An equipment early warning system based on port digitalization

Abstract

The present invention belongs to the field of equipment early warning technology. The present invention discloses an equipment early warning system based on port digitization, including: a data processing module for quantifying operation tasks into operation quantitative indicators and extracting equipment operation characteristics; a prediction model module for building an equipment operation prediction model based on historical data; a real-time prediction module for obtaining the quantitative indicators of the next operation task and predicting the equipment operation characteristics; a prediction correction module for correcting the prediction results based on historical deviations; a task evaluation module for evaluating task adaptability from three dimensions: safety, maintenance requirements, and timeliness; and an alarm module for performing graded early warnings based on task adaptability. Through quantitative analysis of operation task characteristics and consideration of individual differences in equipment, accurate predictions of future operation risks are achieved, and through multi-dimensional evaluation and graded early warning mechanisms, the efficiency and accuracy of risk management are significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment early warning, and more particularly to an equipment early warning system based on port digitization. Background Art

[0002] As core equipment for port loading and unloading operations, the safe and stable operation of container cranes is crucial to ensuring port productivity and the smooth flow of international logistics chains. Container cranes typically withstand high-intensity and high-frequency loads, and their operating conditions directly impact port throughput capacity and service quality. Therefore, condition monitoring and risk warning for container cranes have become crucial aspects of port equipment management.

[0003] Currently, the prevalent early warning systems used in container crane management primarily predict risks based on historical equipment operating data and current state parameters. For example, these systems monitor key indicators such as motor temperature and hydraulic system pressure. Historical operating data captures the changing trends of specific indicators, triggering corresponding alarm mechanisms when these indicators exceed preset thresholds. While this current-state-based early warning approach can respond to immediate equipment anomalies, it has certain limitations. First, traditional early warning systems fail to consider the characteristics of the tasks the equipment will perform and are unable to predict the potential impact of future operations on the equipment. For example, when the equipment is about to perform a high-load, high-intensity task, even if the current equipment state parameters do not exceed the threshold, the accumulated load may still lead to equipment failure or safety incidents during the operation. Furthermore, existing early warning systems are relatively single-dimensional in their risk assessment, focusing primarily on equipment safety indicators and insufficiently considering aspects such as maintenance needs and operational efficiency. This one-sided risk assessment results in a lack of targeted early warning information, hindering detailed classification based on risk type and severity. This leads to over- or under-alarming, compromising the efficiency and accuracy of risk management. Summary of the Invention

[0004] In order to overcome the above problems of the prior art, the present invention proposes an equipment early warning system based on port digitization to solve the above problems.

[0005] The present invention provides the following technical solutions:

[0006] An equipment early warning system based on port digitalization, including:

[0007] The data processing module is used to quantify the operation tasks into quantitative operation indicators, and is also used to collect operation data when executing the operation tasks and extract equipment operation characteristics;

[0008] A prediction model module is used to use historical operation quantitative indicators and corresponding equipment operation characteristics to form a training data set, and to build an equipment operation prediction model based on the training data set, wherein the equipment operation prediction model is used to predict the equipment operation characteristics based on the operation quantitative indicators;

[0009] The real-time prediction module is used to obtain the quantitative indicators of the next task to be executed and predict the equipment operation characteristics through the equipment operation prediction model;

[0010] A prediction correction module is used to obtain a performance deviation vector based on the equipment operation characteristics predicted by the equipment operation prediction model in the past and the corresponding actual equipment operation characteristics extracted by the executing equipment; and to correct the predicted equipment operation characteristics based on the performance deviation vector to obtain corrected operation characteristic data;

[0011] A task evaluation module is used to evaluate the task adaptability of the execution equipment to the next task to be executed based on the modified task characteristic data and the task quantitative indicators;

[0012] The alarm module is used to issue corresponding warnings based on the assessed task adaptability.

[0013] Preferably, the quantifying of the work tasks into work quantification indicators includes:

[0014] Extract container data indicators, operation time indicators, and operation space indicators from operation tasks;

[0015] The container data indicators include the total number of containers, the total weight of containers, the average weight of containers, the maximum and minimum weights of containers, and the container weight distribution index. The container weight distribution index is calculated by multiplying the proportion of the number of containers in each weight interval to the total number by the median of the corresponding interval. The operation time indicator includes the planned operation time.

[0016] The working space indicators include the maximum and minimum loading and unloading heights, the average loading and unloading heights, the loading and unloading height distribution index, the maximum and minimum loading and unloading radiuses, the average loading and unloading radiuses, and the loading and unloading radius distribution index, wherein the loading and unloading height distribution index and the loading and unloading radius distribution index are calculated by multiplying the proportion of each height interval and each radius interval by the median of the corresponding interval respectively;

[0017] Standardize the packing data indicators, operation time indicators and operation space indicators to form quantitative operation indicators.

