacity, etc.), AI model prediction characteristics (accuracy, response speed, etc.) and working condition identification dynamic adaptation characteristics (accuracy, switching response, etc.), filters and classifies effective elements, constructs adjustment logic and adaptation standards for various elements, and integrates them to form working condition-power demand prediction rules, including mutual drive signal adjustment logic and power demand prediction adaptation standards.

[0087] Step S131: The dynamic rule evolution unit receives the continuous time series of working condition-power mutual drive iteration signals, sorts the working condition-power mutual drive iteration signals in time order to form an ordered mutual drive signal sequence, which reflects the dynamic mutual drive process between the hydraulic system, the AI ​​prediction model and the working condition identification module.

[0088] The dynamic rule evolution unit receives continuous working condition-power mutual drive iteration signals, sorts them according to time identifiers to form an ordered mutual drive signal sequence, reflecting the dynamic mutual drive process of the three.

[0089] Step S132: From each signal in the ordered mutual drive signal sequence, extract the adjustment command field in the feedback drive signal, the prediction value field in the prediction signal, and the characteristic parameter field in the associated drive signal, and store all the extracted fields as a core drive element set; from the hydraulic system technical manual or configuration file, read the response delay time, adjustable range of key parameters, communication protocol between functional modules, and rated load parameters of the hydraulic system as hydraulic system operating characteristic data.

[0090] The adjustment instructions, predicted values, and characteristic parameter fields of each signal in the ordered mutual drive signal sequence are analyzed and stored as a set of core drive elements. The hydraulic system response delay time, adjustable range of key parameters, communication protocol, and rated load parameters are read from the technical manual or configuration file as hydraulic system operating characteristic data.

[0091] Step S133: Analyze the prediction characteristics of the AI ​​prediction model, which include the prediction accuracy characteristics, drive response speed characteristics, parameter adjustment range characteristics, and multi-scenario adaptation characteristics of the model, forming a description of the prediction characteristics of the AI ​​prediction model; and analyze the dynamic adaptation characteristics of working condition recognition, which include the response speed characteristics of working condition recognition, feature extraction accuracy characteristics, working condition switching recognition sensitivity characteristics, and power demand correlation adaptation characteristics, forming a description of working condition recognition adaptation characteristics.

[0092] Analyze the predictive characteristics of the AI ​​prediction model, including prediction accuracy (prediction bias rate, mean squared error), drive response speed (response time), parameter adjustment range (adjustable upper and lower limits of parameters, step size), and multi-scenario adaptation characteristics (prediction performance under different operating conditions), to form a description of prediction characteristics. Analyze the dynamic adaptation characteristics of operating condition recognition, including response speed (recognition time), feature extraction accuracy (extraction error), operating condition switching recognition sensitivity (minimum change in switching features), and adaptation characteristics related to power demand (correlation between recognition results and power demand), to form a description of adaptation characteristics.

[0093] Step S134: Compare each adjustment instruction field, predicted value field, or feature parameter field in the core drive element set with the corresponding parameter range or constraint condition in the hydraulic system operating characteristic data to determine whether it is within the allowable range, and record the comparison result as the element hydraulic adaptation result; perform association matching between the core drive element set and the AI ​​prediction model prediction characteristic description to identify the degree of adaptation between each core drive element and the AI ​​prediction model prediction characteristic, forming the element model adaptation result; perform association matching between the core drive element set and the working condition identification adaptation characteristic description to identify the degree of adaptation between each core drive element and the working condition identification adaptation characteristic, forming the element working condition adaptation result.

[0094] The core driving elements are compared with the hydraulic system operating characteristic data to determine if they are within the allowable range, and the hydraulic adaptation results of the elements are recorded. The core driving elements are then matched with the AI ​​model's predicted characteristic descriptions to identify the degree of adaptation and generate element model adaptation results. Finally, the core driving elements are matched with the operating condition identification adaptation characteristic descriptions to identify the degree of adaptation and generate element operating condition adaptation results.

[0095] Step S135: Select the core driving elements that are judged to be within the allowable range in the element hydraulic adaptation results, element model adaptation results, and element working condition adaptation results, and form a set of effective elements. Classify and organize the effective core driving elements in the set of effective elements, and divide them into parameter adjustment elements, prediction adaptation elements, working condition related elements, and collaborative optimization elements according to their functions, and form a set of classified elements.

[0096] Core driving elements whose hydraulic parameters, model, and operating condition adaptation results are all within acceptable ranges are selected to form a set of effective elements. Based on their function, these effective elements are categorized into parameter adjustment (adjustment instruction fields), prediction adaptation (predicted value fields and adaptation adjustment suggestions), operating condition association (feature parameter fields), and collaborative optimization (elements requiring coordinated adjustment of all three), thus forming a set of categorized elements.

[0097] Step S136: For each category of element in the categories of parameter adjustment, prediction adaptation, working condition association, and collaborative optimization, construct the corresponding adjustment logic and adaptation standard. The adjustment logic of the category determines the correlation between the change pattern of the category and the power demand prediction. The adaptation standard of the category determines the adaptation requirements between the category and the hydraulic system, AI prediction model, and working condition identification module.

[0098] For each category of elements, adjustment logic and adaptation standards are constructed. The adjustment logic determines the correlation between the element's change patterns and the forecast of driving demand, while the adaptation standards determine the adaptation requirements between the elements and various parts of the system.

[0099] Step S1361: For parameter adjustment elements, analyze the relationship between the change pattern of the parameter adjustment elements and the hydraulic system control parameters and AI prediction model driving parameters, and establish the correspondence criteria between the change direction of the parameter adjustment elements and the adjustment direction of the hydraulic system control parameters and AI prediction model driving parameters.

[0100] Analyze the relationship between the change patterns of parameter adjustment elements and the control parameters of the hydraulic system and the driving parameters of the AI ​​model, and establish a criterion for the direction of change. For example, when the parameter adjustment elements increase, the control parameters of the hydraulic system and the driving parameters of the AI ​​model increase in the same direction.

[0101] Step S1362: Based on the correspondence between the change direction of the parameter adjustment category element and the adjustment direction of the hydraulic system control parameters and AI prediction model driving parameters, construct the adjustment logic of the parameter adjustment category element. The data processing function of the parameter adjustment category element defines the value of the parameter adjustment category element and maps it to the adjustment amount of the hydraulic system control parameters or AI prediction model driving parameters through a preset conversion function. When multiple parameter adjustment category elements are triggered at the same time, the execution order of their data processing functions is determined by a preset priority list.

[0102] Based on corresponding criteria, adjustment logic is constructed. The data processing function maps the numerical values ​​of parameter adjustment elements to the adjustment amounts of hydraulic system control parameters or AI model driven parameters through a transformation function. When multiple elements are triggered simultaneously, the data processing function is executed according to a preset priority list.

[0103] Step S1363: Combining the parameter adjustment sensitivity characteristics of the hydraulic system and the parameter adjustment range characteristics of the AI ​​prediction model, construct the adaptation standard for the parameter adjustment category elements. The adaptation standard for the parameter adjustment category elements determines the allowable range of the change amplitude of the parameter adjustment category elements and the safety boundary for the adjustment of hydraulic system control parameters and AI prediction model drive parameters.

[0104] By combining the sensitivity of hydraulic system parameter adjustment (the change in system output caused by a unit adjustment) and the parameter adjustment range of AI model, an adaptation standard is constructed to determine the allowable range of parameter adjustment element changes and the safety boundary of parameter adjustment.

[0105] Step S1364: For the predictive adaptation category elements, analyze the relationship between the change pattern of the predictive adaptation category elements and the prediction parameters of the AI ​​prediction model and the prediction results of power demand, and establish the correspondence criteria between the change direction of the predictive adaptation category elements and the adjustment direction of the prediction parameters of the AI ​​prediction model and the prediction results of power demand.

[0106] Analyze the correlation between the changing patterns of adaptive elements and the AI ​​model's prediction parameters and the predicted results of energy demand, and establish a criterion for the direction of change. For example, when the predicted data increases, the model's prediction parameters and the predicted results of energy demand increase in the same direction.

[0107] Step S1365: Based on the correspondence between the direction of change of the prediction adaptation element and the direction of adjustment of the AI ​​prediction model prediction parameters and the power demand prediction results, construct the adjustment logic of the prediction adaptation element. The adjustment logic of the prediction adaptation element includes the correlation between the change of the prediction adaptation element and the adjustment of the AI ​​prediction model prediction parameters and the power demand prediction results, as well as the priority ranking method when multiple prediction adaptation elements act simultaneously.

[0108] The adjustment logic is constructed based on the corresponding criteria, including the correlation between changes in factors and adjustments to model prediction parameters and power demand prediction results, as well as the priority ranking method when multiple factors act simultaneously.

[0109] Step S1366: Combining the prediction accuracy characteristics of the AI ​​prediction model with the adaptation requirements of power demand prediction, construct the adaptation standard for the prediction adaptation category elements. The adaptation standard for the prediction adaptation category elements determines the allowable range of the change range of the prediction adaptation category elements and the accuracy boundary of the adjustment of the prediction parameters of the AI ​​prediction model and the prediction results of power demand.

[0110] By combining the prediction accuracy characteristics of AI models with the adaptation requirements of power demand prediction, an adaptation standard is constructed to determine the allowable range of changes in prediction adaptation elements and parameters, as well as the accuracy boundaries for result adjustment.

[0111] Step S1367: For the working condition related elements, analyze the relationship between the change pattern of the working condition related elements and the working condition identification parameters and power demand prediction results, and establish the correspondence criteria between the change direction of the working condition related elements and the adjustment direction of the working condition identification parameters and power demand prediction results.

[0112] Analyze the relationship between the changing patterns of operating condition-related elements and the operating condition identification parameters and power demand prediction results, and establish a criterion for the direction of change. For example, when the load weight increases, the operating condition identification parameters and power demand prediction results increase in the same direction.

[0113] Step S1368: Based on the correspondence between the change direction of the working condition related elements and the adjustment direction of the working condition identification parameters and power demand prediction results, construct the adjustment logic of the working condition related elements. The adjustment logic of the working condition related elements includes the correlation between the change of the working condition related elements and the adjustment of the working condition identification parameters and power demand prediction results, as well as the priority sorting method when multiple working condition related elements act simultaneously.

[0114] The adjustment logic is constructed based on the corresponding criteria, including the correlation between changes in elements and the adjustment of operating condition identification parameters and power demand forecast results, as well as the priority ranking method when multiple elements act simultaneously.

[0115] Step S1369: Combining the dynamic adaptation characteristics of working condition identification and the adaptation requirements of power demand prediction, construct the adaptation standard for the working condition-related elements. The adaptation standard for the working condition-related elements determines the allowable range of changes in the working condition-related elements and the adaptation boundary for adjusting the working condition identification parameters and power demand prediction results.

[0116] By combining the dynamic adaptation characteristics of working condition identification and the adaptation requirements of power demand prediction, an adaptation standard is constructed to determine the allowable range of changes in working condition-related elements and the adaptation boundaries for parameter and result adjustment.

[0117] Step S13610: For collaborative optimization elements, analyze the relationship between the change pattern of the collaborative optimization elements and the hydraulic system parameters, AI prediction model parameters, and working condition identification parameters, and establish the correspondence criteria between the change direction of the collaborative optimization elements and the adjustment direction of the hydraulic system parameters, AI prediction model parameters, and working condition identification parameters.

[0118] The correlation between the changing patterns of collaborative optimization elements and hydraulic system parameters, AI model parameters, and working condition identification parameters is analyzed, and a criterion for the direction of change is established. For example, when the overall efficiency index decreases, the hydraulic system parameters, model parameters, and working condition identification parameters are adjusted in a specific direction.

[0119] Step S13611: Based on the correspondence between the change direction of the collaborative optimization element and the adjustment direction of the hydraulic system parameters, AI prediction model parameters, and working condition identification parameters, construct the adjustment logic of the collaborative optimization element. The adjustment logic of the collaborative optimization element includes the correlation between the change of the collaborative optimization element and the adjustment of the hydraulic system parameters, AI prediction model parameters, and working condition identification parameters, as well as the priority sorting method when multiple collaborative optimization elements act simultaneously.

[0120] The adjustment logic is constructed based on the corresponding criteria, including the relationship between changes in elements and adjustments of three types of parameters, as well as the priority ranking method when multiple elements act simultaneously.

[0121] Step S13612: Combining the operating characteristics of the hydraulic system, the predictive characteristics of the AI ​​prediction model, and the dynamic adaptation characteristics of the working condition identification, construct the adaptation standard for the collaborative optimization element. The adaptation standard for the collaborative optimization element determines the allowable range of the variation of the collaborative optimization element and the collaborative boundary for adjusting the hydraulic system parameters, AI prediction model parameters, and working condition identification parameters.

[0122] By combining the operating characteristics of the hydraulic system, the predictive characteristics of the AI ​​model, and the dynamic adaptation characteristics of working condition identification, an adaptation standard is constructed to determine the allowable range of changes in collaborative optimization elements and the collaborative boundaries of the adjustment of three types of parameters.

[0123] Step S137: The adjustment logic and adaptation standards of parameter adjustment elements, prediction adaptation elements, operating condition related elements and collaborative optimization elements are structurally integrated to form operating condition-power demand prediction rules. These operating condition-power demand prediction rules include mutual drive adjustment schemes and power demand prediction standards under all scenarios.

[0124] Integrate the adjustment logic and adaptation standards of various elements, establish the correlation and interaction mechanism between elements, form a mutual drive adjustment scheme and power demand prediction standard under the whole scenario, and describe it in a structured way as the working condition-power demand prediction rule.

[0125] Step S140: Using the working condition-power demand prediction rule as the processing logic, calculate the continuously input real-time operating information of the hydraulic system, the predicted state data of the AI ​​prediction model, and the real-time characteristic data of the working condition, and output the hydraulic power demand prediction result. The hydraulic power demand prediction result is transmitted to the lifting machine hydraulic system control unit through a dynamic coupling association structure.

[0126] Using the working condition-power demand prediction rule as the processing logic, the system continuously receives real-time operating information change data of the hydraulic system, AI model prediction status update data, and real-time characteristic change data of the working condition. This data is accumulated to form a data sequence, which is then input into the rule matching unit to filter the target adjustment logic. Each data sequence is analyzed to determine the degree of influence, optimization direction, and related influence direction. The adjustment direction is matched collaboratively and conflicts are eliminated. Data processing functions are called to calculate parameter values ​​and generate control sequences. The results are integrated to form the hydraulic power demand prediction result, which is then transmitted to the hydraulic system control unit through a dynamic coupling and correlation structure.

[0127] Step S141: Continuously receive change data of the real-time operation information of the lifting hydraulic system through the change data acquisition port of the dynamic coupling and association structure. The change data includes fluctuation data, state transition data, and response change data of the operating parameters of each functional module of the hydraulic system. Continuously receive update data of the predicted state of the AI ​​prediction model through the update data acquisition port of the dynamic coupling and association structure. The update data includes update data of the model input parameters, adjustment data of the prediction calculation logic, and optimization data of the output signal. Continuously receive change data of the real-time characteristics of the working conditions through the working condition change data acquisition port of the dynamic coupling and association structure. The change data includes working condition type switching data, working condition load fluctuation data, working condition environment adaptation change data, and working condition characteristic parameter adjustment data.

[0128] The dynamic coupling and correlation structure change data acquisition port receives fluctuation data (minor parameter changes), state transition data (working state transitions), and response change data (response to control signals) from various functional modules of the hydraulic system. The update data acquisition port receives AI model input parameter update data, predictive calculation logic adjustment data, and output signal optimization data. The operating condition change data acquisition port receives operating condition type switching data, load fluctuation data, environmental adaptation change data, and characteristic parameter adjustment data.

[0129] Step S142: Accumulate hydraulic system change data in chronological order to form a hydraulic system change data sequence, which reflects the dynamic evolution of the real-time operating state of the hydraulic system; accumulate AI prediction model update data in chronological order to form an AI prediction model update data sequence, which reflects the dynamic evolution of the predicted state by the AI ​​prediction model; accumulate real-time operating condition characteristic change data in chronological order to form an operating condition characteristic change data sequence, which reflects the dynamic evolution of the real-time operating state.

[0130] Hydraulic system change data is accumulated chronologically to form a hydraulic system change data sequence, reflecting the dynamic evolution of the system's operating state. AI model update data is accumulated to form an AI model update data sequence, reflecting the dynamic evolution of the model's predicted state. Real-time operating condition characteristic change data is accumulated to form an operating condition characteristic change data sequence, reflecting the dynamic evolution of the real-time operating state.

[0131] Step S143: Input the hydraulic system change data sequence, AI prediction model update data sequence, and working condition characteristic change data sequence into the matching unit of the working condition-power demand prediction rule, and filter out the target adjustment logic corresponding to the current change data according to the adaptation criteria in the working condition-power demand prediction rule.

[0132] The three data sequences are input into the matching unit of the working condition-power demand prediction rule. After preprocessing, key features are extracted and compared with the feature templates in the adaptation standard to select the target adjustment logic.