[0018] Preferably, the collecting of operation data during the execution of the operation task and extracting the equipment operation characteristics includes:

[0019] The main motor temperature, hydraulic system pressure, main structure stress, and equipment vibration frequency are collected during the operation process, and the maximum value of the above values is extracted to form the equipment operation characteristics;

[0020] The hydraulic oil contamination degree and the wear degree of each connection part are collected during the execution of the task. Based on the values at the beginning and after the task is completed, the corresponding contamination change degree and wear change degree are obtained to form the equipment loss characteristics;

[0021] Obtain the total time required to complete the task as the task time feature;

[0022] Equipment operation characteristics are composed of equipment operation characteristics, equipment loss characteristics and task time characteristics.

[0023] Preferably, constructing the equipment operation prediction model based on the training data set includes:

[0024] Perform data cleaning on historical operation quantitative indicators and corresponding equipment operation characteristics, remove outliers and missing values, and form an effective training data set;

[0025] Divide the effective training data set into training set and validation set according to the preset ratio;

[0026] Construct a deep neural network model structure, including an input layer, multiple hidden layers, and an output layer. The number of nodes in the input layer is the same as the dimension of the job quantification indicator, and the number of nodes in the output layer is the same as the dimension of the equipment operation feature.

[0027] Using the job quantitative indicators in the training set as input and the equipment operation characteristics as output labels, the gradient descent optimization algorithm is used to optimize the model parameters;

[0028] Through multiple iterative training of the training set data, the model learns the mapping relationship between the quantitative indicators of the operation and the characteristics of the equipment operation;

[0029] During the training process, the validation set is used to evaluate the model performance. When the error value on the validation set no longer decreases after multiple consecutive iterations, the training process is terminated.

[0030] Use cross-validation to evaluate model performance and calculate the error between the predicted and actual values;

[0031] The model structure and hyperparameters are adjusted based on the evaluation results, and the training set is repeatedly used for training until the model performance meets the predetermined threshold requirements to obtain the equipment operation prediction model.

[0032] Preferably, the step of obtaining the performance deviation vector includes:

[0033] Collect the predicted equipment operation characteristics and the actually extracted equipment operation characteristic data from at least the last five operations of the executing equipment;

[0034] For each operation's predicted and actual features, calculate the difference between the actual and predicted values for each dimension;

[0035] Arrange the differences of each dimension in chronological order to form a deviation sequence of each dimension;

[0036] Predict the expected deviation value of each dimension based on the deviation sequence of each dimension;

[0037] The expected deviation values of each dimension are combined to form a performance deviation vector.

[0038] Preferably, predicting the expected deviation value of each dimension according to the deviation sequence of each dimension includes:

[0039] Detect and remove outliers in deviation sequences of each dimension;

[0040] For the deviation series of each dimension after removing outliers, calculate the weighted mean, deviation change rate and volatility index;

[0041] Taking the most recent deviation value in the deviation sequence as a benchmark, the preliminary deviation is obtained by combining the change rate and time interval;

[0042] Determine the volatility index. If the volatility index is lower than the set threshold, directly use the preliminary deviation as the expected deviation value;

[0043] Otherwise, the weighted mean and the preliminary deviation are fused and their arithmetic mean is taken as the expected deviation value.

[0044] Preferably, the step of evaluating the task adaptability of the execution device to the next task to be performed includes:

[0045] Compare the values of each dimension of the equipment operation characteristics in the corrected operation characteristic data with the corresponding safe operation thresholds; mark the dimensions that exceed the corresponding safety thresholds as safety risk dimensions;

[0046] Obtain the current hydraulic oil contamination degree and wear degree data of each connection part of the execution equipment, and add them to the corresponding contamination change degree and wear change degree in the corrected operation characteristic data to obtain the expected contamination degree and expected wear degree;

[0047] Compare the predicted contamination and wear with the corresponding maintenance thresholds respectively, and mark the dimensions exceeding the corresponding maintenance thresholds as maintenance risk dimensions;

[0048] Compare the task duration characteristics in the revised task feature data with the task time index. If the task duration is not greater than the planned task duration, add a timeliness risk flag.

[0049] The task adaptability is determined based on the safety risk dimension, the maintenance risk dimension, and the timeliness risk marker, and the task adaptability is ranked from 1 to 6 from high to low.

[0050] Preferably, determining task adaptability based on the security risk dimension, the maintenance risk dimension, and the timeliness risk marker includes:

[0051] If there is no safety risk dimension mark, no maintenance risk mark, and no time risk mark, the task adaptability is determined to be level 1 adaptability;

[0052] If there is no safety risk dimension mark, no maintenance risk mark, and a time-limited risk mark, the task adaptability is determined to be level 2 adaptability;

[0053] If there is no safety risk dimension mark, there is a maintenance risk mark, and no timeliness risk mark, the task adaptability is determined to be level 3 adaptability;

[0054] If there is no safety risk dimension mark, one maintenance risk mark, and one time-limited risk mark, the task adaptability is determined to be level 4 adaptability;

[0055] If there is no safety risk dimension marker and more than one maintenance risk marker, the task adaptability is determined to be level 5 adaptability;

[0056] If any of the security risk dimensions are marked, the mission adaptability is determined to be level 6.

[0057] Preferably, the corresponding warning according to the assessed task adaptability includes:

[0058] Establish corresponding warning levels according to the task adaptability level. The higher the adaptability level, the lower the warning level.

[0059] Different warning methods are used for different warning levels, including indicator lights of different colors and sound and light prompts of different frequencies;

[0060] Warning information is sent to management departments in different ranges for different warning levels. The higher the warning level, the wider the scope of information sending.

[0061] The present invention provides an equipment early warning system based on port digitization, which has the following beneficial effects:

[0062] By quantifying tasks into quantitative indicators, the system can fully capture the characteristics of the tasks, including key factors such as the number of containers, weight distribution, and workspace distribution. Furthermore, by constructing a deep neural network model, a mapping relationship is established between quantitative indicators and equipment operation characteristics, enabling predictions of equipment operating characteristics, wear characteristics, and task duration characteristics. This prediction method, which considers task characteristics, significantly improves prediction accuracy. Furthermore, a prediction correction module personalizes the prediction results, taking into account individual differences such as the length of use and maintenance status of each device, making the predictions more realistic.

[0063] By assessing the equipment's adaptability to operational tasks from the perspectives of safety, maintenance requirements, and timeliness, a task adaptability grading mechanism has been established. This multi-dimensional, multi-level assessment mechanism makes risk warnings more precise and targeted, clearly distinguishing between different types and degrees of risk. Consequently, corresponding warning levels can be established, intuitively conveying risk information through indicator lights of varying colors and sound and light cues of varying frequencies. Furthermore, depending on the warning level, the system sends warning information to management departments within different scopes, ensuring that the information reaches those with appropriate processing authority. This tiered warning mechanism avoids unnecessary management burdens caused by low-risk events while ensuring timely and effective responses to high-risk events, significantly improving the efficiency and accuracy of risk management. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a module schematic diagram of an equipment early warning system based on port digitization of the present invention. DETAILED DESCRIPTION

[0065] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0066] Example 1

[0067] See also Figure 1 In this embodiment, an equipment early warning system based on port digitalization includes:

[0068] The data processing module is used to quantify the operation tasks into quantitative operation indicators, and is also used to collect operation data when executing the operation tasks and extract equipment operation characteristics;

[0069] The quantification of the task into the quantitative task index includes:

[0070] Extract container data indicators, operation time indicators, and operation space indicators from operation tasks;

[0071] The container data indicators include the total number of containers, the total weight of containers, the average weight of containers, the maximum and minimum weights of containers, and the container weight distribution index. The container weight distribution index is calculated by multiplying the proportion of the number of containers in each weight interval to the total number by the median of the corresponding interval. The operation time indicator includes the planned operation time.

[0072] The working space indicators include the maximum and minimum loading and unloading heights, the average loading and unloading heights, the loading and unloading height distribution index, the maximum and minimum loading and unloading radiuses, the average loading and unloading radiuses, and the loading and unloading radius distribution index, wherein the loading and unloading height distribution index and the loading and unloading radius distribution index are calculated by multiplying the proportion of each height interval and each radius interval by the median of the corresponding interval respectively;

[0073] Standardize the packing data indicators, operation time indicators and operation space indicators to form quantitative operation indicators.

[0074] Collecting operational data during task execution and extracting equipment operational features include:

[0075] The main motor temperature, hydraulic system pressure, main structure stress, and equipment vibration frequency are collected during the operation process, and the maximum value of the above values is extracted to form the equipment operation characteristics;

[0076] The hydraulic oil contamination degree and the wear degree of each connection part are collected during the execution of the task. Based on the values at the beginning and after the task is completed, the corresponding contamination change degree and wear change degree are obtained to form the equipment loss characteristics;

[0077] Obtain the total time required to complete the task as the task time feature;

[0078] Equipment operation characteristics are composed of equipment operation characteristics, equipment loss characteristics and task time characteristics.