[0133] Step S144: Based on the target adjustment logic, analyze the key fluctuation data in the hydraulic system change data sequence to determine the degree of impact and adjustment direction of the hydraulic system operating state changes on power demand; based on the target adjustment logic, analyze the key update data in the AI ​​prediction model update data sequence to determine the optimization direction and adjustment parameters of the AI ​​prediction model's prediction of state changes on power demand prediction; based on the target adjustment logic, analyze the key change data in the working condition characteristic change data sequence to determine the direction of the correlation between dynamic changes in working conditions and power demand, and adaptation adjustment suggestions.

[0134] Based on the target adjustment logic, we analyze key fluctuation data of the hydraulic system to determine the degree of impact on power demand and the direction of adjustment; we analyze key update data of the AI ​​model to determine the optimization direction and adjustment parameters for power demand prediction; and we analyze key change data of operating conditions to determine the direction of the related impact on power demand and make adaptation adjustment suggestions.

[0135] Step S145: Perform coordinated matching of the hydraulic system influence adjustment direction, the AI ​​prediction model prediction optimization direction, and the working condition related influence direction. If a conflict is detected, modify one or more adjustment directions according to the preset conflict resolution strategy until the conflict is eliminated. Based on the adjustment direction after the conflict is eliminated, call the corresponding data processing function in the working condition-power demand prediction rule to calculate the final value of the power demand prediction parameter, the adjustment value of the input parameter of the AI ​​prediction model, and the feature weight adjustment value of the working condition identification module, and generate the corresponding control sequence.

[0136] In the application scenario of car lifts, the direction of hydraulic system adjustment might be "increase hydraulic power demand by 10% to cope with pressure fluctuations," the direction of AI prediction model optimization might be "reduce the load weight in the model input parameters by 5% to balance prediction bias," and the direction of working condition-related influence might be "based on the characteristics of heavy-load working conditions, it is recommended to increase power demand by 15%." First, these three directions are coordinated and matched to check for any inconsistencies in objectives.

[0137] Step S1451: List all specific contents of the hydraulic system's influence on the adjustment direction. For each hydraulic system's influence on the adjustment direction, determine the corresponding hydraulic system parameter category and the specific impact on power demand, forming a list of hydraulic influence directions; list all specific contents of the AI ​​prediction model's prediction optimization direction. For each AI prediction model's prediction optimization direction, determine the corresponding AI prediction model parameter category and the specific optimization direction for power demand prediction, forming a list of model optimization directions; list all specific contents of the working condition's related influence direction. For each working condition's related influence direction, determine the corresponding working condition characteristic parameter category and the specific correlation with power demand, forming a list of working condition related directions.

[0138] The list of hydraulic impact directions details each adjustment, such as "Adjustment of hydraulic pump outlet pressure fluctuation: the corresponding hydraulic system parameter category is pressure control parameter, and the specific impact on power demand is an increase of 8%" and "Adjustment of insufficient hydraulic oil flow: the corresponding parameter category is flow control parameter, and the impact is an increase of 5% in power demand." Each item clearly specifies the parameter category and the direction of impact.

[0139] The list of model optimization directions includes items such as "Load parameter weight adjustment: the corresponding AI prediction model parameter category is input layer weight parameter, and the specific optimization direction for power demand prediction is to reduce the predicted value by 3%" and "Model output filter coefficient adjustment: the corresponding parameter category is output layer processing parameter, and the optimization direction is to smooth the prediction curve and reduce the fluctuation amplitude by 2%", clearly defining the parameter categories and optimization directions.

[0140] The list of operating condition correlation directions includes items such as "Heavy load condition load characteristics: the corresponding operating condition characteristic parameter category is load intensity parameter, and the specific correlation with power demand is an increase of 12%" and "Ambient temperature rise characteristics: the corresponding parameter category is environmental adaptation parameter, and the correlation with power demand is an increase of 3%".

[0141] Step S1452: Pair the hydraulic influence direction list, model optimization direction list, and working condition association direction list according to their functional association to form a set of functionally associated adjustment direction pairs. Each adjustment direction pair includes a hydraulic system influence adjustment direction, an AI prediction model prediction optimization direction, and a working condition association influence direction.

[0142] Based on functional relevance, adjustment directions targeting the same optimization objective or related parameters in the three lists are paired. For example, "Hydraulic pump outlet pressure fluctuation adjustment (increase by 8%)", "load parameter weight adjustment (decrease by 3%)", and "heavy-load condition load characteristics (increase by 12%)" are paired as one adjustment direction because they are all related to load and pressure, jointly affecting the prediction and adjustment of power demand. Through this method, multiple sets of functionally related adjustment direction pairs are formed.

[0143] Step S1453: Analyze the consistency of the objectives of the hydraulic system's influence on the adjustment direction, the AI ​​prediction model's prediction of the optimization direction, and the working condition's influence on the direction in each adjustment direction pair. Determine whether the three are predicting the same power demand optimization objective and form an objective consistency judgment result.

[0144] For the adjustment direction pairings in the above examples, analyze whether the objectives of the three are consistent. Hydraulic system adjustment aims to ensure power output in response to pressure fluctuations; model optimization aims to balance the impact of load weights on predictions; and operating condition correlation determines power demand based on heavy-load operating condition characteristics. The ultimate goal of all three is to make the power demand prediction more consistent with the actual needs of current heavy-load lifting. Therefore, the result of the objective consistency judgment is "consistent." If, in a pairing, the hydraulic system direction is to increase power, the model direction is to decrease the predicted value, and the operating condition direction is to remain unchanged, and the three are not targeting the same load or speed objective, then the result of the objective consistency judgment is "inconsistent."

[0145] Step S1454: For adjustment direction pairs with consistent target consistency judgment results, analyze whether the hydraulic system's influence on the execution of the adjustment direction and the AI ​​prediction model's prediction of the optimization direction can support the execution of the working condition-related influence direction in the adjustment direction pair, forming a complementary analysis result; for adjustment direction pairs with inconsistent target consistency judgment results, analyze whether the hydraulic system's influence on the execution of the adjustment direction and the AI ​​prediction model's prediction of the optimization direction in the adjustment direction pair will hinder the execution of the working condition-related influence direction in the adjustment direction pair or lead to a decrease in the overall prediction effect, forming a conflict analysis result.

[0146] In the pairing with consistent objectives, the analysis shows that an 8% increase in power in the hydraulic system can directly provide basic support for the 12% increase in demand related to the working conditions. The model's reduction of the load weight by 3% may be due to a certain overestimation in the current load detection. Optimization can make the prediction more accurate, thereby better matching the working conditions. The execution of the three can support each other, and the result of the complementarity analysis is "strongly complementary".

[0147] In pairings where objectives are inconsistent, the hydraulic system needs to increase power, the model needs to significantly reduce the predicted value, while the operating conditions need to be maintained. The execution of the hydraulic system and the model may result in the operating conditions not being met, hindering the overall prediction effect. The conflict analysis result is "severe conflict".

[0148] Step S1455: Based on the complementarity analysis results, retain the adjustment direction pairs with strong complementarity, determine the collaborative execution order and mutual cooperation mode of the adjustment direction pairs, and form a collaborative pair set; based on the conflict analysis results, adjust the adjustment direction pairs with conflicts, modify the specific influence direction of the hydraulic system on the adjustment direction, the specific optimization direction of the AI ​​prediction model prediction optimization direction, or the specific association direction of the working condition correlation influence direction in the adjustment direction pair, or reallocate the adjustment priority of the three in the adjustment direction pair to eliminate the conflict.

[0149] For "strongly complementary" pairings, the order of collaborative execution is determined, such as first adjusting the load weight in the direction of model optimization, then increasing power in the direction of hydraulic system, and finally fine-tuning in the direction of working condition correlation. The mutual cooperation method is also clarified, such as using the data after model optimization as the input of hydraulic adjustment to form a collaborative pairing set.

[0150] For pairings with "serious conflicts", the hydraulic system and model orientation are modified according to the preset conflict resolution strategy, such as the "operating condition priority" principle. The hydraulic increase power is changed to maintenance, and the model decrease prediction value is changed to a slight decrease of 1% to adapt to the operating condition maintenance requirements and eliminate the conflict.

[0151] Step S1456: Perform a target consistency review on the adjusted conflict-free adjustment direction pairings to ensure that the modified adjustment direction pairings can serve the unified power demand prediction optimization target, forming a set of qualified pairings after review. Integrate the collaborative pairing set and the set of qualified pairings after review to form a collaborative matching result that includes all hydraulic system influence adjustment directions, AI prediction model prediction optimization directions, and working condition related influence directions. Determine the execution order, coordination method, and priority of each hydraulic system influence adjustment direction, AI prediction model prediction optimization direction, and working condition related influence direction.