[0079] In this embodiment, the operation task information of the container crane can be obtained first through the port production management system, the port digital system, etc., and the operation task can be quantified into operation quantitative indicators. The specific steps of quantification can be: extracting container data indicators from the operation task, including the total number of containers, total weight, average weight, maximum and minimum weights, and weight distribution index; the weight distribution index can accurately reflect the distribution characteristics of the container weight in the operation task. At the same time, the planned operation time in the operation time indicator is extracted, which can be used as a reference for the task volume and also as a benchmark value for evaluating the timeliness of the task. For the operation space indicator, the maximum and minimum loading and unloading heights, the average height, the height distribution index, and the maximum and minimum loading and unloading radius, the average radius, and the radius distribution index are calculated. The distribution index of loading and unloading height and radius can fully characterize the spatial distribution characteristics of the operation task. Finally, all extracted indicators are standardized. The Z-score standardization method can be used to convert indicators of different dimensions into comparable standard scales to form a unified operation quantitative indicator vector.

[0080] To extract equipment operational characteristics, the data processing module uses a sensor network deployed on the container crane to collect real-time data during the execution of operational tasks. First, operational parameters such as the main motor temperature, hydraulic system pressure, main structural stress, and equipment vibration frequency are collected. The maximum values of these parameters throughout the entire operation process are extracted to form an equipment operational characteristic vector. These maximum values reflect the equipment's operational status and are important indicators for assessing equipment safety. Second, data on hydraulic oil contamination and wear at various joints is collected. By comparing the measured values at the start and end of the operation, the contamination and wear changes are calculated to form an equipment wear characteristic vector. This before-and-after comparison method accurately quantifies the cumulative wear and tear caused by a single operation, providing a direct basis for predicting equipment maintenance needs. Finally, the actual completion time of the operation is recorded as a task duration feature. The equipment operational characteristics, equipment wear characteristics, and task duration features are combined to form a complete equipment operational characteristic vector, which comprehensively reflects the equipment's operational status, wear and tear, and efficiency performance during the execution of a specific operation.

[0081] A prediction model module is used to use historical operation quantitative indicators and corresponding equipment operation characteristics to form a training data set, and to build an equipment operation prediction model based on the training data set, wherein the equipment operation prediction model is used to predict the equipment operation characteristics based on the operation quantitative indicators;

[0082] The constructing of the equipment operation prediction model according to the training data set includes:

[0083] Perform data cleaning on historical operation quantitative indicators and corresponding equipment operation characteristics, remove outliers and missing values, and form an effective training data set;

[0084] Divide the effective training data set into training set and validation set according to the preset ratio;

[0085] Construct a deep neural network model structure, including an input layer, multiple hidden layers, and an output layer. The number of nodes in the input layer is the same as the dimension of the job quantification indicator, and the number of nodes in the output layer is the same as the dimension of the equipment operation feature.

[0086] Using the job quantitative indicators in the training set as input and the equipment operation characteristics as output labels, the gradient descent optimization algorithm is used to optimize the model parameters;

[0087] Through multiple iterative training of the training set data, the model learns the mapping relationship between the quantitative indicators of the operation and the characteristics of the equipment operation;

[0088] During the training process, the validation set is used to evaluate the model performance. When the error value on the validation set no longer decreases after multiple consecutive iterations, the training process is terminated.

[0089] Use cross-validation to evaluate model performance and calculate the error between the predicted and actual values;

[0090] The model structure and hyperparameters are adjusted based on the evaluation results, and the training set is repeatedly used for training until the model performance meets the predetermined threshold requirements to obtain the equipment operation prediction model.

[0091] In this example, we first collected historical operation records of port container cranes, including various quantitative operational indicators and corresponding equipment operational characteristics, to form an initial training dataset. We then cleaned this historical data to remove outliers and missing values, improving data quality.

[0092] The valid training dataset was randomly split into a training set and a validation set in an 8:2 ratio. Next, a deep neural network model structure was constructed. The model consists of an input layer, three hidden layers, and an output layer. The number of nodes in the input layer matches the dimension of the job quantification indicators, which is 15 nodes. The first hidden layer can have 64 neurons, the second hidden layer can have 128 neurons, and the third hidden layer can have 64 neurons. The number of nodes in the output layer matches the dimension of the device operation characteristics, which is 12 nodes. The hidden layers can use the ReLU activation function.

[0093] Using the job metrics from the training set as input and the device job characteristics as output labels, we optimized the model parameters using a gradient descent optimization algorithm. Specifically, we used the Adam optimizer, with an initial learning rate of 0.001, a batch size of 64, and the mean squared error (MSE) loss function. To prevent overfitting, we added a dropout layer after each hidden layer with a dropout rate of 0.2 and L2 regularization of 0.001.

[0094] Through multiple iterations of training data, the model learns the mapping between job metrics and device operational characteristics. The maximum number of iterations can be set to 500, and after each round of training, the validation set is used to evaluate model performance. Training is terminated early when the error on the validation set stops decreasing after 15 consecutive iterations to avoid overfitting.

[0095] To evaluate model performance, use a 5-fold cross-validation method. Split the training dataset into five equal parts, rotating four of them as training data and one as validation data. For each validation set, calculate metrics such as the mean squared error, mean absolute error, and R² coefficient of determination between the predicted and actual values. The average of the five validation runs is used to evaluate the overall model performance.

[0096] Based on the evaluation results, the model structure and hyperparameters are adjusted, including the number of hidden layers, the number of neurons per layer, the learning rate, and the regularization coefficient. The system uses a combination of grid search and random search to try different hyperparameter combinations, repeatedly training on the training set. When the model's mean squared error on the validation set falls below the predetermined threshold of 0.05 and the R² coefficient of determination exceeds 0.9, the model is considered to meet the requirements and the final equipment operation prediction model is obtained.

[0097] This deep learning-based prediction model captures the complex, nonlinear relationships between quantitative job metrics and equipment operating characteristics, improving prediction accuracy compared to traditional statistical methods. By learning patterns from extensive historical data, the model accurately predicts the load and wear and tear imposed on equipment by different tasks, providing a reliable basis for subsequent task suitability assessments.

[0098] The real-time prediction module is used to obtain the quantitative indicators of the next task to be executed and predict the equipment operation characteristics through the equipment operation prediction model;

[0099] In this embodiment, when a new work task is assigned to a container crane, the real-time prediction module, through the data processing module, processes the task into a standardized vector of quantitative operational indicators. The real-time prediction module then inputs these quantitative indicators into a trained equipment operation prediction model. The model automatically calculates and outputs predicted equipment operation characteristics based on the input indicators. This real-time prediction capability enables the system to assess the potential impact of the task on the equipment before the operation begins, providing data support for subsequent risk warnings.

[0100] A prediction correction module is used to obtain a performance deviation vector based on the equipment operation characteristics predicted by the equipment operation prediction model in the past and the corresponding actual equipment operation characteristics extracted by the executing equipment; and to correct the predicted equipment operation characteristics based on the performance deviation vector to obtain corrected operation characteristic data;

[0101] The step of obtaining the performance deviation vector includes:

[0102] Collect the predicted equipment operation characteristics and the actually extracted equipment operation characteristic data from at least the last five operations of the executing equipment;

[0103] For each operation's predicted and actual features, calculate the difference between the actual and predicted values for each dimension;

[0104] Arrange the differences of each dimension in chronological order to form a deviation sequence of each dimension;

[0105] Predict the expected deviation value of each dimension based on the deviation sequence of each dimension;

[0106] The expected deviation values of each dimension are combined to form a performance deviation vector.

[0107] The step of predicting the expected deviation value of each dimension according to the deviation sequence of each dimension includes:

[0108] Detect and remove outliers in deviation sequences of each dimension;

[0109] For the deviation series of each dimension after removing outliers, calculate the weighted mean, deviation change rate and volatility index;

[0110] Taking the most recent deviation value in the deviation sequence as a benchmark, the preliminary deviation is obtained by combining the change rate and time interval;

[0111] Determine the volatility index. If the volatility index is lower than the set threshold, directly use the preliminary deviation as the expected deviation value;

[0112] Otherwise, the weighted mean and the preliminary deviation are fused and their arithmetic mean is taken as the expected deviation value.

[0113] In this embodiment, although the deep learning model can establish a mapping relationship between the quantitative indicators of the operation and the characteristics of the equipment operation, due to factors such as the usage time and maintenance status of each equipment, there is often a certain deviation between the predicted results and the actual situation. The prediction correction module analyzes the historical prediction deviation pattern of a specific equipment and makes personalized corrections to the prediction results to improve the prediction accuracy. The modification process can be specifically as follows:

[0114] First, the system collects the most recent operation records for the executing equipment, including the predicted and actual characteristic data for at least five complete operations. For a container crane, for example, the system records the predicted and actual maximum temperature values of the main motor for the equipment's five most recent operations. The difference between these values is then calculated to form a deviation sequence. Outlier detection within the deviation sequence can be performed using a modified Z-score method, which identifies outliers by calculating the median absolute deviation (MAD). Appropriate outlier thresholds are set, and when a deviation value exceeds the threshold, it is considered an outlier and removed from the sequence to ensure the accuracy of subsequent analysis.

[0115] Next, the weighted mean, deviation change rate, and volatility index are calculated. For the weighted mean, a time-decay weighting method can be used to give higher weight to recent deviation values, reflecting the importance of recent data to the current forecast. The deviation change rate uses linear regression to calculate the overall trend of the deviation series, yielding the average deviation growth value for each operation. The volatility index, calculated by calculating the ratio of the standard deviation to the mean, is used to assess the stability of the deviation series. The system can preset a volatility threshold for subsequent decision-making.