[0152] The adjusted pairings undergo a target consistency recheck to ensure that all modified directions serve a unified objective. If, after the aforementioned conflicting pairing adjustments, all three serve the objective of "stable power output under heavy load conditions," the recheck is passed, forming a rechecked and qualified pairing set. The collaborative pairing set is then integrated with the rechecked and qualified pairing set, clarifying the execution order (e.g., operating condition-related directions take precedence over hydraulic and model directions), coordination methods (e.g., data sharing, real-time feedback), and priorities (e.g., safety-related adjustments have the highest priority), resulting in a collaborative matching result.

[0153] Based on the adjustment direction after conflict resolution, such as determining the final adjustment direction to be influenced by both the overall working condition and the hydraulic system, the power demand is increased by 10%, and the model load weight is adjusted to decrease by 2%. The corresponding "heavy-load power calculation function" in the working condition-power demand prediction rule is invoked. This function calculates the final value of the power demand prediction parameters based on the input hydraulic system pressure fluctuation data, working condition load data, and model parameter adjustment values, using a preset algorithm (such as weighted summation, multi-factor regression, etc.), e.g., "The final hydraulic power demand is XXX Pascals." Simultaneously, the input parameter adjustment values ​​for the AI ​​prediction model are calculated, e.g., "The load parameter input weight is adjusted to 0.75," and the feature weight adjustment values ​​for the working condition identification module are calculated, e.g., "The heavy-load working condition feature weight is increased by 0.2." Based on these calculation results, a corresponding control sequence is generated. This control sequence contains control instructions arranged in chronological order, such as "At time t1: Adjust the hydraulic pump displacement to XXX; at time t2: Update the model load weight parameters; at time t3: Adjust the working condition identification feature threshold," ensuring that all adjustments are executed in an orderly and accurate manner.

[0154] Step S146: Integrate the power demand prediction parameter adjustment scheme, model optimization scheme, and working condition adaptation scheme to form a hydraulic power demand prediction result that includes the core parameters of hydraulic power demand prediction, AI prediction model optimization instructions, working condition adaptation adjustment instructions, and the collaborative execution rules of the three.

[0155] By integrating power demand forecasting parameter adjustment schemes, model optimization schemes, and operating condition adaptation schemes, a hydraulic power demand forecasting result is formed, which includes core parameters, optimization instructions, adaptation adjustment instructions, and collaborative execution rules.

[0156] Step S147: The hydraulic power demand prediction results are transmitted to the lift hydraulic system control unit through a dynamic coupling association structure.

[0157] After the hydraulic power demand forecast is generated, it needs to be transmitted quickly and accurately to the lift's hydraulic system control unit so that the control unit can adjust the hydraulic system's operating status in a timely manner. The dynamic coupling and correlation structure acts as an information superhighway, ensuring that the forecast results are transmitted according to a predetermined path and format, and maintaining data integrity and security during transmission. The transmission process must adhere to certain communication protocols and timing requirements to match the receiving capabilities of the hydraulic system control unit. Upon receiving the forecast results, the control unit will precisely control the hydraulic pumps, hydraulic valves, hydraulic cylinders, and other actuators based on the core parameters, optimization instructions, adaptation adjustment instructions, and collaborative execution rules. This enables on-demand supply of hydraulic power to the lift, improving the lift's working efficiency and safety.

[0158] For example, in step S1471: according to the protocol defined by the transmission port of the dynamic coupling association structure, the hydraulic power demand prediction result is encapsulated and a check code is added to the encapsulated data; the hydraulic power demand prediction result after format conversion is split into multiple transmission data blocks according to functional modules, and each transmission data block corresponds to a core control module of the hydraulic system control unit, so as to facilitate the hydraulic system control unit to receive and process the data in modules.

[0159] The transmission ports of the dynamically coupled associated structure define specific communication protocols, such as Modbus and Profinet protocols. These protocols specify the data transmission format, rate, and verification method. First, the hydraulic power demand prediction results are encapsulated according to the protocol requirements. The core parameters of the hydraulic power demand prediction, AI prediction model optimization instructions, working condition adaptation adjustment instructions, and collaborative execution rules in the prediction results are organized according to the data frame format specified by the protocol, such as frame header, data length, data body, and frame tail.

[0160] To ensure the accuracy of data transmission, a checksum is added to the encapsulated data. The checksum can be calculated using algorithms such as CRC (Cyclic Redundancy Check) or checksum. For example, a CRC16 calculation is performed on all bytes in the data body, and the resulting checksum is appended to the checksum field of the data frame.

[0161] Then, the encapsulated and verified hydraulic power demand forecast results are broken down into functional modules. A hydraulic system control unit typically contains multiple core control modules, such as a hydraulic pump control module, a hydraulic valve control module, a hydraulic cylinder control module, and a safety monitoring module. Data related to hydraulic pump control (such as target pump pressure and flow commands) in the forecast results is split into one transmission data block; data related to hydraulic valve control (such as valve opening / closing commands and opening adjustment commands) is split into another transmission data block; and so on. Each transmission data block contains all the data required by the corresponding core control module.

[0162] Step S1472: Attach a function identifier and a timing identifier to each transmitted data block. The function identifier is used by the hydraulic system control unit to identify the control module corresponding to the transmitted data block, and the timing identifier is used to ensure that the receiving order of the transmitted data blocks is consistent with the execution order.

[0163] Each transmitted data block is appended with a function identifier, which is a unique code or string. For example, "PUMP_CTRL" indicates a data block for the hydraulic pump control module, and "VALVE_CTRL" indicates a data block for the hydraulic valve control module. After receiving the transmitted data block, the hydraulic system control unit can determine which core control module the data block should be sent to for processing by parsing the function identifier.

[0164] A timing identifier is a sequence number or timestamp generated in chronological order, such as "001", "002", ... or "t1", "t2", ... . A timing identifier is attached to each transmitted data block to ensure that the hydraulic system control unit receives the transmitted data blocks in the order specified by the timing identifier and executes the control commands within them in that order, avoiding control errors caused by disordered data block reception. For example, the data block with timing identifier "001" should be received and executed before the data block with timing identifier "002".

[0165] Step S1473: Through the prediction result transmission port of the dynamic coupling association structure, send each transmission data block sequentially according to the timing identifier, and enable the acknowledgment and retransmission mechanism of the transport layer.

[0166] The prediction result transmission port of the dynamically coupled associated structure is a physical or logical interface specifically used for transmitting prediction result data. Data blocks, appended with function and timing identifiers, are sequentially sent through this port according to the order of their timing identifiers. During transmission, the transport layer's acknowledgment and retransmission mechanisms are enabled to ensure reliable data transmission.

[0167] After sending a data block, the sender waits for the receiver (hydraulic system control unit) to return an acknowledgment (ACK). If an ACK is received within a preset time, the sender continues to send the next data block. If no ACK is received or a negative acknowledgment (NACK) is received, the data block is considered to have failed to be transmitted, and the sender will retransmit the data block. The number of retransmissions can be preset (e.g., 3 times). If multiple retransmissions still fail, an alarm mechanism is triggered.

[0168] Step S1474: The prediction result receiving unit of the lifting hydraulic system control unit receives the transmission data block in real time, and allocates the transmission data block to the corresponding core control module processing unit according to the function identifier. Each core control module processing unit parses the received transmission data block, verifies the integrity of the data, and after the verification is passed, extracts the power demand prediction parameters, control commands, execution timestamps and action identifiers that need to be synchronized with other modules allocated to this module.

[0169] The prediction result receiving unit of the lift hydraulic system control unit continuously monitors the prediction result transmission port of the dynamically coupled associated structure and receives transmitted data blocks in real time. Based on the function identifier of each transmitted data block, the receiving unit allocates it to the corresponding core control module processing unit through an internal routing mechanism; for example, it allocates the data block with the function identifier "PUMP_CTRL" to the hydraulic pump control module processing unit.

[0170] Each core control module processing unit parses the received transmitted data block, first extracting the checksum field, and then using the same checksum algorithm as the sender (such as CRC16) to perform checksum calculation on the data body in the data block. The calculated checksum is compared with the checksum field in the data block. If they match, the data integrity check passes; if they do not match, the data block is discarded, and a NACK message is returned to the sender requesting retransmission.

[0171] After the verification is passed, the core control module processing unit performs structured parsing on the data block and extracts the power demand prediction parameters (such as the target pressure value of the hydraulic pump), control commands (such as "start the hydraulic pump" and "adjust the displacement to XXX"), execution timestamps (such as "execute at t5"), and action identifiers that need to be synchronized with other modules (such as "coordinate with the hydraulic valve module A").