[0116] Using the most recent deviation as a benchmark, combined with the calculated rate of change and the time interval (e.g., number of operation cycles), a preliminary deviation value for the current operation can be calculated. The system then determines whether the fluctuation index is below a preset threshold. If the fluctuation index is low, indicating relatively stable deviation changes, the preliminary deviation can be used directly as the expected deviation value. If the fluctuation index is high, indicating significant deviation fluctuations, the weighted mean and preliminary deviation are combined, and the arithmetic mean is taken as a more robust expected deviation value. The same calculation process is performed for all dimensions of the equipment's operating characteristics, ultimately forming a complete performance deviation vector.

[0117] Apply the performance deviation vector to the prediction results for the current task and make corrections to the corresponding dimension. For example, add the predicted maximum temperature of the main motor to the corresponding deviation value to obtain the corrected temperature value. Corrections are made to other dimensions in the same way, ultimately obtaining complete corrected task feature data.

[0118] This approach corrects the model's predicted equipment operating characteristics, capturing the individual operating characteristics of the equipment and making them more realistic. This approach also eliminates the need to retrain the prediction model, resulting in high computational efficiency and suitability for real-time applications.

[0119] A task evaluation module is used to evaluate the task adaptability of the execution equipment to the next task to be executed based on the modified task characteristic data and the task quantitative indicators;

[0120] The step of evaluating the task adaptability of the execution device to the next task to be executed includes:

[0121] Compare the values of each dimension of the equipment operation characteristics in the corrected operation characteristic data with the corresponding safe operation thresholds; mark the dimensions that exceed the corresponding safety thresholds as safety risk dimensions;

[0122] Obtain the current hydraulic oil contamination degree and wear degree data of each connection part of the execution equipment, and add them to the corresponding contamination change degree and wear change degree in the corrected operation characteristic data to obtain the expected contamination degree and expected wear degree;

[0123] Compare the predicted contamination and wear with the corresponding maintenance thresholds respectively, and mark the dimensions exceeding the corresponding maintenance thresholds as maintenance risk dimensions;

[0124] Compare the task duration characteristics in the revised task feature data with the task time index. If the task duration is not greater than the planned task duration, add a timeliness risk flag.

[0125] The task adaptability is determined based on the safety risk dimension, the maintenance risk dimension, and the timeliness risk marker, and the task adaptability is ranked from 1 to 6 from high to low.

[0126] Determining task adaptability based on the security risk dimension, maintenance risk dimension, and timeliness risk marker includes:

[0127] If there is no safety risk dimension mark, no maintenance risk mark, and no time risk mark, the task adaptability is determined to be level 1 adaptability;

[0128] If there is no safety risk dimension mark, no maintenance risk mark, and a time-limited risk mark, the task adaptability is determined to be level 2 adaptability;

[0129] If there is no safety risk dimension mark, there is a maintenance risk mark, and no timeliness risk mark, the task adaptability is determined to be level 3 adaptability;

[0130] If there is no safety risk dimension mark, one maintenance risk mark, and one time-limited risk mark, the task adaptability is determined to be level 4 adaptability;

[0131] If there is no safety risk dimension marker and more than one maintenance risk marker, the task adaptability is determined to be level 5 adaptability;

[0132] If any of the security risk dimensions are marked, the mission adaptability is determined to be level 6.

[0133] In this embodiment, the task evaluation module evaluates the adaptability of the equipment to perform the next task from three dimensions: safety, maintenance requirements, and timeliness, by comparing the corrected operation feature data with various thresholds. The various thresholds can be pre-set by comprehensively considering factors such as the specific model and service life of the equipment.

[0134] By comparing equipment operating characteristics with safety thresholds, projected contamination and wear levels with maintenance thresholds, and task duration with planned duration, the system comprehensively assesses potential risks during task execution. This data-driven approach eliminates the subjectivity of traditional empirical judgments and makes task suitability assessments more scientific.

[0135] At the same time, the hierarchical design provides structured input data for subsequent alarm modules, enabling early warning information to accurately reflect the type and severity of risk. This allows different staff members to make timely decisions based on the adaptability results at different levels, such as adjusting task assignments, replacing execution equipment, or arranging equipment maintenance.

[0136] The alarm module is used to issue corresponding warnings based on the assessed task adaptability.

[0137] The corresponding warning according to the assessed task adaptability includes:

[0138] Establish corresponding warning levels according to the task adaptability level. The higher the adaptability level, the lower the warning level.

[0139] Different warning methods are used for different warning levels, including indicator lights of different colors and sound and light prompts of different frequencies;

[0140] Warning information is sent to management departments in different ranges for different warning levels. The higher the warning level, the wider the scope of information sending.

[0141] In this embodiment, the alarm module establishes a corresponding early warning mechanism based on the mission adaptability level output by the mission assessment module, ensuring timely and effective communication of risks. The early warning level is inversely proportional to the mission adaptability level: the higher the mission adaptability level (lower the risk), the lower the early warning level; the lower the mission adaptability level (higher the risk), the higher the early warning level.

[0142] Warning information is delivered through multiple sensory channels, including visual and auditory prompts. For visual prompts, an intuitive color-coding system is used. For example, Level 1 adaptability corresponds to a green indicator light, indicating safety and no risk; Level 2 adaptability corresponds to a blue indicator light, indicating a minor failure risk; Levels 3 and 4 adaptability correspond to yellow indicators, indicating maintenance risks requiring attention; Level 5 adaptability corresponds to an orange indicator light, indicating multiple maintenance risks; and Level 6 adaptability corresponds to a red indicator light, indicating a safety risk. The frequency of audio and visual prompts also increases with the warning level, from silent prompts to continuous high-frequency alarms, allowing operators to quickly perceive the risk level.

[0143] The scope of early warning information transmission is appropriately expanded based on risk level. For example, low-level warnings (e.g., Levels 1-2) are only notified to equipment operators and direct managers; mid-level warnings (e.g., Levels 3-4) are extended to maintenance departments and dispatch centers; and high-level warnings (e.g., Levels 5-6) are reported to safety management departments and port management. This hierarchical transmission mechanism ensures that information reaches those with appropriate processing authority, avoiding unnecessary management burdens for low-risk incidents while ensuring timely and effective responses to high-risk incidents.

[0144] This early warning mechanism transforms complex risk assessment results into simple and clear early warning signals through standardized information transmission methods, making it easier for personnel at all levels to quickly understand and respond.

[0145] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only one type. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0146] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

[0147] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An equipment early warning system based on port digitalization, characterized in that: include: The data processing module is used to quantify the operation tasks into quantitative operation indicators, and is also used to collect operation data when executing the operation tasks and extract equipment operation characteristics; A prediction model module is used to use historical operation quantitative indicators and corresponding equipment operation characteristics to form a training data set, and to build an equipment operation prediction model based on the training data set, wherein the equipment operation prediction model is used to predict the equipment operation characteristics based on the operation quantitative indicators; The real-time prediction module is used to obtain the quantitative indicators of the next task to be executed and predict the equipment operation characteristics through the equipment operation prediction model; A prediction correction module is used to obtain a performance deviation vector based on the equipment operation characteristics predicted by the equipment operation prediction model in the past and the corresponding actual equipment operation characteristics extracted by the executing equipment; and to correct the predicted equipment operation characteristics based on the performance deviation vector to obtain corrected operation characteristic data; A task evaluation module is used to evaluate the task adaptability of the execution equipment to the next task to be executed based on the modified task characteristic data and the task quantitative indicators; Alarm module, used to issue corresponding warnings based on the assessed task adaptability; The step of obtaining the performance deviation vector includes: Collect the predicted equipment operation characteristics and the actually extracted equipment operation characteristic data from at least the last five operations of the executing equipment; For each operation's predicted and actual features, calculate the difference between the actual and predicted values for each dimension; Arrange the differences of each dimension in chronological order to form a deviation sequence of each dimension; Predict the expected deviation value of each dimension based on the deviation sequence of each dimension; Combine the expected deviation values of each dimension to form a performance deviation vector; The step of evaluating the task adaptability of the execution device to the next task to be executed includes: Compare the values of each dimension of the equipment operation characteristics in the corrected operation characteristic data with the corresponding safe operation thresholds; mark the dimensions that exceed the corresponding safety thresholds as safety risk dimensions; Obtain the current hydraulic oil contamination degree and wear degree data of each connection part of the execution equipment, and add them to the corresponding contamination change degree and wear change degree in the corrected operation characteristic data to obtain the expected contamination degree and expected wear degree; Compare the predicted contamination and wear with the corresponding maintenance thresholds respectively, and mark the dimensions exceeding the corresponding maintenance thresholds as maintenance risk dimensions; Compare the task duration characteristics in the revised task feature data with the task time index. If the task duration is not greater than the planned task duration, add a timeliness risk flag. The task adaptability is determined based on the safety risk dimension, the maintenance risk dimension, and the timeliness risk marker, and the task adaptability is ranked from 1 to 6 from high to low.