[0172] Step S1475: The core control module processing unit compares the parameters in the extracted control command with its own safe operating parameter range. If they are within the safe range, it prepares to execute; otherwise, it triggers a safety anomaly. The core control module processing units share coordination requirement information through internal communication links and coordinate the execution timing so that multiple core control modules can operate collaboratively according to the requirements of the hydraulic power demand prediction results. The overall coordination unit of the hydraulic system control unit summarizes the execution preparation status information of each core control module and confirms that all core control modules have completed control information parsing and adaptation preparation.

[0173] The core control module processing unit (such as the hydraulic pump control module processing unit) compares the parameters (such as the target pressure value) extracted from the control command with its own stored safe operating parameter range (such as the maximum allowable working pressure of the hydraulic pump). If the target pressure value is within the maximum allowable working pressure range, the control command parameters are deemed safe, and the module enters the execution preparation state, such as initializing relevant control registers and setting control algorithm parameters. If the target pressure value exceeds the safe range, a safety anomaly is immediately triggered, and the safety anomaly information is sent to the fault handling module of the hydraulic system control unit. The fault handling module may take measures such as shutdown, alarm, or execution of preset safety plans.

[0174] Each core control module processing unit shares coordination requirement information, such as the execution timestamp of a module and the action identifier that needs to be synchronized, through internal communication links (e.g., CAN bus, Ethernet). By exchanging this information, the modules can coordinate their execution timing. For example, the hydraulic pump control module needs to receive a synchronization signal of "valve open" from the hydraulic valve control module before it starts increasing the output pressure to avoid excessive system pressure that could damage components.

[0175] The overall coordination unit of the hydraulic system control unit continuously polls or receives execution preparation status information sent by each core control module processing unit, such as "ready," "waiting for synchronization," and "safety anomaly." When the overall coordination unit confirms that all core control module processing units have returned to the "ready" state and there are no unresolved safety anomalies, it determines that all core control modules have completed control information parsing and adaptation preparation.

[0176] Step S1476: The overall coordination unit issues an execution start command, and each core control module starts running according to the execution sequence and control commands in the module-specific control information.

[0177] After confirming that all core control modules are ready, the overall coordination unit sends an execution start command to each core control module processing unit. This execution start command can be a broadcast signal or a sequentially sent trigger signal. Upon receiving the execution start command, each core control module processing unit begins to execute the corresponding operation based on the execution sequence (i.e., execution timestamp) and control commands in the module-specific control information.

[0178] For example, at time t5, the hydraulic pump control module starts the hydraulic pump and adjusts the displacement according to the control command "adjust displacement to XXX"; at time t6, the hydraulic valve control module opens the corresponding hydraulic valve according to the command; after receiving the coordinated action signal from the hydraulic pump and hydraulic valve, the hydraulic cylinder control module begins to control the extension and retraction of the hydraulic cylinder according to the predetermined speed and force. Through this method, the core control modules of the lift's hydraulic system can operate collaboratively according to the predicted hydraulic power demand, achieving precise control of hydraulic power and meeting the lift's power requirements under the current operating conditions.

[0179] Figure 2 The following is a schematic diagram of the hardware structure of a hydraulic power demand prediction system 100 for a lift, which is used to implement the above-described method for predicting hydraulic power demand for a lift based on fusion working condition identification, provided by an embodiment of the present invention. Figure 2 As shown, the lifting hydraulic power demand prediction system 100 that integrates working condition identification may include a processor 110, a machine-readable storage medium 120, a bus 130, and a communication unit 140.

[0180] Machine-readable storage medium 120 can store data and / or instructions. In some embodiments, machine-readable storage medium 120 can store data acquired from an external terminal. In some embodiments, machine-readable storage medium 120 can store data and / or instructions used by the fusion working condition identification lift hydraulic power demand prediction system 100 to perform or use in order to complete the exemplary methods described in this invention. In a specific implementation, one or more processors 110 execute the computer-executable instructions stored in machine-readable storage medium 120, enabling processor 110 to perform the fusion working condition identification lift hydraulic power demand prediction method as described in the above method embodiments. Processor 110, machine-readable storage medium 120, and communication unit 140 are connected via bus 130, and processor 110 can be used to control the transmission and reception actions of communication unit 140. The specific implementation process of processor 110 can be found in the various method embodiments executed by the fusion working condition identification lift hydraulic power demand prediction system 100 described above, and their implementation principles and technical effects are similar, so they will not be repeated here.

[0181] Furthermore, this embodiment of the invention also provides a readable storage medium containing computer-executable instructions. When the processor runs the computer-executable instructions, the above-mentioned method for predicting the hydraulic power demand of a lifting machine based on integrated working condition identification is realized.

[0182] It should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of the embodiments of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof. Similarly, it should be noted that, in order to simplify the description of this invention and thus aid in the understanding of one or more embodiments, the foregoing description of the embodiments of this invention sometimes combines multiple features into a single embodiment, drawing, or description thereof.

Claims

1. A method for predicting the hydraulic power demand of a lifting platform by integrating working condition identification, characterized in that, The method includes: Acquire basic operating information of the lift hydraulic system, basic driving information of the pre-trained AI prediction model, and basic identification information of the lift working condition identification module. Based on the basic operating information, basic driving information, and basic identification information, construct a dynamic coupling and association structure. Based on the aforementioned dynamic coupling and correlation structure, real-time operating information of the lifting hydraulic system, real-time drive status information of the AI ​​prediction model, and real-time working condition feature information of the working condition identification module are collected synchronously to generate a working condition-power mutual drive iterative signal. The working condition-power mutual drive iterative signal is input into the dynamic rule evolution unit, and combined with the operating characteristics of the lifting hydraulic system, the prediction characteristics of the AI ​​prediction model and the dynamic adaptation characteristics of working condition identification, the working condition-power demand prediction rule is evolved. Using the aforementioned working condition-power demand prediction rule as processing logic, the system calculates the continuously input real-time operating information of the hydraulic system, the predicted state data of the AI ​​prediction model, and the real-time characteristic data of the working condition, and outputs the hydraulic power demand prediction result. The hydraulic power demand prediction result is transmitted to the lifting machine hydraulic system control unit through a dynamic coupling association structure.

2. The method for predicting the hydraulic power demand of a lifting machine based on fusion working condition identification according to claim 1, characterized in that, The process involves acquiring basic operational information of the lift's hydraulic system, basic driving information of the pre-trained AI prediction model, and basic identification information of the lift's operating condition identification module. Based on this basic operational information, basic driving information, and basic identification information, a dynamically coupled and related structure is constructed, including: The system collects basic operational information about the lifting platform's hydraulic system. This basic operational information includes hydraulic oil storage status information, hydraulic pump operating basic information, hydraulic valve control basic information, and hydraulic actuator operation basic information. The hydraulic oil storage status information includes oil quantity, oil temperature, and oil pressure. The hydraulic pump operating basic information includes rated flow rate, rated pressure, and operating mode. The hydraulic valve control basic information includes valve type, control signal range, and response time. The hydraulic actuator operation basic information includes cylinder bore, rod diameter, and stroke. Extract the basic driving information of the pre-trained AI prediction model. The basic driving information includes the input parameter types, output signal formats, basic prediction calculation logic, and basic characteristics of the driving response of the AI ​​prediction model. The system collects basic identification information from the lifting machine operating condition identification module. This basic identification information includes operating condition classification standard information, basic information for operating condition feature extraction, basic information for the correlation between operating conditions and power demand, and basic information for dynamic switching identification of operating conditions. The operating condition classification standard information includes a preset list of operating condition categories; the basic information for operating condition feature extraction includes a list of feature sensors to be extracted; the basic information for the correlation between operating conditions and power demand includes typical power values ​​corresponding to various operating conditions in historical data; and the basic information for dynamic switching identification of operating conditions includes threshold conditions for determining operating condition switching. The basic operating information of the hydraulic system is divided into multiple basic operating units according to functional attributes. Each basic operating unit corresponds to a core functional module of the hydraulic system. Each basic operating unit contains the key basic parameters and operating constraints of the core functional module. The basic driving information of the AI ​​prediction model is divided into multiple driving basic units according to the function dimension. Each driving basic unit corresponds to a core driving module of the AI ​​prediction model. Each driving basic unit contains the key driving parameters and response constraints of the core driving module. The basic identification information of the working condition identification module is divided into multiple working condition identification basic units according to the identification dimension. Each working condition identification basic unit corresponds to a core identification module of the working condition identification module. Each working condition identification basic unit contains the key identification parameters and adaptation constraints of the core identification module. Establish a preliminary mapping relationship between the basic operating unit, the basic driving unit, and the basic working condition identification unit. The mapping is based on the parameter type compatibility, functional correlation, and constraint compatibility of the basic operating unit, the basic driving unit, and the basic working condition identification unit, forming a preliminary mapping set. Construct a test dataset containing basic operating state parameters of the hydraulic system, basic drive state parameters of the AI ​​prediction model, and basic state parameters of the working condition identification; input the test dataset into each mapping relationship in the preliminary mapping set, and record the success rate and delay time of data transmission; Based on the success rate and latency of data transmission, adjust the data conversion parameters or link paths of the mapping relationships in the initial mapping set, and then repeat the test dataset until the data transmission success rate of all mapping relationships exceeds the preset threshold and the latency is lower than the preset threshold. The optimized mapping relationship is structurally integrated according to the operation process of the hydraulic system, the driving process of the AI ​​prediction model, and the recognition process of the working condition recognition module, forming a multi-level mapping structure that includes input mapping, output mapping, and feedback mapping. A dynamic adjustment method is embedded in the multi-level mapping structure. This dynamic adjustment method can automatically adjust the parameter association weights or mapping logic in the mapping relationship according to the changes in the hydraulic system operation state, the driving state of the AI ​​prediction model, and the working condition recognition state, thus solidifying the multi-level mapping structure containing the dynamic adjustment method into a dynamically coupled association structure.