2. The equipment early warning system based on port digitalization according to claim 1 is characterized in that: The quantification of the task into the quantitative task index includes: Extract container data indicators, operation time indicators, and operation space indicators from operation tasks; The container data indicators include the total number of containers, the total weight of containers, the average weight of containers, the maximum and minimum weights of containers, and the container weight distribution index. The container weight distribution index is calculated by multiplying the proportion of the number of containers in each weight interval to the total number by the median of the corresponding interval. The operation time indicator includes the planned operation time. The working space indicators include the maximum and minimum loading and unloading heights, the average loading and unloading heights, the loading and unloading height distribution index, the maximum and minimum loading and unloading radiuses, the average loading and unloading radiuses, and the loading and unloading radius distribution index, wherein the loading and unloading height distribution index and the loading and unloading radius distribution index are calculated by multiplying the proportion of each height interval and each radius interval by the median of the corresponding interval respectively; Standardize the packing data indicators, operation time indicators and operation space indicators to form quantitative operation indicators.

3. The equipment early warning system based on port digitalization according to claim 2 is characterized in that: The collecting of operation data during the execution of the operation task and extracting the equipment operation characteristics includes: The main motor temperature, hydraulic system pressure, main structure stress, and equipment vibration frequency are collected during the operation process, and the maximum value of the above values is extracted to form the equipment operation characteristics; The hydraulic oil contamination degree and the wear degree of each connection part are collected during the execution of the task. Based on the values at the beginning and after the task is completed, the corresponding contamination change degree and wear change degree are obtained to form the equipment loss characteristics; Obtain the total time required to complete the task as the task time feature; Equipment operation characteristics are composed of equipment operation characteristics, equipment loss characteristics and task time characteristics.

4. The equipment early warning system based on port digitalization according to claim 3 is characterized in that: The constructing of the equipment operation prediction model according to the training data set includes: Perform data cleaning on historical operation quantitative indicators and corresponding equipment operation characteristics, remove outliers and missing values, and form an effective training data set; Divide the effective training data set into training set and validation set according to the preset ratio; Construct a deep neural network model structure, including an input layer, multiple hidden layers, and an output layer. The number of nodes in the input layer is the same as the dimension of the job quantification indicator, and the number of nodes in the output layer is the same as the dimension of the equipment operation feature. Using the job quantitative indicators in the training set as input and the equipment operation characteristics as output labels, the gradient descent optimization algorithm is used to optimize the model parameters; Through multiple iterative training of the training set data, the model learns the mapping relationship between the quantitative indicators of the operation and the characteristics of the equipment operation; During the training process, the validation set is used to evaluate the model performance. When the error value on the validation set no longer decreases after multiple consecutive iterations, the training process is terminated. Use cross-validation to evaluate model performance and calculate the error between the predicted and actual values; The model structure and hyperparameters are adjusted based on the evaluation results, and the training set is repeatedly used for training until the model performance meets the predetermined threshold requirements to obtain the equipment operation prediction model.

5. The equipment early warning system based on port digitalization according to claim 4 is characterized in that: The step of predicting the expected deviation value of each dimension according to the deviation sequence of each dimension includes: Detect and remove outliers in deviation sequences of each dimension; For the deviation series of each dimension after removing outliers, calculate the weighted mean, deviation change rate and volatility index; Taking the most recent deviation value in the deviation sequence as a benchmark, the preliminary deviation is obtained by combining the change rate and time interval; Determine the volatility index. If the volatility index is lower than the set threshold, directly use the preliminary deviation as the expected deviation value; Otherwise, the weighted mean and the preliminary deviation are fused and their arithmetic mean is taken as the expected deviation value.

6. The equipment early warning system based on port digitalization according to claim 5 is characterized in that: Determining task adaptability based on the security risk dimension, maintenance risk dimension, and timeliness risk marker includes: If there is no safety risk dimension mark, no maintenance risk mark, and no time risk mark, the task adaptability is determined to be level 1 adaptability; If there is no safety risk dimension mark, no maintenance risk mark, and a time-limited risk mark, the task adaptability is determined to be level 2 adaptability; If there is no safety risk dimension mark, there is a maintenance risk mark, and no timeliness risk mark, the task adaptability is determined to be level 3 adaptability; If there is no safety risk dimension mark, one maintenance risk mark, and one time-limited risk mark, the task adaptability is determined to be level 4 adaptability; If there is no safety risk dimension marker and more than one maintenance risk marker, the task adaptability is determined to be level 5 adaptability; If any of the security risk dimensions are marked, the mission adaptability is determined to be level 6.

7. The equipment early warning system based on port digitalization according to claim 6 is characterized in that: The corresponding warning according to the assessed task adaptability includes: Establish corresponding warning levels according to the task adaptability level. The higher the adaptability level, the lower the warning level. Different warning methods are used for different warning levels, including indicator lights of different colors and sound and light prompts of different frequencies; Warning information is sent to management departments in different ranges for different warning levels. The higher the warning level, the wider the scope of information sending.

Images