3. The method for predicting the hydraulic power demand of a lifting machine based on fusion working condition identification according to claim 1, characterized in that, Based on the aforementioned dynamic coupling and correlation structure, the system synchronously collects real-time operating information of the lifting machine's hydraulic system, real-time drive status information of the AI ​​prediction model, and real-time working condition feature information of the working condition identification module to generate a working condition-power mutual drive iterative signal, including: Through the hydraulic information acquisition port of the dynamically coupled and associated structure, real-time operating information of the lifting machine's hydraulic system is continuously acquired at a preset acquisition frequency. The real-time operating information includes the real-time status of hydraulic oil flow, the real-time status of hydraulic pump operation, the real-time status of hydraulic valve control, and the real-time status of hydraulic actuator action. Through the model information acquisition port of the dynamically coupled and associated structure, real-time driving status information of the AI ​​prediction model is acquired at a rhythm synchronized with the hydraulic information acquisition frequency. The real-time driving status information includes the real-time values ​​of the model's input parameters, the real-time progress of prediction calculation, the real-time status of output signals, and the real-time effect of driving response. Through the working condition information acquisition port of the dynamically coupled and associated structure, real-time working condition feature information of the working condition identification module is acquired at a rhythm synchronized with the hydraulic information acquisition frequency. The real-time working condition feature information includes current working condition type identification information, working condition load feature information, working condition operating speed feature information, and working condition environment adaptation feature information. The collected real-time operating information, real-time drive status information, and real-time working condition characteristic information of the hydraulic system are time-series marked. The set of hydraulic real-time information with time-series marking, the set of model real-time information with time-series marking, and the set of working condition real-time information with time-series marking are paired according to time identifier to form an information pairing set. Each information pairing contains hydraulic real-time information, model real-time information, and working condition real-time information at the same point in time. For each information pair containing real-time hydraulic information, calculate the difference, moving average, or derivative of a specific parameter at the current time relative to the previous N times, and use the calculation results as real-time hydraulic features; for each information pair containing real-time model information, extract the values ​​of its input parameters, model calculation progress identifier, and output signal, and use the extracted content as real-time model features; for each information pair containing real-time operating condition information, extract its current operating condition type identifier, load sensor reading, speed sensor reading, and environmental sensor reading, and use the extracted content as real-time operating condition features. The hydraulic real-time features, model real-time features, and working condition real-time features in the same information pair are input into the correlation analysis model, and the output is a result matrix representing the correlation coefficient or correlation rule between the hydraulic real-time features, model real-time features, and working condition real-time features, which is used as the three-dimensional feature correlation result. When the three-dimensional feature association result indicates that a specific parameter in the real-time hydraulic feature exceeds the preset range, an adjustment instruction is generated. This adjustment instruction includes the name of the AI ​​prediction model driving parameter to be adjusted and the adjustment amount. Based on the three-dimensional feature association results, a prediction signal for hydraulic power demand by the AI ​​prediction model is generated. This prediction signal includes preliminary prediction data and adaptation adjustment suggestions for hydraulic power demand under the coupling of real-time features of the model and real-time features of the working condition. Based on the three-dimensional feature association results, a correlation driving signal for power demand based on working condition features is generated. This correlation driving signal includes key feature parameters and adjustment direction suggestions in the real-time features of the working condition that affect hydraulic power demand. The feedback drive signal, prediction signal and associated drive signal are integrated according to time identifier to form a working condition-power mutual drive iterative signal containing three-way drive information. Each working condition-power mutual drive iterative signal corresponds to a collaborative prediction demand at a certain time point.

4. The method for predicting the hydraulic power demand of a lifting machine based on fusion working condition identification according to claim 1, characterized in that, The step involves inputting the working condition-power mutual drive iterative signal into the dynamic rule evolution unit, combining the operating characteristics of the lifting machine hydraulic system, the predictive characteristics of the AI ​​prediction model, and the dynamic adaptation characteristics of working condition identification to evolve and form working condition-power demand prediction rules, including: The dynamic rule evolution unit receives continuous time series of working condition-power mutual drive iteration signals, sorts the working condition-power mutual drive iteration signals in time order to form an ordered mutual drive signal sequence, which reflects the dynamic mutual drive process between the hydraulic system, the AI ​​prediction model and the working condition identification module. From each signal in the ordered mutual drive signal sequence, the adjustment command field in the feedback drive signal, the predicted value field in the prediction signal, and the characteristic parameter field in the associated drive signal are parsed out, and all parsed fields are stored as a core drive element set; from the hydraulic system technical manual or configuration file, the response delay time, the adjustable range of key parameters, the communication protocol between each functional module, and the rated load parameters of the hydraulic system are read as hydraulic system operating characteristic data; The predictive characteristics of the AI ​​prediction model are analyzed, including the model's prediction accuracy, drive response speed, parameter adjustment range, and multi-scenario adaptation characteristics, forming a description of the AI ​​prediction model's predictive characteristics; and the dynamic adaptation characteristics of the working condition recognition are analyzed, including the working condition recognition response speed, feature extraction accuracy, working condition switching recognition sensitivity, and power demand-related adaptation characteristics, forming a description of the working condition recognition adaptation characteristics. Each adjustment instruction field, predicted value field, or feature parameter field in the core driving element set is compared with the corresponding parameter range or constraint condition in the hydraulic system operating characteristic data to determine whether it is within the allowable range, and the comparison result is recorded as the element hydraulic adaptation result; the core driving element set is associated and matched with the AI ​​prediction model prediction characteristic description to identify the degree of adaptation between each core driving element and the AI ​​prediction model prediction characteristic, forming the element model adaptation result; the core driving element set is associated and matched with the working condition identification adaptation characteristic description to identify the degree of adaptation between each core driving element and the working condition identification adaptation characteristic, forming the element working condition adaptation result; Core driving elements that are judged to be within the allowable range in the element hydraulic adaptation results, element model adaptation results, and element working condition adaptation results are selected and formed into a set of effective elements. The effective core driving elements in the set of effective elements are classified and organized according to their functions into parameter adjustment elements, prediction adaptation elements, working condition related elements, and collaborative optimization elements, forming a set of classified elements. For each category of element, including parameter adjustment elements, prediction adaptation elements, working condition correlation elements, and collaborative optimization elements, an adjustment logic and adaptation standard are constructed for that category of element. The adjustment logic of that category of element determines the correlation between the change pattern of that category of element and the power demand prediction. The adaptation standard of that category of element determines the adaptation requirements between that category of element and the hydraulic system, AI prediction model, and working condition identification module. The adjustment logic and adaptation standards of parameter adjustment elements, prediction adaptation elements, operating condition related elements and collaborative optimization elements are structurally integrated to form operating condition-power demand prediction rules. These operating condition-power demand prediction rules include mutual drive adjustment schemes and power demand prediction standards for all scenarios.

5. The method for predicting the hydraulic power demand of a lifting machine based on fusion working condition identification according to claim 1, characterized in that, The process uses the aforementioned working condition-power demand prediction rule as processing logic to calculate the continuously input real-time operating information of the hydraulic system, the predicted state data of the AI ​​prediction model, and the real-time characteristic data of the working condition, and outputs the hydraulic power demand prediction result, including: The system continuously receives real-time operational data changes from the lifting platform's hydraulic system via a dynamically coupled and correlated structure's change data port. This change data includes fluctuation data, state transition data, and response change data of the operating parameters of each functional module of the hydraulic system. It also continuously receives update data of the AI ​​prediction model's predicted state via the dynamically coupled and correlated structure's update data port. This update data includes updated model input parameters, adjusted prediction calculation logic, and optimized output signals. Finally, it continuously receives real-time operational characteristic change data via the dynamically coupled and correlated structure's working condition change data port. This change data includes working condition type switching data, working condition load fluctuation data, working condition environment adaptation change data, and working condition characteristic parameter adjustment data. Hydraulic system change data is accumulated in chronological order to form a hydraulic system change data sequence, which reflects the dynamic evolution of the real-time operating status of the hydraulic system; AI prediction model update data is accumulated in chronological order to form an AI prediction model update data sequence, which reflects the dynamic evolution of the predicted state by the AI ​​prediction model; and real-time operating condition characteristic change data is accumulated in chronological order to form an operating condition characteristic change data sequence, which reflects the dynamic evolution of the real-time operating condition status. Input the hydraulic system change data sequence, AI prediction model update data sequence, and working condition characteristic change data sequence into the matching unit of the working condition-power demand prediction rule, and filter out the target adjustment logic corresponding to the current change data according to the adaptation criteria in the working condition-power demand prediction rule; Based on the target adjustment logic, key fluctuation data in the hydraulic system change data sequence are analyzed to determine the degree of impact and adjustment direction of hydraulic system operating state changes on power demand; based on the target adjustment logic, key update data in the AI ​​prediction model update data sequence are analyzed to determine the optimization direction and adjustment parameters of the AI ​​prediction model's prediction of state changes on power demand prediction; based on the target adjustment logic, key change data in the working condition characteristic change data sequence are analyzed to determine the correlation and influence of dynamic changes in working conditions on power demand and adaptation adjustment suggestions. The hydraulic system influence adjustment direction, AI prediction model prediction optimization direction, and working condition related influence direction are coordinated and matched. If a conflict is detected, one or more adjustment directions are modified according to the preset conflict resolution strategy until the conflict is eliminated. Based on the adjustment direction after the conflict is eliminated, the corresponding data processing function in the working condition-power demand prediction rule is called to calculate the final value of the power demand prediction parameter, the adjustment value of the input parameter of the AI ​​prediction model, and the feature weight adjustment value of the working condition identification module, and generate the corresponding control sequence. The power demand forecasting parameter adjustment scheme, model optimization scheme, and working condition adaptation scheme are integrated to form a hydraulic power demand forecasting result that includes core parameters for hydraulic power demand forecasting, AI prediction model optimization instructions, working condition adaptation adjustment instructions, and rules for the coordinated execution of the three.

6. The method for predicting the hydraulic power demand of a lifting machine based on fusion working condition identification according to claim 2, characterized in that, The method of embedding a dynamic adjustment mechanism in the multi-level mapping structure can automatically adjust the parameter association weights or mapping logic in the mapping relationship based on changes in the hydraulic system's operating state, changes in the AI ​​prediction model's driven state, and changes in the working condition identification state. This includes: Define the triggering factors for changes in the operating status of the hydraulic system. The triggering factors include: load sensor reading changes exceeding a threshold, ambient temperature sensor reading changes exceeding a threshold, system cumulative operating time reaching a maintenance cycle threshold, or fault codes of critical hardware components being triggered. The analysis of the triggering factors for changes in the state driven by the AI ​​prediction model includes various situations that may lead to changes in the state driven by the AI ​​prediction model, such as changes in the distribution of input data, changes in the prediction scenario, drift of model parameters, and fluctuations in operating resources. The analysis of the triggering factors for changes in the state of the working condition identification includes various situations that may lead to changes in the state of the working condition identification, such as switching of working condition type, load fluctuations, changes in environmental conditions, and drift of identification parameters. A monitoring index is configured for each triggering factor of the hydraulic system. The monitoring index for load change is the load sensor reading; the monitoring index for environmental condition change is the temperature sensor reading; the monitoring index for accumulated running time is the system timer reading; and the monitoring index for hardware wear is a preset vibration or current characteristic value. Corresponding monitoring indexes are set for various triggering factors that drive state changes by AI prediction model and for various triggering factors that identify state changes by working condition. Each monitoring index can reflect the degree of change and the scope of influence of the corresponding triggering factor. Construct a query table that records the type and level of different triggering factors, and the correspondence between them and the weights associated with the parameters to be adjusted or the mapping logic to be switched. A monitoring indicator acquisition unit is set up to collect various monitoring indicator data from the hydraulic system, AI prediction model, and working condition identification module in real time. A monitoring data analysis unit is established to analyze the collected monitoring indicator data in real time, identify the changing status of triggering factors, and determine whether it is necessary to start adjusting the mapping parameters. When the monitoring data analysis unit identifies that a triggering factor has been activated, it queries the query table according to the type and level of the triggering factor to obtain the corresponding adjustment strategy. The adjustment strategy includes parameter association weight adjustment value or mapping logic identifier. Based on the determined adjustment strategy, the parameter association weights or mapping logic of the corresponding mapping relationships in the multi-level mapping structure are adjusted so that the mapping relationships can adapt to the changed hydraulic system operating state, AI prediction model driving state, and working condition identification state. In an offline test environment, historical data or simulated data are used to verify the data transmission success rate and latency time under the adjusted mapping relationships and confirm whether they meet the preset verification threshold.

7. The method for predicting the hydraulic power demand of a lifting machine based on fusion working condition identification according to claim 3, characterized in that, The process involves performing correlation analysis on the real-time hydraulic features, real-time model features, and real-time operating condition features within the same information pair to identify the collaborative changes and mutual influences among these features, forming a three-dimensional feature correlation result, including: Key parameters of hydraulic real-time features are extracted from the same information pairing. These key parameters directly reflect the core operating state of the hydraulic system at that point in time. The key parameters of hydraulic real-time features include hydraulic oil flow change parameters, hydraulic pump operating state parameters, hydraulic valve control and adjustment parameters, and hydraulic actuator action response parameters. Key parameters of model real-time features are also extracted from the same information pairing. These key parameters directly reflect the core driving state of the AI ​​prediction model at that point in time. The key parameters of model real-time features include input parameter change parameters, calculation progress parameters, output signal strength parameters, and response effect feedback parameters. Finally, key parameters of working condition real-time features are extracted from the same information pairing. These key parameters directly reflect the core state of the current working condition. The key parameters of working condition real-time features include working condition type identification parameters, load intensity parameters, operating speed parameters, and environmental adaptation parameters. The key parameters of the real-time hydraulic features, the real-time model features, and the real-time operating conditions features are grouped by functional category to form hydraulic parameter group, model parameter group, and operating condition parameter group. Each of the hydraulic parameter group, model parameter group, and operating condition parameter group contains multiple functionally related key parameters. A time-series variation analysis is performed on each key parameter item in the hydraulic parameter group to determine the variation trend and magnitude of each key parameter item at the current time point relative to historical adjacent time points, thus forming a time-series variation description of the hydraulic parameters; a time-series variation analysis is also performed on each key parameter item in the model parameter group to determine the variation trend and magnitude of each key parameter item at the current time point relative to historical adjacent time points, thus forming a time-series variation description of the model parameters; a time-series variation analysis is also performed on each key parameter item in the operating condition parameter group to determine the variation trend and magnitude of each key parameter item at the current time point relative to historical adjacent time points, thus forming a time-series variation description of the operating condition parameters. The time-series changes of hydraulic parameters, model parameters, and operating conditions parameters are compared parameter by parameter. The correlation coefficient between any two parameters in the hydraulic parameter group, model parameter group, and operating conditions parameter group is calculated in time. A pair of parameters with a correlation coefficient greater than a preset positive threshold is marked as having a cooperative change relationship. For parameter pairs marked as having a synergistic change relationship, calculate the average change of the other parameter corresponding to the unit change of one parameter, and record the average change as a description of the synergistic effect. Calculate the time-series correlation coefficient between any two parameters in the hydraulic parameter group, model parameter group, and working condition parameter group, and mark a pair of parameters whose correlation coefficient is less than a preset negative threshold as having a restrictive relationship. Analyze the triggering conditions for changes in each parameter in the constraint relationship, determine the mutual constraint mechanism between each parameter in the constraint relationship, and form a description of the constraint influence; By integrating the descriptions of synergistic and restrictive effects, and supplementing them with analyses of the temporal synchronicity of parameter changes, functional correlation, and working condition-power correlation, a comprehensive three-dimensional feature correlation result is formed that reflects the relationship between real-time hydraulic characteristics, real-time model characteristics, and real-time working condition characteristics.

8. The method for predicting the hydraulic power demand of a lifting machine based on fusion working condition identification according to claim 4, characterized in that, For each category of classification elements—parameter adjustment elements, prediction adaptation elements, operating condition related elements, and collaborative optimization elements—the adjustment logic and adaptation criteria corresponding to that category of classification elements are constructed, including: For parameter adjustment elements, analyze the relationship between the change pattern of these parameter adjustment elements and the control parameters of the hydraulic system and the driving parameters of the AI ​​prediction model, and establish a correspondence criterion between the change direction of these parameter adjustment elements and the adjustment direction of the control parameters of the hydraulic system and the driving parameters of the AI ​​prediction model. Based on the correspondence between the change direction of the parameter adjustment category element and the adjustment direction of the hydraulic system control parameters and AI prediction model driving parameters, the adjustment logic of the parameter adjustment category element is constructed. The data processing function of the parameter adjustment category element defines the value of the parameter adjustment category element and maps it to the adjustment amount of the hydraulic system control parameters or AI prediction model driving parameters through a preset conversion function. When multiple parameter adjustment category elements are triggered at the same time, the execution order of their data processing functions is determined by a preset priority list. Combining the parameter adjustment sensitivity characteristics of the hydraulic system and the parameter adjustment range characteristics of the AI ​​prediction model, an adaptation standard for this parameter adjustment category is constructed. The adaptation standard for this parameter adjustment category determines the allowable range of the variation amplitude of this parameter adjustment category and the safety boundary for the adjustment of hydraulic system control parameters and AI prediction model driving parameters. For predictive adaptation elements, analyze the relationship between the changing patterns of these elements and the prediction parameters of the AI ​​prediction model and the prediction results of power demand, and establish a correspondence criterion between the changing direction of these predictive adaptation elements and the adjustment direction of the prediction parameters of the AI ​​prediction model and the prediction results of power demand. Based on the correspondence between the direction of change of the prediction adaptation element and the direction of adjustment of the AI ​​prediction model prediction parameters and the power demand prediction results, the adjustment logic of the prediction adaptation element is constructed. The adjustment logic of the prediction adaptation element includes the relationship between the change of the prediction adaptation element and the adjustment of the AI ​​prediction model prediction parameters and the power demand prediction results, as well as the priority sorting method when multiple prediction adaptation elements act simultaneously. Combining the prediction accuracy characteristics of the AI ​​prediction model with the adaptation requirements of power demand prediction, an adaptation standard for the prediction adaptation category of elements is constructed. The adaptation standard for the prediction adaptation category of elements determines the allowable range of the variation of the prediction adaptation category of elements and the accuracy boundary of the adjustment of AI prediction model prediction parameters and power demand prediction results. For operating condition-related elements, analyze the relationship between the changing patterns of these elements and the operating condition identification parameters and power demand forecasting results, and establish a correspondence between the changing direction of these operating condition-related elements and the adjustment direction of the operating condition identification parameters and power demand forecasting results. Based on the correspondence between the change direction of the working condition-related elements and the adjustment direction of the working condition identification parameters and power demand prediction results, the adjustment logic of the working condition-related elements is constructed. The adjustment logic of the working condition-related elements includes the relationship between the change of the working condition-related elements and the adjustment of the working condition identification parameters and power demand prediction results, as well as the priority sorting method when multiple working condition-related elements act simultaneously. Combining the dynamic adaptation characteristics of working condition identification and the adaptation requirements of power demand prediction, an adaptation standard for the related elements of the working condition is constructed. The adaptation standard for the related elements of the working condition determines the allowable range of the variation of the related elements of the working condition and the adaptation boundary for adjusting the working condition identification parameters and power demand prediction results. For collaborative optimization elements, we analyze the relationship between the variation pattern of these collaborative optimization elements and the parameters of the hydraulic system, the parameters of the AI ​​prediction model, and the parameters of the working condition identification, and establish the corresponding criteria between the variation direction of these collaborative optimization elements and the adjustment direction of the parameters of the hydraulic system, the parameters of the AI ​​prediction model, and the parameters of the working condition identification. Based on the correspondence between the change direction of the collaborative optimization element and the adjustment direction of hydraulic system parameters, AI prediction model parameters, and working condition identification parameters, the adjustment logic of the collaborative optimization element is constructed. The adjustment logic of the collaborative optimization element includes the correlation between the change of the collaborative optimization element and the adjustment of hydraulic system parameters, AI prediction model parameters, and working condition identification parameters, as well as the priority sorting method when multiple collaborative optimization elements act simultaneously. By combining the operating characteristics of the hydraulic system, the predictive characteristics of the AI ​​prediction model, and the dynamic adaptation characteristics of the working condition identification, an adaptation standard for this type of collaborative optimization element is constructed. This adaptation standard determines the allowable range of the variation of the collaborative optimization element and the collaborative boundary for adjusting the hydraulic system parameters, AI prediction model parameters, and working condition identification parameters.

9. The method for predicting the hydraulic power demand of a lifting machine based on fusion working condition identification according to claim 5, characterized in that, The process involves collaboratively matching the adjustment direction influenced by the hydraulic system, the optimization direction predicted by the AI ​​prediction model, and the influence direction related to the working conditions. If a conflict is detected, one or more adjustment directions are modified according to a preset conflict resolution strategy until the conflict is eliminated. This includes: List all specific aspects of the hydraulic system's influence on the adjustment direction. For each hydraulic system's influence on the adjustment direction, determine the corresponding hydraulic system parameter category and its specific impact on power demand, forming a list of hydraulic influence directions. List all specific aspects of the AI ​​prediction model's prediction optimization direction. For each AI prediction model's prediction optimization direction, determine the corresponding AI prediction model parameter category and its specific optimization direction for power demand prediction, forming a list of model optimization directions. List all specific aspects of the working condition's associated influence direction. For each working condition's associated influence direction, determine the corresponding working condition characteristic parameter category and its specific association with power demand, forming a list of working condition-related influence directions. The list of hydraulic influence directions, the list of model optimization directions, and the list of working condition related directions are paired according to their functional correlation to form a set of adjustment direction pairs with functional correlation. Each adjustment direction pair includes a hydraulic system influence adjustment direction, an AI prediction model prediction optimization direction, and a working condition related influence direction. Analyze the consistency of the hydraulic system's influence on the adjustment direction, the AI ​​prediction model's prediction of the optimization direction, and the working condition's influence on the direction of each adjustment direction pair. Determine whether the three are predicting and optimizing the same power demand, and form a result of the consistency of the objectives. For adjustment direction pairs with consistent target consistency, we analyze whether the hydraulic system's influence on the execution of the adjustment direction and the AI ​​prediction model's prediction of the optimized direction can support the execution of the working condition-related influence direction in the same pair, thus forming a complementary analysis result. For adjustment direction pairs with inconsistent target consistency, we analyze whether the hydraulic system's influence on the execution of the adjustment direction and the AI ​​prediction model's prediction of the optimized direction in the same pair will hinder the execution of the working condition-related influence direction or lead to a decrease in the overall prediction effect, thus forming a conflict analysis result. Based on the complementarity analysis results, adjustment direction pairs that satisfy the set strength of complementarity are retained, and the cooperative execution order and mutual cooperation mode of the adjustment direction pairs are determined to form a cooperative pair set; based on the conflict analysis results, the adjustment direction pairs with conflicts are adjusted, modifying the specific influence direction of the hydraulic system on the adjustment direction, the specific optimization direction of the AI ​​prediction model prediction optimization direction, or the specific correlation direction of the working condition correlation influence direction in the adjustment direction pair, or reallocating the adjustment priority of the three in the adjustment direction pair to eliminate the conflict; The adjusted non-conflicting adjustment direction pairings are rechecked for target consistency to ensure that the modified adjustment direction pairings can serve the unified power demand prediction optimization goal, forming a set of qualified pairings. The collaborative pairing set and the set of qualified pairings are integrated to form a collaborative matching result that includes all hydraulic system influence adjustment directions, AI prediction model prediction optimization directions, and working condition related influence directions. The execution order, coordination method, and priority of each hydraulic system influence adjustment direction, AI prediction model prediction optimization direction, and working condition related influence direction are determined.

10. A lifting platform hydraulic power demand prediction system integrating working condition identification, characterized in that, The lift hydraulic power demand prediction system based on fusion working condition identification includes a processor and a memory. The memory and the processor are connected. The memory is used to store programs, instructions, or code. The processor is used to run the programs, instructions, or code in the memory to implement the lift hydraulic power demand prediction method based on fusion working condition identification as described in any one of claims 1-9.

Images